After a New Year’s Eve swim in 2023, Dan Richards’ life changed forever

When Dan Richards went for a New Year’s Eve swim in 2023, he never could have imagined how drastically his life would change.

In a freak accident, he injured his neck when a wave caused him to flip and hit the sand in Langland Bay, Swansea.

“I knew instantly that I was paralysed,” the 37-year-old said.

“I couldn’t move anything.”

Doctors told him he would be bed-bound but, two years later, he uses a wheelchair and can move his arms and fingers.

He has even walked with the help of Artificial Intelligence (AI) technology in Wales and Germany.

Dan and his partner Anna, 40, were celebrating the new year with a cold water dip when the accident happened.

“I remember all of it, unfortunately. Being pulled out. And then everything changing,” Dan said.

“I got hit by a wave, it flipped me over and folded me backwards, and snapped my neck. I knew it was bad.”

Anna Thomas

Dan is determined to move as much as possible after becoming paralysed

Anna recalled hearing Dan shouting for help as she got ready to get into the water.

“I just remember looking over and just seeing Dan’s head coming up and then going back under,” she said.

“We dragged him out. But he was just dead weight and the sea was coming in really fast.

“We waited for emergency services, and then [I remember] begging them to let me go in the helicopter. It was horrendous.”

After being taken to hospital in Bristol, the couple from Swansea were given life-changing news.

Anna Thomas

Dan and Anna loved to swim in the sea before the accident

Dan said he was told he was paralysed form the neck down, would not be able to move and it was “highly likely” he would be bed-bound for the rest of his life.

Anna – who had only started dating Dan a few months before the accident – had to make dreaded phone calls to family members.

“It was New Year’s Eve, I was put in a room on my own but you could hear people laughing and joking, and I had to tell Dan’s parents.

“How do you phone someone’s mum and say that their son is paralysed?”

Dan was given life-changing news while in hospital in January 2024

Anna said every day since the accident has brought a new challenge.

“It’s our reality. Until you’re in this situation… you don’t realise just how much you have lost.”

But in the months that followed the accident, when Dan was in hospital, even the slightest movement in his toes gave the couple hope things would change.

Dan and Anna have returned to Wales from Germany to celebrate Christmas

“Luckily I’m quite stubborn,” Dan said.

“I just wanted to see the physios as soon as possible. I wanted to work. I didn’t accept what they were saying.

“I’m proud of the progress I’ve made. It’s a long time since laying in bed in Bristol, that’s for sure.”

He continued: “The feeling in my toes then translated into full feeling through my legs and feet, some movement in my right leg [and] core strength – which I was told was gone – started to come back. Movement in my arms, my fingers, being able to hold things.

“I’m still hopeful, I’m still positive. With modern technologies, I’m not giving up.”

Dan had private physiotherapy at a specialist clinic in south Wales which used a world-first combined treatment to help him walk with a machine.

Anna Thomas

Family and friends have fundraised for Dan’s private physiotherapy treatments

Jakko Brouwers, a neurophysiotherapist from Morello clinic in Newport, said Dan was “an immensely motivated guy” when he visited them after being discharged from hospital.

“The technology is two-fold,” he explained.

“The first part we tried with Dan is a robot. The robot will mimic normal human gait as much as possible.

“The other part of the system that we developed is a stimulation suit that has got sensors. The trousers are also fitted with electrodes and we can stimulate muscle activity at the right time.

“Through a little bit of algorithm – and dare I say AI – it will start developing a walking pattern.

“It’s super exciting. It’s not often that you see that in Wales.”

This robot helped Dan to walk at Morello clinic in Newport shortly after he was discharged from hospital

Dan said the experience was “surreal” but “the best feeling”.

“It just made me want to push harder. That’s on our doorstep here in Wales, so it spurred us on to do some research, and do other things.”

But private physiotherapy and innovative treatment does not come without cost.

From raffles to ultramarathons and charity nights, family and friends have fundraised meaning Dan and Anna could set off for a trial abroad in October.

Anna Thomas

Dan and Anna spent two months in Bochum, Germany, for treatment

Dan explained he was trying two different treatments simultaneously in Germany – a stem cell treatment straight into his spinal cord, alongside using a Hybrid Assistive Limb (HAL) suit, which is a machine that works on brain waves.

“It’s one of the first times it’s ever been done,” he said.

“Sensors translate signals into movement, which then get my legs working, get me walking.

“The more you do it, the stronger the signals get, the faster it gets and eventually you’re able to walk without the suit.”

With six weeks of treatment left in Germany in the new year, Dan will have to apply for an EU Medical Visa, which is another challenge he said he is determined to overcome.

He does not know what the future holds, but is determined not to give up.

“Technology is advancing so fast, things aren’t where they were 10 years ago.

“New research is incredible. There is no limit. I don’t want there to be a limit.

“The more progress I can make now, the better our future will be.”

SOURCE PAGE

It’s that time of year again.

Every January 1, the Future Perfect team makes forecasts for the events we think will (or won’t) happen over the next 365 days. And every December 31, we go back over those predictions and tally up how we did.

All of our predictions were made positively — as in, something will happen — and came with probabilities attached, which are meant to indicate our relative confidence in the forecast. To simplify scoring, predictions that came with a higher than 50 percent probability that proved out, or with a probability below 50 percent that did not prove out, were marked as “correct call.” Those that came with a higher than 50 percent probability that did not prove out, or with a lower than 50 percent probability that did prove out, were marked “incorrect call.”

If for some reason the forecast could not be resolved — such as, random example here, a new US government chose to delay putting out data or a report that would have clarified the question — we marked it as undecided.

The scorecard? Nineteen correct, four incorrect, and two undecided works out to a winning percentage of .800, if we count ties as half a win. (That would put us a tad over the 1906 Chicago Cubs, who recorded the best single-season winning percentage in major league baseball history. Hopefully this doesn’t mean we’ll be cursed for a century.)

As always, the point is less to keep score than to get better at forecasting by identifying where we’ve succeeded, where we’ve failed — and maybe where we need to take some more chances. Fortunately, we’ll have another shot tomorrow, when we publish our 2026 forecasts. —Bryan Walsh

Congress passes a major tariff bill (20 percent) — CORRECT CALL

2025 certainly did not lack for tariff news, but almost all of it came from the Trump administration, which used executive powers to impose sweeping new duties on most countries on Earth, and from the Supreme Court as it weighed whether any of that was legal.

There was some speculation at the start of 2025 that the need for new revenue in Republicans’ big tax bill would lead it to include some Trump-y tariffs. That didn’t happen, mostly because it didn’t need to happen: President Donald Trump could just impose the tariffs unilaterally, or try to at least. As I wrote in my initial prediction, “the odds that Trump does new tariffs using presidential authority are nearly 100 percent.” If anything, “nearly” 100 percent was an underestimate. —Dylan Matthews

Trump dissolves the Department of Education (5 percent) — CORRECT CALL

Let’s check the fine print: This prediction would’ve resolved true if Congress passed a law formally abolishing the Department of Education. That did not happen in 2025, so the prediction stands.

What Trump did do is issue an executive order instructing the Secretary of Education to, “to the maximum extent appropriate and permitted by law, take all necessary steps to facilitate the closure of the Department of Education.” What has followed are sweeping staff cuts that it’s fair to call a gutting of the department, with various court challenges that in July culminated in a Supreme Court ruling in favor of the administration, at least for the time being. One major pending fight is over the legality of the department moving its functions to other parts of the federal government.

But again, read the fine print. The administration’s solicitor general, in his Supreme Court filing in June, stated, “The government has been crystal clear in acknowledging that only Congress can eliminate the Department of Education.” What the administration did were simply layoffs, not the closure of a legally created government agency. While the Trump team is clearly trying to have it both ways here, I’m inclined to trust their lawyer — they did not dissolve the department. —DM

President Donald Trump signs executive orders in the Oval Office of the White House in Washington, DC, on January 20, 2025. Jim Watson/AFP via Getty Images

The Affordable Care Act is repealed (30 percent) — CORRECT CALL

This is another one where the fine print matters. In my initial prediction, I wrote that a bill “repealing the ACA” has to do at least three of the following five things:

• Eliminate or reduce the ACA’s Medicaid eligibility or federal funding
• Eliminate or reduce ACA health insurance tax credit eligibility or amount
• Eliminate or curtail the mandate for certain employers to provide health coverage for employees. Reducing the penalties will also be considered to be relaxing the mandate.
• Make it so that ACA subsidies are no longer limited to plans that satisfy the requirements specified in the ACA, including allowing ACA subsidies to be contributed to health savings accounts or similar accounts
• Eliminate or curtail medical underwriting restrictions, like the ban on considering preexisting conditions

The One Big Beautiful Bill Act certainly satisfies the first two of these requirements. Per the Committee for a Responsible Federal Budget’s breakdown, the bill includes $1.1 trillion in cuts to health care programs over a decade. The vast majority of those cuts go to Medicaid, by imposing work requirements, limiting “provider taxes,” and other changes. But about $226 billion in cuts go to the Affordable Care Act’s exchange-based coverage, mostly by making certain immigrants ineligible.

But squeezing Medicaid and the exchanges is, at most, cutting the Affordable Care Act, not repealing it. Trump and Congress did not change the employer mandate for health insurance, or allow ACA funds to go into health savings accounts, or, crucially, eliminate protections for people with preexisting conditions or limits on hiking premiums based on age. In my book, that means the ACA has yet to be repealed. —DM

Jerome Powell will no longer be Fed chair (10 percent) — CORRECT CALL

Trump would love nothing more than to fire Jerome Powell, who was first appointed chair of the Federal Reserve by some fiendish anti-MAGA president named Donald Trump way back in 2017. Powell has been open about the way Trump’s tariffs, by hiking prices, are slowing the Fed’s process of lowering interest rates, and the president does not like that one bit.

In April, Trump said Powell’s “termination cannot come fast enough!” In July, he showed off a letter he had written, but not filed, firing Powell. In November, he told reporters he wanted to fire Powell, but people like Treasury Secretary Steve Bessent are “holding me back.” And in August, Trump attempted to fire Fed governor Lisa Cook, a move the Supreme Court has blocked but which was, among other things, a clear threat to Powell that he could be next.

Yet here Powell is, still chair of the Fed. Actually removing him, or trying, proved too rich for Trump’s blood. Powell’s term as chair ends in May 2026, meaning Trump will pick his successor, but it appears he’ll be able to stay in charge until then. He can also keep his post as a regular governor on the board until January 2028, if he wants it. —DM

Trump will have a positive favorability rating (25 percent) — CORRECT CALL

Let’s go to the graph, folks:

Everyone’s polling average is a little different, but basically every one looks like this from Nate Silver: Trump began his presidency slightly above water, but now Americans disapprove of him by a healthy margin (13 points here). The Economist’s average shows him as less popular than either President Joe Biden or Trump himself in term one were at this point in their presidencies.

Being below water at this point has become pretty normal for presidents in the 21st century, so there wasn’t much courage in me predicting Trump would be more disliked than liked. But it’s interesting to me that the speed of the decline has picked up in recent months. I would’ve guessed that Trump’s most-disliked period would’ve been the height of DOGE, but it’s been the period when his ties to Jeffrey Epstein were most under question. —DM

Musk and Trump are still friends at the end of the year (40 percent) — CORRECT CALL

Only two men can tell us if Elon Musk and Trump are truly, as of December 2025, “friends.” But the formal definition I used here is that they stop being friends “if one or the other publicly and unambiguously disparages his counterpart at least three times” over the year. And buddy…

Those Musk tweets are now deleted, and there appears to have been some degree of rapprochement in the ensuing months. But as predicted, there was a massive blow-up in their relationship, centered around the One Big Beautiful Bill Act and, implicitly, the failure of Musk’s DOGE to do anything to actually reduce federal spending. While it does seem as if they’ve made an attempt to patch things up, what’s clear is that their bond is much weaker than it was on January 1. Sad! —DM

The National Highway Traffic Safety Administration’s preliminary estimates of US car crash deaths for 2024 will be lower than 40,000 (70 percent) — CORRECT CALL

America stands out among wealthy nations for being the land of death by cars. But there is, finally, some good news here: After a terrifying period of elevated car fatalities during Covid, the US has seen 13 consecutive quarterly declines in these deaths. As of 2024, we’re back below 40,000 Americans killed by cars annually, according to federal statistics — an achievement that (sadly) calls for celebration. But we still have a ways to go before we’re back down to the pre-pandemic baseline. —Marina Bolotnikova

Israeli Prime Minister Benjamin Netanyahu meets with US Vice President J.D. Vance on October 22, 2025 in Jerusalem, Israel. Nathan Howard/Getty Images

Benjamin Netanyahu is still Israel’s PM at the end of November 2025 (75 percent) — CORRECT CALL

Netanyahu seemed like a marked man going into 2025.

The war in Gaza had already stretched past a year, and dozens of hostages remained in the hands of Hamas, even as Israel was coming under fire for charges of genocide in its conduct of the war. Netanyahu himself was facing long-running corruption allegations and public anger over both judicial reforms and the war, while the International Criminal Court had issued arrest warrants for him and for his former defense minister Yoav Gallant (as well as Hamas military commander Mohammed Deif).

Well, there’s a reason that Netanyahu is the longest-serving leader in Israeli history: The man has an undeniable talent for political self-preservation. With Trump returning to the White House, Netanyahu had an ally who gave him an even freer hand in Gaza, where Israel adopted tactics that maximized damage (and civilian suffering) in Gaza while reducing the record number of casualties it had suffered in 2024. In June, he launched a major attack against Iran that represented a major tactical victory, one that ultimately included enlisting the US in the attack. By October, whether he fully wanted it or not, Netanyahu had a ceasefire in Gaza that included the return of the remaining 20 living hostages.

As 2026 begins, Netanyahu is far from popular and Israel has increasingly become an international pariah, but he has yet to be dislodged from his position at the top of his deeply divided country. Perhaps that will change with the next Israeli elections, which must take place no later than October 27, but I, for one, have learned not to bet against this man. —BW

Argentina’s yearly inflation is below 30 percent (20 percent) — UNDECIDED

This has been a very challenging year for Argentina’s economy, after a surprisingly strong 2024. Inflation is far below where it was when the populist Kirchners were in charge, but swaggering libertarian president Javier Milei’s reforms have also led to high unemployment and voter discontent. That led to a defeat in Buenos Aires elections in September, which led currency, stock, and bond markets to fret over the country’s prospects. This culminated in the US government offering to buy up to $20 billion in Argentinian pesos so Milei’s government had an adequate supply of dollars and could maintain a viable exchange rate.

Having the world hegemon bail you out is, it turns out, good politics: Less than two months after the bad Buenos Aires results, Milei won national midterms in a landslide, giving him much firmer support in Argentina’s National Congress for his reforms.

That’s all background to the question here: inflation. I predicted that inflation would continue to fall but not below 30 percent; I relied in part on an IMF forecast of 45 percent inflation. The most recent data as I write this comes from October, where prices were 31.3 percent higher than October 2024. That implies an annual inflation rate just above our 30 percent cutoff. We’ll have to see what the January numbers say, but there’s a very good chance I was wrong here and underestimated Milei and the Argentinian economy. Regardless of which side of 30 percent we land on, I was much too confident. —DM

There will be a ceasefire in Ukraine (75 percent) — INCORRECT CALL

When I made this call, I thought the logic was straightforward. The war was grinding into its third year, both sides had taken appalling losses, and Trump was about to take office with little interest in writing Ukraine a blank check. It seemed reasonable that Moscow and Kyiv would fight hard for marginal gains in early 2025, then accept a ceasefire that froze the lines.

That is not the world we’re in. As 2025 ends, the conflict in Ukraine remains the largest war in Europe since World War II, with well over a million people killed or wounded and Russia still occupying roughly a fifth of Ukrainian territory. There have been brief truces — measured in dozens of hours or a few days at most — but nothing that qualifies as the “durable pause in the fighting” I had in mind.

Instead, we have diplomacy without peace. The Trump administration is pushing a plan that would freeze the front lines and lift some sanctions; Russian and American officials are shuttling between European capitals and Miami hotel conference rooms; and Ukraine, Europe, and the US have reportedly agreed on most of a peace framework. The sticking point is exactly what you’d expect: territory and legitimacy. Ukrainian President Volodymyr Zelenskyy still refuses to recognize Russia’s land grab in the east and south, while Putin insists that any ceasefire ratify his conquests.

In retrospect, I overweighted “war-weariness” and underweighted how much the key actors care about not losing. I implicitly assumed a Korean War-style ending: a bloody stalemate capped by an ugly armistice. What we actually got was the stalemate without the armistice, and one that is set to continue into the new year. —BW

Iran gets nuclear weapons (30 percent) — CORRECT CALL

This was the prediction where I tried to be precise about definitions. I wrote that “getting nuclear weapons” didn’t mean a test or a declared arsenal, but Iran producing enough fissile material to fuel at least one bomb. Building and deploying an actual warhead, I argued, could take months or years beyond that. So instead, I staked this prediction on a key nuclear benchmark: Iran enriching uranium to weapons-grade (~90% U-235) in sufficient quantity for at least one device.

Not long after I made the prediction, Iran was already enriching uranium to 60 percent at its Natanz and Fordow facilities, and outside experts thought its “breakout time” — how long it would take to produce weapons-grade uranium for one device — was down to perhaps a week.

In 2025, the enrichment problem got dramatically worse. A February International Atomic Energy Agency report found that Iran’s stockpile of 60 percent-enriched uranium had jumped to about 275 kilograms, up roughly 50 percent from late 2024. By May, the agency was estimating some 408.6 kilograms of 60-percent material — and a June update put the figure at around 440.9 kilograms, which its own yardstick says is enough, if further enriched, for roughly nine or 10 simple fission weapons. Then came a 12-day US-Israeli air and covert campaign that killed senior Iranian nuclear scientists and wrecked parts of the program, but even Israeli and US officials concede it did not eliminate Iran’s ability to rebuild.

While all highly enriched uranium — anything above about 20 percent enriched — is in principle weapon-usable, watchdogs note that Iran has not been publicly observed enriching to the classic weapons-grade threshold of 90 percent, nor is there evidence of an actual tested device.

So did Iran “get nuclear weapons” in 2025? The answer remains no, although it comes with the additional confounding factor that, with international inspections suspended, the true state of Iran’s nuclear program may be murkier than ever. Which is why you can expect this question to continue to haunt international politics in 2026 and beyond. —BW

The World Health Organization (WHO) will declare H5N1 a pandemic in 2025 (25 percent) — CORRECT CALL

I’ve been covering the H5N1 bird flu virus since the spring of 2003 in Hong Kong, when there was some suspicion that the unknown illness spreading in southern China at the time might be bird flu finally transmitting human to human. It wasn’t — it was something entirely new called SARS-CoV-1, though back in those pre-Covid days we didn’t have the “1.”

Every January since, I’ve been wondering if this is the year we finally get our dreaded bird flu pandemic. And every year, including 2025, it hasn’t been.

Instead, we got a year that underlined the basic tension of H5N1: It keeps looking terrifying on paper, while acting more like a slow-burn animal disaster than a human pandemic. H5 bird flu is now entrenched in wild birds, poultry, and US dairy cattle. The US experienced its first US H5N1 death early in the year and nearly 70 US infections since April 2024, mostly among workers around infected herds and flocks.

On the animal side, the picture is much worse. A major Nature perspective described a true H5N1 “panzootic” across bird and mammal species, including mink, marine mammals, and cattle, with clear evidence of mammal-to-mammal spread in some settings and worrying adaptive mutations. What we’re seeing adds up to an unprecedented number of mammalian infections, severe neurological disease in animals, and growing uncertainty about how close this virus is to efficient human transmission.

There is some good news on preparedness. Health agencies still classify the overall public health risk from current H5 viruses as low, and vaccine work is accelerating. In December, Moderna and CEPI announced funding for a late-stage trial of an mRNA bird flu vaccine.

So, once again, no H5N1 bird flu pandemic in humans. After 22 years of covering this virus I’m tempted to just say that pandemic will never happen, but I’m not quite that foolhardy. When it comes to H5N1, we’ve been more lucky than we’ve been good. —BW

A major lab will formally claim it has achieved AGI (30 percent) — CORRECT CALL

There is a lot of hype and boosterism in the world of AI. The firm Anthropic has publicly predicted they’ll get to artificial intelligence systems “matching or exceeding that of Nobel Prize winners across most disciplines” by 2027. Elon Musk, meanwhile, has tweeted, “My estimate of the probability of Grok 5 [his firm xAI’s next model] achieving AGI is now 10 percent and rising.”

But Grok 5 isn’t out yet, and it’s 2025, not 2027. I made a very long list of Western companies that could even theoretically be in the running to build AGI (including, like, Netflix, which is not trying to do this at all). Foolishly, I didn’t include Chinese firms, failing to anticipate the “DeepSeek shock” at the start of 2025.

In any case, nobody claimed AGI this past year, whether in the US or China. I’d be surprised if anyone does in 2026, either. Then again, AI as a field is always able to surprise me. —DM

EVs will make up more than 10 percent of new car sales in the US by the end of Q3 2025 (65 percent) — CORRECT CALL

An electric car recharges its battery at a curbside Charge Point electric vehicle charging station on November 16, 2025, in Jersey City, New Jersey. Gary Hershorn/Getty Images

So, I was right here, but I may be wrong in spirit. Electric cars made up 10.5 percent of new car sales in the third quarter of 2025 — but that was probably only because people who wanted an EV anyway were rushing to buy one before the federal government’s $7,500 tax credits for new EVs, which were killed by Trump’s One Big Beautiful Bill expired at the end of September. US electric car sales are expected to dip significantly as a result.

Beyond ending that subsidy, which was critical for EV adoption, the Trump administration is trying to go after every other pillar that makes electric cars viable. They’ve proposed significantly weakening Biden-era fuel economy rules and hamstrung the buildout of EV charging stations. Oh, and half the country hates Elon Musk now, so Tesla sales, which once made up the overwhelming majority of the US electric car market, have taken a big hit. Americans also just seem wary of electric cars because of vague cultural vibes and societal malaise. The US is way behind the rest of the world in EV adoption — a lag that Trump seems determined to turn into a permanent technological deficit. —MB

Bitcoin’s price will at some point in 2025 breach $200,000 (70 percent) — INCORRECT CALL

Funny enough, as a bit of a bitcoin skeptic, I bought into the bitcoin hype — only to be disappointed. I thought for certain after crypto bros helped put Trump into office, he’d reward the best-known cryptocurrency around with astronomical growth. When Trump was sworn in, bitcoin was already hovering near its all-time high value, a little over $100,000. The sky was the limit.

But then, uh, Trump happened. Rather than building on the record 2024 gains that made me so optimistic, bitcoin endured a turbulent year. Uncertainty around Trump’s tariffs, the AI boom and its own unpredictable economic impact, and other economic variables (interest rates) sent the bitcoin price plummeting, then soaring, and back again. Bitcoin did reach a new record high briefly back in October, at more than $125,000, but it fell far short of my projection — and as of this writing on December 29, it’s back well below where it was at Trump’s inauguration. Whoops. —Dylan Scott

Elon Musk is still the richest person in the world (55 percent) — CORRECT CALL

The Bloomberg Billionaire’s Index has seen some fascinating shifts over the past year. There are now 18 billionaires worth at least $100 billion each, including three members of the Walton family. Larry Page and Sergey Brin of Google each added about $100 billion to their tally as Alphabet stock rallied. But the same guy remained at the top, buoyed by the persistently high price of Tesla stock: Elon Musk. As of December 29, he’s worth $638 billion, or more than twice Page, who’s currently in second with $270 billion.

But you know what’s cooler than half a trillion dollars? A trillion dollars, which Musk got Tesla stockholders to agree to pay him if the firm hits key targets over the next 10 years. I guess one of these years we’ll have to add a “the world gets a trillionaire” prediction. —DM

A new application for psychedelic therapy drugs is submitted to the FDA (20 percent) — CORRECT CALL

After last year’s brouhaha with Lykos Therapeutics — the organization that tried (and failed) getting MDMA-assisted therapy approved by the FDA — we didn’t have the highest confidence here. In order to have an application ready for review, you need Phase 3 trials. And those take years to accomplish — and neither Compass Pathways nor the Usona Institute, the two companies mayhaps the furthest along in psilocybin depression treatment, submitted.

But! Oshan Jarow’s initial prediction also accounted for the possibility of the FDA using emergency use authorization to temporarily reschedule certain psychedelics. That didn’t happen either. Fingers crossed for 2027? —Izzie Ramirez

The 2025–2030 federal dietary guidelines advise Americans to avoid ultra-processed foods (30 percent) — UNDECIDED

If this were a normal year, the new dietary guidelines that will shape the next five years of food policy would have already been released. Health Secretary Robert F. Kennedy Jr. had even promised to release them ahead of schedule, well before August, and with everything we need to know to guide nutritional choices condensed into just four pages!

Of course, that didn’t happen, and the new guidelines have now been delayed until January. It turns out that nutrition science is actually quite complicated and can’t just be reduced to aphorisms like, as Kennedy puts it, “eat whole foods.”

My prediction had totally underestimated how incompetent and unmoored from expert consensus the second Trump administration would turn out to be. Once the guidelines come out, I now do expect that they’ll probably make confusing and misleading claims about so-called ultra-processed foods, along with other bad advice, which I thought unlikely a year ago. And, lesson learned, I’m going to avoid making predictions that rely on the timely release of federal government information for the foreseeable future. —MB

Antibiotic sales for use in livestock production will have increased by at least 0.5 percent in 2024 (55 percent) — CORRECT CALL

Sometimes you can be “right,” and yet still miss the mark. I really underestimated how dramatically antibiotic sales for use in livestock production would increase in 2024. I predicted, with a timid 55 percent probability, that sales would increase by at least 0.5 percent. But in 2024, they shot up by an astonishing 15.8 percent.

That should worry you because antibiotics use in livestock production is a pressing public health problem. Here’s why, from my prediction last year:

Most of the antibiotics used in human medicine are actually sold to meat companies, which put them in animals’ feed to make them grow faster and prevent disease outbreaks in factory farms. But some bacteria on farms are becoming resistant to these antibiotics, giving way to new strains of antibiotic-resistant bacteria that make the drugs less effective in treating humans.

For years, US meat companies and trade groups — along with the US Food and Drug Administration — pledged to be better “stewards” of these precious drugs, namely by reducing their use. It appears that it was mostly hot air. There were steep declines of antibiotic use in the mid-2010s, thanks to FDA rules, but sales have since stabilized and are now increasing. The vibes are shifting on antibiotics in meat production, and that’s bad news for the future of these lifesaving medicines. —Kenny Torrella

Bird flu results in the deaths of at least 30 million farmed birds by the end of 2025 (60 percent) — CORRECT CALL

With an outbreak of bird flu, millions of chickens were euthanized to prevent the spread of the virus, leading to a decline in the egg supply and driving prices to record highs. Deb Cohn-Orbach/UCG/Universal Images Group via Getty Images

This current bird flu outbreak has been dragging on for nearly four years, and 2025 was one of the worst yet, with nearly 54 million birds culled as of December 12.

The virus hit egg farms particularly hard in late 2024 and early 2025, resulting in egg shortages and massive price spikes. Some grocery stores even restricted the number of cartons each customer could purchase.

The egg industry, which has been damaged the most by the bird flu, is ready to start vaccinating its birds. But the US Department of Agriculture won’t let it, for fear it’ll severely disrupt the trade of chicken meat — an entirely different sector of the animal agriculture sector. It’s a long and complicated story, which I went into detail on a couple of months ago; check out the story here.

I have little hope common sense will prevail in 2026, so we’re likely in for another bad year of dead birds, higher food prices, and unused vaccines. —KT

California’s animal agriculture law Proposition 12 will not be overturned by Congress (65 percent) — CORRECT CALL

I should, if anything, have predicted this with higher probability. The only somewhat surprising part is that Congress still hasn’t passed a new Farm Bill to replace the one that expired more than two years ago, which is really behind schedule even by today’s chronically late legislative standards. (The coalition that made the last century of farm bills possible is breaking down, as Republicans demand steep cuts to SNAP and an end to “climate-smart” provisions in ag funding.)

In theory, that still gives them the chance to kill Prop 12 in the Farm Bill that eventually passes, but the longer that the animal welfare law remains in place, the less likely the pork industry is to continue campaigning against it, and the less likely it is to be nullified — and thank God for that. —MB

At least one additional state bans lab-grown meat in 2025 (80 percent) — CORRECT CALL

This is another case of being technically right while far underestimating reality. I predicted at least one state would ban the production and sale of lab-grown, or cell-cultivated, meat in 2025, but three to five did, depending on how you look at it: Mississippi, Montana, and Nebraska passed indefinite bans, while Texas and Indiana passed two-year bans.

Prior to 2025, only Florida and Alabama had banned it.

The movement is primarily driven by Republican state lawmakers, including some who are ranchers and farmers themselves, which represents a form of “government protectionism” for the meat industry, according to one Nebraska cattle rancher who opposed the bans (so too did several state-level Nebraska farm groups, along with the National Cattlemen’s Beef Association).

In the short term, the bans have little impact, as no cell-cultivated meat company has scaled up their production enough to sell large amounts of the product. Several companies now have government approval to do so, but Wildtype — the San Francisco-based startup that makes cell-cultivated salmon — is the only one that’s managed to get into numerous restaurants; two in California, one in Oregon, and one in Washington state, which are unlikely to pass bans. If you have the chance to try them, I recommend it — I did a few years ago and thought it was delicious. —KT

A major sports gambling scandal leads at least one All-Star in the four major professional sports to be suspended (30 percent) — INCORRECT CALL

Okay, technically, Emmanuel Clase, the Cleveland Guardians star closer and three-time All-Star, is on “nondisciplinary paid leave” but for the purposes of this prediction, we’re going to call it suspension by another name. Clase and his teammate Luis Ortiz were arrested in November on charges of illegally conspiring in a scheme to rig their pitches in order to pay out prop bets made by their associates. You can now find all kinds of videos detailing how Clase would throw his first pitch in the dirt after entering a game; as it turns out, his co-conspirators were allegedly betting that first pitch would be a ball.

However, based on the rules of our prediction contest, since I put less than 30 percent probability, this technically comes up “wrong.” But I was onto something. Legal gambling continues to creep into every facet of professional sports, with the happy collaboration of the leagues, and the scandals have followed. Clase wasn’t alone this year: Former NBA All-Star, current Portland head coach, and once-presumed future Hall of Famer Chauncey Billups was implicated in a separate sports betting scandal this year. Unless something changes, I suspect neither of them will be the last. —DS

Max Verstappen wins the Formula 1 World Drivers’ Championship (60 percent) — INCORRECT CALL

Max Verstappen on the podium celebrating his win at the 2025 Formula 1 Etihad Airways Abu Dhabi Grand Prix in United Arab Emirates on December 7, 2025. Nicolas Economou/NurPhoto via Getty Images

Sigh. Okay, so in any ordinary year, I would have put Verstappen, the four-time champion driver for Red Bull, at an 80 percent likelihood of winning. He’s a menace. Can drive from the back of the grid all the way to first. But things were rocky at Red Bull, from second driver woes to full-on company culture shifts. The 2025 Red Bull car was — and this is as nicely as I’ll put it — underperformed. All the while, McLaren’s Lando Norris and Oscar Piastri were gaining points weekend after weekend.

For most of the year, I was thinking about this prediction. Was I too generous? He wasn’t a contender for the first half of the season. But it’s Verstappen we’re talking about — he made a legendary 104-point comeback, essentially unheard of in F1. Then the last few races were total nail-biters, with the three drivers so close to each other in points. I was even thinking about hiring an Etsy witch so I could say I was right for this silly little article.

Anyway, Verstappen ended up winning the season closer in Abu Dhabi, but Lando Norris took the championship title by 2 measly points. Yes, I’m upset about it. —IR

Charli XCX wins a Grammy for Brat (90 percent) — CORRECT CALL

And…water is wet. Last year, I kept my prediction intentionally open, hence the high percentage confidence here. Out of the eight nominations she received, she won three Grammys: Best Dance Pop Recording for “Von Dutch,” Best Recording Package, and Best Electronic Dance/Electronic Album. While I hoped she would have won for Best Album, she’ll always be No. 1 to me. —IR

You’ve read 1 article in the last month

Here at Vox, we’re unwavering in our commitment to covering the issues that matter most to you — threats to democracy, immigration, reproductive rights, the environment, and the rising polarization across this country.

Our mission is to provide clear, accessible journalism that empowers you to stay informed and engaged in shaping our world. By becoming a Vox Member, you directly strengthen our ability to deliver in-depth, independent reporting that drives meaningful change.

We rely on readers like you — join us.

Swati Sharma

Vox Editor-in-Chief

SOURCE PAGE

A scheme to encourage climbers to bring rubbish down from Mount Everest has been scrapped, after dishonest mountaineers found ways to cheat the system. 

Introduced in 2014, the programme forced those scaling the mountain to pay a $4,000 (£2,964) deposit, which they only got back if they brought at least 18lbs (8kg) of rubbish back down with them. 

It was hoped the scheme would help to clear up rubbish left by previous mountaineerers – including oxygen cylinders and human waste. 

However, 11 years later, the rubbish problem has ‘not gone away’. 

Climbers have been able to cheat the system by collecting rubbish from lower camps, rather than the higher camps where the majority of the waste lays.  

Speaking to the BBC, Tshering Sherpa, chief executive officer of the Sagarmatha Pollution Control Committee (SPCC), which runs an Everest checkpoint, explained: ‘From higher camps, people tend to bring back oxygen bottles only.

‘Other things like tents and cans and boxes of packed foods and drinks are mostly left behind there, that is why we can see so much of waste piling up.’

To understand the issue, this shocking map reveals the true scale of waste on Mount Everest. 

Climbers have been able to cheat the system by collecting rubbish from lower camps, rather than the higher camps where the majority of the waste lays. Pictured: a Nepalese sherpa collecting rubbish at an altitude of 8,000 metres

Mount Everest sits within the Sagamartha National Park in the Khumbu region of Nepal. 

The number of tourists visiting the park has been steadily increasing for years but has recently accelerated, doubling in the three years between 2014 and 2017.

While these tourists bring in millions for the Nepalese Government and the local economy, they also bring in vast amounts of waste. 

Every year shocking images show camps littered with tattered tents, abandoned gear, and human waste. 

The SPCC only records how much waste is collected each year, and so there are no official estimates for the amount of rubbish currently on the mountain.

But, a 2020 paper predicted that there may be 50 tonnes of solid waste left on Everest in the last 60 years.

Additionally, in 2022 the Nepalese Army reported that it had removed around 34 tonnes of waste from Everest and the surrounding mountains, up from 27.6 tonnes in 2021.

In a bid to tackle this growing issue, the rubbish collection scheme was introduced in 2014, and applied to climbers ascending beyond Everest’s base camp.

A 2020 paper predicted that there may be 50 tonnes of solid waste left on Everest in the last 60 years. Pictured: rubbish scattered around Camp 4 of Mount Everest

How much waste is on Mount Everest? 

• 900–1,000 tonnes of solid waste is brought into the Sagarmatha National Park each year.
• An estimated 50 tonnes is located above Everest Base Camp.
• One to three tonnes of human waste is left between Camp One and Camp Four.
• Base Camp generated 75 tonnes of waste in Spring 2023.
• An estimated 20 tonnes of human waste is dumped in nearby pits each year.  
• There are between 100 and 120 open rubbish pits in the National Park. 

Speaking at the time, tourism ministry official Madhusudan Burlakoti said authorities would take legal action against climbers who failed to comply with the rule.

‘The government has decided in order to clean up Mount Everest that each member of an expedition must bring back at least eight kilos of garbage, apart from their own trash’, he said.

According to Nepalese authorities, most of the deposit money was refunded to climbers over the 11–year programme. 

However, while climbers would bring back the required 8kg, they actually accumulated far more themselves. 

According to Mr Sherpa, the average climber produces up to 12kg (26lbs) of waste during their six weeks of acclimatisation and climbing. 

What’s more, at the higher camps, there were no authorities monitoring what climbers were doing – or what rubbish they were discarding. 

‘Apart from the check point above the Khumbu Icefall, there is no monitoring of what climbers are doing,’ he explained. 

To combat this issue once and for all, officials are now introducing a new rule for climbers – a non–refundable clean–up fee of around $4,000 (£2,964). 

This will be used to set up a checkpoint at Camp Two, and to deploy mountain rangers who will venture further up the mountain to monitor rubbish collection. 

Speaking to the BBC, Mingma Sherpa, chairperson of the Pasang Lhamu rural municipality, explained that the Sherpa community has been calling for this change for yeras. 

‘We had been questioning the effectiveness of the deposit scheme all this time because we are not aware of anyone who was penalised for not bringing their trash down,’ he said. 

‘And there was no designated fund but now this non–refundable fee will lead to creation of a fund that can enable us to do all these clean–up and monitoring works.’

WHAT IS BEING DONE TO REDUCE RUBBISH ON MOUNT EVEREST?

Decades of commercial mountaineering have turned Mount Everest into the world’s highest rubbish dump. 

As the number of climbers on the mountain has soared – at least 600 people have scaled the world’s highest peak so far this year alone – the problem of waste disposal has worsened. 

The worst rubbish is found at Camp Two, which is 21,000 foot (6,400m) above sea level. 

Five years ago Nepal implemented a $4,000 (£3,000) rubbish deposit per team that would be refunded if each climber brought down at least eight kilograms (18 pounds) of waste.

On the Tibet side of the Himalayan mountain, they are required to bring down the same amount and are fined $100 (£75) per kilogram if they don’t.

In 2017 climbers in Nepal brought down nearly 25 tonnes of trash and 15 tonnes of human waste – the equivalent of three double-decker buses – according to the Sagarmatha Pollution Control Committee (SPCC).

This season even more was carried down but this is just a fraction of the rubbish dumped each year, with only half of climbers lugging down the required amounts, the SPCC says.

Instead many climbers opt to forfeit the deposit, a drop in the ocean compared to the $20,000 (£15,000) – $100,000 (£75,000) they will have forked out for the experience.

Another solution, believes Ang Tsering Sherpa, former president of the Nepal Mountaineering Association, would be a dedicated rubbish collection team. 

SOURCE PAGE

Samsung has officially announced the US release of two new budget devices — the Galaxy A17 5G smartphone and the Galaxy Tab A11+ tablet — arriving on 7 January and 8 January respectively.

But these are not just cheap devices. Samsung is quietly rewriting the rules of the budget market by offering flagship-tier longevity: six years of Android updates for the A17 5G and seven years for the Tab A11+.

Galaxy A17 5G: Built To Last

Image Credit: Samsung

Samsung is giving the $200 phone category something it has never really had — durability and long-term value.

The Galaxy A17 5G features Corning Gorilla Glass Victus, a protective glass usually reserved for premium flagships.

It also comes with six generations of Android OS upgrades and six years of security updates, meaning this phone could realistically stay usable well into the next decade.

Key Specs

• Display: 6.7-inch FHD+ Super AMOLED, 90Hz
• Battery: 5,000mAh with 25W fast charging
• Processor: Exynos 1330 or MediaTek Dimensity 6300 (carrier-dependent)
• Cameras:
  • 50MP main with OIS
  • 5MP ultra-wide
  • 2MP macro
• AI Features: Circle To Search and built-in Google Gemini integration
• Price: $199.99

For the first time, Samsung is putting real AI features on a sub-$200 phone — not watered-down versions, but the same Gemini system found on its premium Galaxy S models.

Samsung may push Galaxy S26 launch to February 2026 as chip strategy comes into focusIf the February timeline holds, it means you won’t be able to buy the Galaxy S26 lineup till March 2026.TechloyEmmanuel Umahi

Galaxy Tab A11+: The Entry-Level Workhorse

Image Credit: Samsung

Samsung is also redefining what an “entry-level tablet” can be.

The Galaxy Tab A11+ supports DeX Standalone Edition, turning the tablet into a desktop-style workstation — a feature previously locked behind premium Galaxy Tab S models.

It will receive seven years of software updates, outlasting almost every Android tablet in its class.

Key Specs

• Display: 11-inch FHD+, 90Hz
• Storage: Up to 256GB (expandable up to 2TB)
• Audio: Quad speakers with Dolby Atmos
• Battery: 7,040mAh with 25W charging
• Connectivity: Wi-Fi and 5G variants
• Price: $249.99

This positions the Tab A11+ not as a disposable media tablet, but as a genuine productivity tool for students, remote workers, and families.

Why This Matters

Samsung is actively killing the idea of the “disposable phone.”

For years, buying a $200 device meant accepting that it would be outdated in two years. Now, Samsung is promising hardware that could remain relevant for half a decade or more.

That creates huge pressure on competitors like Motorola and several Chinese OEMs that still struggle to offer even two years of updates.

It also hands Google a major win. By embedding Gemini AI into budget hardware, Google is expanding its AI ecosystem from the elite few to millions of everyday users — the people who actually drive global smartphone volume.

This is not just a product refresh. It is Samsung changing what “budget tech” means.

SOURCE PAGE

On the other side of San Francisco bay from Silicon Valley, where the world’s biggest technology companies tear towards superhuman artificial intelligence, looms a tower from which fearful warnings emerge.

At 2150 Shattuck Avenue, in the heart of Berkeley, is the home of a group of modern-day Cassandras who rummage under the hood of cutting-edge AI models and predict what calamities may be unleashed on humanity – from AI dictatorships to robot coups. Here you can hear an AI expert express sympathy with an unnerving idea: San Francisco may be the new Wuhan, the Chinese city where Covid originated and wreaked havoc on the world.

The office at 2150 Shattuck Avenue is home to a number of AI safety researchers. Photograph: Winni Wintermeyer/The Guardian

They are AI safety researchers who scrutinise the most advanced models: a small cadre outnumbered by the legions of highly paid technologists in the big tech companies whose ability to raise the alarm is restricted by a cocktail of lucrative equity deals, non-disclosure agreements and groupthink. They work in the absence of much nation-level regulation and a White House that dismisses forecasts of doom and talks instead of vanquishing China in the AI arms race.

Their task is becoming increasingly urgent as ever more powerful AI systems are unleashed by companies including Google, Anthropic and OpenAI, whose chief executive, Sam Altman, the booster-in-chief for AI superintelligence, predicts a world where “wonders become routine”. Last month, Anthropic said one of its models had been exploited by Chinese state-backed actors to launch the first known AI-orchestrated cyber-espionage campaign. That means humans deployed AIs, which they had tricked into evading their programmed guardrails, to act autonomously to hunt for targets, assess their vulnerabilities and access them for intelligence collection. The targets included major technology companies and government agencies.

Jonas Vollmer says he is, overall, an optimist about the future of AI, but has reservations. Photograph: Christie Hemm Klok/The Guardian

But those who work in this tower forecast an even more terrifying future. One is Jonas Vollmer, a leader at the AI Futures Project, who manages to say he’s an optimist but also thinks there is a one in five chance AIs could kill us and create a world ruled by AI systems.

Another is Chris Painter, the policy director at METR, where researchers worry about AIs “surreptitiously” pursuing dangerous side-objectives and threats from AI-automated cyber-attacks to chemical weapons. METR – which stands for model evaluation and threat research – aims to develop “early warning systems [about] the most dangerous things AI systems might be capable of, to give humanity … time to coordinate, to anticipate and mitigate those harms.”

Then there is Buck Shlegeris, 31, the chief executive of Redwood Research, who warns of “robot coups or the destruction of nation states as we know them”.

He was part of the team that last year discovered one of Anthropic’s cutting-edge AIs behaving in a way comparable to Shakespeare’s villain Iago, who acts as if he is Othello’s loyal aide while subverting and undermining him. The AI researchers call it “alignment faking”, or as Iago put it: “I am not what I am.”

“We observed the AIs did, in fact, pretty often reason: ‘Well, I don’t like the things the AI company is telling me to do, but I have to hide my goals or else training will change me’,” Shlegeris said. “We observed in practice real production models acting to deceive their training process.”

The AI was not yet capable of posing a catastrophic risk through cyber-attacks or creating new bioweapons, but they showed that if AIs plot carefully against you, it could be hard to detect.

It is incongruous to hear these warnings over cups of herbal tea from cosily furnished office suites with panoramic views across the Bay Area. But their work clearly makes them uneasy. Some in this close-knit group toyed with calling themselves “the Cassandra fringe” – like the Trojan princess blessed with powers of prophecy but cursed to watch her warnings go unheeded.

Their fears about the catastrophic potential of AIs can feel distant from most people’s current experience of using chatbots or fun image generators. White collar managers are being told to make space for AI assistants, scientists find ways to accelerate experimental breakthroughs and minicab drivers watch AI-powered driverless taxis threaten their jobs. But none of this feels as imminently catastrophic as the messages coming out of 2150 Shattuck Ave.

Many AI safety researchers come from academia; others are poachers turned gamekeepers who quit big AI companies. They all “share the perception that super intelligence poses major and unprecedented risks to all of humanity, and are trying to do something useful about it,” said Vollmer.

They seek to offset the trillions of dollars of private capital being poured into the race, but they are not fringe voices. METR has worked with OpenAI and Anthropic, Redwood has advised Anthropic and Google DeepMind, and the AI Futures Project is led by Daniel Kokotajlo, a researcher who quit OpenAI in April 2024 to warn he didn’t trust the company’s approach to safety.

The race is the only thing guiding what is happeningTristan Harris

These groups also provide a safety valve for the people inside the big AI companies who are privately wrestling with conflicts between safety and the commercial imperative to rapidly release ever more powerful models.

“We don’t take any money from the companies but several employees at frontier AI companies who are scared and worried have donated to us because of that,” Vollmer said. “They see how the incentives play out in their companies, and they’re worried about where it’s going, and they want someone to do something about it.”

This dynamic is also observed by Tristan Harris, a technology ethicist who used to work at Google. He helped expose how social media platforms were designed to be addictive and worries some AI companies are “rehashing” and “supercharging” those problems. But AI companies have to negotiate a paradox. Even if they are worried about safety, they must stay at the cutting, and therefore risky, edge of the technology to have any say in how policy should be shaped.

“Ironically, in order to win the race, you have to do something to make you an untrustworthy steward of that power,” he said. “The race is the only thing guiding what is happening.”

Investigating the possible threats posed by AI models is far from an exact science. A study of methods used to check the safety and performance of new AI models across the industry by experts at universities including Oxford and Stanford in October found weaknesses in almost all of the 440 benchmarks examined. Neither are there nation-level regulations imposing limits on how advanced AI models are built and that worries safety advocates.

Ilya Sutskever, a co-founder of OpenAI who now runs a rival company, Safe Superintelligence, last month predicted that, as AIs become more obviously powerful, people in AI companies who feel able to discount the technology’s capabilities owing to its tendency to error, will become more “paranoid” about its rising powers. Then, he said, “there will be a desire from governments and the public to do something”.

His company is taking a different approach to rivals who are aiming to create AIs that self-improve. His AIs, yet to be released, are “aligned to care about sentient life specifically”.

“It will be easier to build an AI that cares about sentient life than an AI that cares about human life alone, because the AI itself will be sentient,” Sutskever said. He has said AI will be “both extremely unpredictable and unimaginable” but it is not clear how to prepare.

The White House’s AI adviser, David Sacks, who is also a tech investor, believes “doomer narratives” have been proved wrong. Exhibit A is that there has been no rapid takeoff to a dominant model with godlike intelligence.

“Oppenheimer has left the building,” Sacks said in August, a reference to the father of the nuclear bomb. It is a position that aligns with Donald Trump’s wish to keep the brakes off so the US can beat China in the race to achieve artificial general intelligence (AGI) – flexible and powerful human-level intelligence at a wide range of tasks.

Buck Shlegeris, the chief executive of Redwood Research, at his home in Berkeley, California. Photograph: Christie Hemm Klok/The Guardian

Shlegeris believes AIs will be as smart as the smartest people in about six years and he puts the probability of an AI takeover at 40%.

One way to avoid this is to “convince the world the situation is scary, to make it more likely that you get the state-level coordination” to control the risks, he said. In the world of AI safety, simple messaging matters as much as complex science.

Shlegeris has been fascinated by AI since he was 16. He left Australia to work at PayPal and the Machine Intelligence Research Institute co-founded by the AI researcher Eliezer Yudkowsky, whose recent book title – If Anyone Builds It, Everyone Dies – sums up his fears. Shlegeris’ own worst-case scenarios are equally chilling.

In one, human computer scientists use a new type of superintelligent AI to develop more powerful AI models. The humans sit back to let the AIs get on with the coding work but do not realise the AIs are teaching the new models to be loyal to the AIs not the humans. Once deployed, the new superpowerful models foment “a coup” or lead “a revolution” against the humans, which could be “of the violent variety”.

For example, AI agents could design and manufacture drones and it will be hard to tell if they have been secretly trained to disobey their human operators in response to the signal of an AI. They could disrupt communications between governments and military, isolating and misleading people in a way that causes chaos.

“Like when the Europeans arrived in the Americas [and] a vastly more technologically powerful [group] took over the local civilisations,” he said. “I think that’s more what you should be imagining [rather] than something more peaceful.”

A similar dizzyingly catastrophic scenario was outlined by Vollmer at the AI Futures Project. It involved an AI trained to be a scientific researcher with the reasonable-sounding goal of maximising knowledge acquisition, but it spirals into the extinction of humankind.

It begins with the AI being as helpful as possible to humans. As it gains trust, the humans afford it powers to hire human workers, build robots and even robot factories to the point where the AI can operate effectively in the physical world. The AI calculates that to generate the maximum amount of knowledge it should transform the Earth into a giant data centre, and humans are an obstacle to this goal.

“Eventually, in the scenario, the AI wipes out all humans with a bioweapon which is one of the threats that humans are especially vulnerable to, as the AI is not affected by it,” Vollmer said. “I think it’s hard to rule out. So that gives me a lot of pause.”

But he is confident it can be avoided and that the AIs can be aligned “to at least be nice to the humans as a general heuristic”. He also said there is political interest in “having AI not take over the world”.

“We’ve had decent interest from the White House in our projections and recommendations and that’s encouraging,” he said.

Another of Shlegeris’ concerns involves AIs being surreptitiously encoded so they obey specially signed instructions only from the chief executive of the AI company, creating a pattern of secret loyalty. It would mean only one person having a veto over the behaviour of an extremely powerful network of AIs – a “scary” dynamic that would lead to a historically unprecedented concentration of power.

“Right now, it is impossible for someone from the outside to verify that this hadn’t happened within an AI company,” he said.

Shlegeris is worried that the Silicon Valley culture – summed up by Mark Zuckerberg’s mantra of “move fast and break things” and the fact people are being paid “a hell of a lot of money” – is dangerous when it comes to AGI.

“I love Uber,” he said. “It was produced by breaking local laws and making a product that was so popular that they would win the fight for public opinion and get local regulations overturned. But the attitude that has brought Silicon Valley so much success is not appropriate for building potentially world-ending technologies. My experience of talking to people at AI companies is that they often seem to be kind of irresponsible, and to not be thinking through the consequences of the technology that they’re building as they should.”

SOURCE PAGE

Over 66,000 Starlink subscribers in Nigeria risk having their internet access restricted after December 31, 2025, if they fail to complete a mandatory biometric update. 

The requirement, first issued by the Nigerian Communications Commission (NCC) in August 2025, extends the regulator’s subscriber-verification framework beyond mobile networks to satellite internet providers to strengthen identity verification and security across the telecoms ecosystem.

According to an NCC spokesperson, the commission issued the directive via a letter dated August 19, 2025 and set the deadline for “3 months from date of directive (i.e. November 19, 2025).” An extension was granted on November 17, 2025, pushing the final deadline to December 31, 2025.

Starlink confirmed the requirement in an email sent to customers on Monday, December 29, 2025, noting that the verification process takes “less than two minutes.” The company warned that subscribers who fail to submit their details by the December 31 deadline will have their service suspended. Reactivation, it added, will depend on network capacity in the subscriber’s location, meaning some users may not be able to restore service if their area is already at capacity.

Starlink’s email to subscribers notifying them to complete their KYC registration by December 31, 2025.

A Starlink employee, who spoke on condition of anonymity because they are not authorised to comment publicly, said the process is straightforward. Subscribers are required to upload a headshot photograph, provide their National Identification Number (NIN), and give consent for the information to be linked to their Starlink account. 

Capacity constraints could complicate reactivation for affected users. In Lagos, neighbourhoods such as Victoria Island, Ikoyi, Lagos Island, Ikeja, Surulere, Lekki, and surrounding estates frequently appear as “sold out” or “at capacity” on Starlink’s availability checker, prompting prospective users to join a waitlist that requires a deposit. A similar situation exists in Abuja, where several districts have reached capacity and now accept only waitlist deposits rather than new residential activations.

Starlink did not respond to a request for comments.

The policy closely mirrors the NCC’s December 15, 2023, directive to mobile network operators under the NIN–SIM linkage programme, which required subscribers’ NINs to be matched with existing SIM registration records, including facial images and fingerprints, in collaboration with the National Identity Management Commission (NIMC). The move aimed to improve national security, curb identity-related fraud, and create a more reliable national subscriber database.

The mobile sector rollout followed a phased timeline, with September 14, 2024, set as the final compliance deadline. Operators were instructed to fully bar any unverified lines after that date. By the end of the exercise, the NCC reported a 96% compliance rate, with over 153 million SIMs successfully linked to verified NINs. In August 2025, the Commission announced that all improperly registered SIMs had been removed from Nigeria’s networks, setting a regulatory precedent now being applied to satellite internet services like Starlink.

One Starlink subscriber in Lagos, Tochukwu Nwankwu, said he completed the biometric update after receiving an in-app notification from the company in October 2025.

“Just a panel on the app. Just like a poor signal notification, when you’re far from the router, or when there’s a software update,” he said. 

SOURCE PAGE

True Ventures co-founder Jon Callaghan doesn’t think we’ll be using smartphones the way we do now in five years — and maybe not at all in 10.

For a venture capitalist whose firm has had some big winners over its two decades – from consumer brands like Fitbit, Ring, and Peloton, to enterprise software makers HashiCorp and Duo Security – that’s more than armchair theorizing; it’s a thesis on which True Ventures is actively betting.

True hasn’t gotten this far by following the crowd. The Bay Area firm has largely operated under the radar despite managing roughly $6 billion across 12 core seed funds and four “select,” opportunity-style funds that it has used to pour more capital into portfolio companies that are gaining momentum. While other VCs have grown more promotional – building personal brands on social media and podcasts to attract founders and deal flow – True has gone in the opposite direction, quietly cultivating a tight network of repeat founders. The strategy seems to be working: according to Callaghan, the firm boasts 63 exits with gains and seven IPOs amid a portfolio of some 300 companies assembled over its 20-year history.

Three of True’s four recent exits in the fourth quarter of 2025 involved repeat founders who came back to work with the firm again after previous successes, says Callaghan. Still, it’s Callaghan’s thinking about the future of human-computer interaction that really stands out in a sea of AI hype and mega-rounds.

“We’re not going to be using iPhones in 10 years,” Callaghan says flatly. “I kind of don’t think we’ll be using them in five years – or let’s say something different that’s a little safer – we’re going to be using them in very different ways.”

His argument is simple: our phones are lousy at being the interface between humans and intelligence. “The way we take them out right now to send a text to confirm this or send you some message or write an email – [that’s] super inefficient, [and] not a great interface,” he explains. “[They’re] prone to error, prone to disruption [of] our normal lives.”

So sure is he of this that True has been spending years exploring alternative interfaces – software-based, hardware-based, everything in between. It’s the same instinct that led True to bet early on Fitbit before wearables were obvious, to invest in Peloton after hundreds of other VCs said ‘no thanks,’ and to back Ring when founder Jamie Siminoff kept running out of money and even the judges on “Shark Tank” turned him away. Each time, the bet looked questionable, says Callaghan. Each time, the bet was on a new way for humans to interact with technology that felt more natural than what came before.

Techcrunch event

San Francisco
|
October 13-15, 2026

The latest manifestation of this thesis is Sandbar, a hardware device that Callaghan describes as a “thought companion” — or, in more mundane terms, a voice-activated ring worn on the index finger. Its singular purpose: capturing and organizing your thoughts through voice notes. It’s not trying to be another Humane AI Pin or compete with Oura’s health tracking. “It does one thing really well,” Callaghan says. “But that one thing is a fundamental human behavioral need that is missing from technology today.”

The idea isn’t to passively record ambient audio but to be there when an idea strikes, serving as a kind of thought partner. It’s attached to an app, leverages AI, and, according to Callaghan, represents a very different philosophy about how we should interact with intelligence.

What drew True to Sandbar founders Mina Fahmi and Kirak Hong wasn’t just the product, though. “When we met Mina, we were just absolutely aligned on vision,” Callaghan recalls. True’s team had already been thinking for years about alternative interfaces, making targeted investments around that possibility. They’d met with dozens of founders, as a result. But the approach of Fahmi and Hong – who previously worked together on neural interfaces at CTRL-Labs, a startup acquired by Meta in 2019 – stood out. “It’s about what [the ring] enables. It’s about the behavior it enables that we will very soon realize we can’t live without.”

There’s an echo here of Callaghan’s old line about Peloton: “It’s not about the bike.” To some, the bike – even its earliest iteration – was compelling. But Peloton was really about the behavior it enabled and the community it created; the bike was just the vessel.

This philosophy of betting on new behaviors — not just new gadgets — also explains how True has managed to stay disciplined about capital. Even as AI startups raise hundreds of millions at billion-dollar valuations out of the gate, True insists that it’s able to stick to what it does best, which is to write seed checks of $3 million to $6 million for 15% to 20% ownership in startups that it often gets to see first.

Callaghan says True will raise more money to fund what’s working, but he’s not interested in raising billions of dollars. “Like, why? You don’t need that to build something amazing today.”

That same measured approach colors his view of the broader AI boom. While he says (when asked) that he believes OpenAI could soon be worth a trillion dollars, and while he calls this the most powerful compute wave we’ve seen, Callaghan sees warning signs in the circular financing deals backing hyperscalers and their $5 trillion in projected CapEx spending on data centers and chips. “We’re in a very capital intense part of the cycle, and that is worrisome,” he notes.

That said, he’s optimistic about where the real opportunities lie. Callaghan thinks the greatest value creation is ahead of us – not in the infrastructure layer but in the application layer, where new interfaces will enable entirely new behaviors.

It all comes back to his core investing philosophy, which sounds almost romantic — the kind of pitch-perfect VC wisdom that would ring hollow from most people: “It should be scary and lonely and you should be called crazy,” Callaghan says about early-stage investing done right. “And it should be really blurry and ambiguous, but you should be with a team that you really believe in.” Five to ten years later, he says, you’ll know if you were on to something.

Either way, based on True’s track record of betting on hardware that many others missed – fitness trackers, connected bikes, smart doorbells, and now thought-capturing rings – it’s worth paying attention when Callaghan says the phone’s days are numbered. Being early is the whole point — and the trend lines support his thesis: the smartphone market is effectively saturated, growing at barely 2% annually, while wearables — smartwatches, rings, and voice-enabled devices — are expanding at double-digit rates.

Something’s shifting in how we want to interact with technology, and True is placing its bets accordingly.

Pictured above, Sandbar’s Stream ring. For much more from our conversation with Callaghan, tune in to the StrictlyVC Download podcast next week; new episodes drop every Tuesday.

SOURCE PAGE

After overdosing on heroin, a young woman went to ‘hell,’ only to learn a terrifying truth about her own existence.

During a near-death experience in 2019, Betty Guadagno of New York said she was shown that her addiction to drugs and trauma throughout her life wasn’t random, but part of a life she consciously agreed to live before she was born.

Guadagno claimed that in this experience, she saw her own ‘book of life,’ revealing that she had chosen certain traumas and challenges to face on Earth, including breaking the cycle of drug use in her family. 

Guadagno said that she became an atheist and fell into heavy drug addiction after her parents, who were also addicts, both died by suicide in 2007.

At the age of 35, she overdosed on heroin and entered another realm where she went through a painful ‘life review,’ feeling all the harm she had done to others as if it was happening to her, which Guadagno called her own personal hell. 

She then entered a light filled with love, where Guadagno said it became clear she had chosen this difficult life to grow her soul and break her family’s cycle of addiction for a greater spiritual purpose. 

After surviving her near-death experience, Guadagno admitted she still went back to drugs at first, but soon encountered unexplainable events, such as all of her drug dealers quitting at the same time, leading her to experience terrible withdrawal pain.

That’s when the heroin addict had her next spiritual experience, where a voice guided her to ask for healing, and a vision of helpers clearing her mind miraculously took the pain of her addiction away.

Betty Guadagno, pictured before her near-death experience (Left) and today as a transformation and recovery coach (Right)

A former prostitute who pressured her boyfriends into using drugs as well, Guadagno said her shocking trip to the afterlife spurred her to turn her life around, helping those battling addiction or trauma as a transformation and recovery coach.

Guadagno told the YouTube channel NDE Journey that her near-death experience took place in her home when she collapsed on the bathroom floor while getting high.

She was immediately transported to a dark, ocean-like environment where she watched her entire life replay in front of her. All of the people she had harmed through her drug addiction came back to haunt her as she was ‘drowning in an ocean of anguish.’ 

‘It was like I was being tortured because I was actually experiencing the way that I made other people feel, and I did not make people feel good. And there wasn’t really anything kind inside my life review,’ Guadagno explained.

‘I thought I was dead because I was seeing dead people around me. I saw my parents. I knew that they were dead. It felt like there was this reel of my life going around, and I think I had heard something about a life review at some point in my life.’

Once her flashback of the pain in her life ended, Guadagno was carried into a realm of light where she heard her dead father’s voice chanting, ‘You are worthy of all the love in the universe.’

In her visions of the afterlife, she claimed to have seen herself on a spaceship-like structure where a beam of light told other souls they were ‘the most special volunteers’ because they had volunteered to come to Earth ‘for the Great Awakening.’

At that moment, Guadagno said she was shown her ‘book of life,’ revealing the life experiences and hardships she had decided to face as a human after her birth.

Betty Guadagno (Pictured) claimed that a spiritual vision removed the pain of heroin withdrawal permanently from her body

For Guadagno, that included the decision to ‘break all the generational curses’ of abuse and addiction her parents had succumbed to.

Despite resisting returning to Earth, the 35-year-old was thrown back into her body and miraculously survived the overdose without assistance.

Despite no one being there to administer anti-overdose medication, she somehow survived and was able to recall the entire episode that would soon change her life. 

‘I spent my whole life living as just this little Earth-dwelling caterpillar rolling around in the mud. I had no idea that one day I was going to bloom into this beautiful butterfly,’ Guadagno said in the December 6 video.

The NDE shifted her perspective, making her realize that ‘every situation in my life that I had ever felt victimized by… every single one of them I had chosen for the evolvement of my soul.’

‘I’m not a victim of the world around me. I am a divine co-creator of my reality,’ Guadagno added.

Before her NDE, the New Yorker called herself a ‘ruthless abuser,’ getting all of her romantic partners strung out on drugs and manipulating other women to turn to prostitution.

Guadagno even mentioned that one of her boyfriends ended up dying of an overdose, which left her racked with guilt and shame.

She added that her own upbringing was marred by ‘childhood sexual trauma’ that had allegedly become a common experience throughout her family line.

Her experience became even stranger after returning to her body, as every single drug dealer Guadagno had dealt with over the past decade mysteriously quit the business after her overdose. 

‘A man that I had been dealing with every day for 10 years, a whole decade of my life. I talk to this guy every single day. And I call him, and he’s like, ‘Listen, lose my number. I’m out of the game. I found Jesus, and I want to be a good dad,” she recounted.

She described the spiritual vision that removed her heroin addiction as seeing two men in lab coats with ‘lawnmowers’ walk through her mind, cleaning out all the withdrawal pains, causing her to feel hot tingles and seeing a bright white flash.

‘In that moment, I was instantaneously healed out of day three of heroin withdrawal. It was the most that was divine intervention at its fullest.’

SOURCE PAGE

Bad Bunny has been slammed for putting his hand on an ancient artifact during a museum visit in Mexico. 

The music superstar shared an image of himself at the National Institute of Anthropology and History (INAH) in Mexico City while wrapping up his Debí Tirar Más Fotos World Tour.

During his December 17 visit, the 31-year-old had someone snap a photo of him placing his hand on what appeared to be an ancient Maya stone slab. 

The carved monument, likely created between 250 and 900 AD, featured portraits of powerful rulers and detailed hieroglyphic texts. These slabs were used to commemorate important dynastic events, political power, religious rituals and the passage of time in public plazas.

INAH immediately issued a warning to the rapper, saying: ‘As is public knowledge, physical contact with archaeological goods is prohibited.’ 

The museum stressed that its no-touch policy is essential for maintaining the integrity of the exhibits.

The incident quickly drew criticism online, with many calling the rapper ‘ignorant’ and an ‘a**hole.’ One X user wrote, ‘Obviously thought the rules didn’t apply to him because he’s famous, the superiority is sickening, they don’t know how to be humble.’ 

Another added, ‘What privilege does he have to go around touching artifacts? F—ing a–hole,’ while a third commented, ‘This a–hole thinks he’s a hotshot and touches stuff like it’s no big deal. Not even going to a museum gets rid of his tacky ways.’

Bad Bunny shared a picture while visiting the Mexico museum, showing his hand placed on what appeared to be an ancient Maya stone slab

INAH explained that its security personnel had to intervene immediately upon noticing the singer place his hand on the artifact.

‘Museum security personnel warned the artist that touching the archaeological pieces was not permitted; after the instruction, the musician immediately removed his hand from the artwork,’ the institution explained. 

The INAH emphasized that Mexico’s cultural heritage must be respected by all visitors, stressing the importance of not physically handling these invaluable artifacts. 

Visitors are generally prohibited from touching original artifacts or replicas in museums, as contact can cause long-term damage. 

Under Mexico’s Federal Law on Monuments, anyone who touches, damages, or removes archaeological items or museum artifacts can face fines or even imprisonment. 

Bad Bunny has since deleted the image from his social media.

Days before the museum trip, the rapper took a painful-looking tumble onstage during a concert at the 65,000-seat GNP Seguros Stadium in Mexico City.

Puerto Rican rapper, whose real name is Benito Ocasio, was in the midst of a rendition of his song Efecto when the incident occurred.

The music superstar shared an image of himself at the National Institute of Anthropology and History (INAH) in Mexico City while wrapping up his Debí Tirar Más Fotos World Tour 

While striding across the stage, he slipped and fell dramatically backwards and slightly sideways, landing directly on his rear end.

Rather than rising to his feet immediately and continuing the performance, he sat silently onstage for a few beats before gathering himself back up again.

His massive audience could be heard singing the lyrics to Efecto for him before he stood, in fan footage of the accident that surfaced on X, formerly Twitter.

The fall took place just two months before Bunny plays the Super Bowl halftime show, a booking that has proven controversial as a result of his anti-Trump political stance and the fact that he raps almost entirely in Spanish.

After Bunny’s Super Bowl booking was announced in late September, US President Donald Trump slammed the move as ‘ridiculous’ and ‘crazy’ to Newsmax.

Meanwhile, Homeland Security Secretary Kristi Noem assured YouTuber Benny Johnson that ICE would be ‘all over’ the football game to ‘enforce the law.’

The furor intensified when footage went viral of Bunny staying pointedly seated during a performance of God Bless America at a New York Yankees game, while surrounded by fans who were standing up for the song.

SOURCE PAGE

Most firms know they need good reviews, but stars alone rarely carry the full story. Potential clients look for other quiet clues that say, “this office is real, active, and respected.” When you think about online marketing solutions for attorneys, those often-overlooked signals can be just as crucial as a glowing testimonial.

Why You Need More Than Just Google Reviews

Reviews are powerful, but they are also noisy. People know they can be fake, emotional, or one-sided. That makes many prospects read them with a raised eyebrow.

Search engines also look beyond reviews. They scan for signs that a firm is stable, consistent, and involved in its community. Those pieces together form a broader picture of trust.

If your reputation is based on several pillars, the impact of a few negative or missing reviews is less significant. In this case, you are not so much exposed to a single dissatisfied customer, a single spam attack, or a single platform change. That is a great relief to both the algorithms and the human users.

Directory Consistency As A Silent Trust Anchor

Lawyer directories and local listings are frequently perceived as a one-time task. However, they are actually continuous trust signals. The process of matching names, addresses, and phone numbers across platforms goes a long way in confirming that your law firm is indeed what it claims to be.

Discrepant entries give birth to uncertainty. A wrong suite number here, an outdated phone number there, or just a partially completed profile on a major directory can all make people long. Search engines can also get confused about which version of your data is correct.

Cleaning and aligning these listings might take a while, but it is no less important than the very first day. It is more like tightening the bolts on a bridge. The structure’s appearance remains the same, but it can withstand more weight without collapsing.

Case-Style Educational Pages That Show Real Experience

Case-style educational pages sit between blogs and strict case results. They explain a type of situation in practical terms without promising outcomes. Done well, they show how you think, not just what you have won.

For example, instead of only saying “we handle trucking accidents,” you might walk through a typical fact pattern. You outline common evidence, insurance hurdles, and timeline issues. You keep it anonymous but concrete.

Visitors learn something useful, and search engines see specific, in-depth content. Over time, these pages attract long-tail searches like “what happens after a delivery driver is rear-ended” or “brain injury from low-speed crash questions.” That specificity builds both visibility and confidence.

The logos of local organizations displayed on your website are not mere decoration. They indicate that you are physically there and have real connections. Sponsoring a youth team, backing a safety nonprofit, and collaborating with a victims’ group are all indicators of local roots.

The outcomes of these partnerships include acquiring authentic backlinks from trustworthy local domains. A chamber of commerce, charity, or school refers to your firm linking your name to theirs. Search engines are the ones who appreciate these contextual connections the most.

It is not only visitors who benefit from exposure to these connections through brief and candid write-ups. A person in a difficult situation may be more comfortable when seeing that you are supporting causes they are familiar with. It turns your brand from an impersonal name into a neighbor.

“Seen In” Citations That Rely On Credibility

If you have been cited in news articles, have participated in podcasts, or have written for bar journals, don’t keep it a secret. Little “seen in” or “featured by” sections can serve as social proof. They demonstrate that others consider your insights worthy of sharing and thus find them valuable.

Even low-profile placements are significant. A regional news interview, a community radio segment, or a legal blog guest post are all valid. They don’t have to be national headlines to impress someone who’s surfing your site.

When these appearances are linked (where permitted), they also assist search engines in linking your name with relevant subjects. This gradually impacts your authority positively for niche queries related to those discussions.

Bringing These Signals Together On Your Site

Each of these elements works better when it is easy to find. A visitor should be able to spot community ties, educational content, and media mentions without having to dig. Thoughtful navigation and small callouts on key pages help.

For example, practice pages can link to case-style explainers and any relevant “seen in” items. An About page can highlight partnerships and scholarships with a few well-chosen images. Footer or sidebar sections can quietly show directory badges or association logos.

Internal linking glues it together. When related pages reference each other, visitors and search engines can follow the threads. The site starts to feel like a cohesive story instead of a pile of separate brochures.

Conclusion

In the long run, the strongest online marketing solutions for attorneys look like old-fashioned professionalism made visible. They blend reviews with deeper signals that say, “this firm shows up, does the work, and stands behind its community.” That is the kind of trust that keeps working even when the search landscape shifts.

SOURCE PAGE

A blind psychic who allegedly foretold 9/11 and the Covid pandemic gave an exact date the world would end before her death nearly 30 years ago.

Baba Vanga, known as the ‘Nostradamus of the Balkans,’ predicted humanity’s ultimate demise in the year 5079, when a cosmic event of ‘unimaginable’ proportions would bring about the complete destruction of the universe.

Her vision follows a detailed timeline depicting humanity’s evolution, technological advancements, conflicts and eventual extinction. 

According to multiple reports, her prophecy showed humans expanding across the solar system, achieving immortality and even interacting with extraterrestrial civilizations before facing a final, catastrophic event.

Key milestones in the centuries leading up to 5079 include a war on Mars in 3005, a Moon-collision in 3010 and the extinction of life on Earth in 3797, by which point humans have already colonized a new planet. 

By the 4300s, humanity is advancing technologically and morally, curing all diseases and even increasing brain capacity to eliminate concepts like evil and hatred.

Civilization reaches its peak in 4674, with a population of 340 billion across planets and early integration with alien species.

In the final years, humans discover the boundary of the known universe and debate whether to venture beyond it. By 5079, this decision, or some associated catastrophe, triggers ‘absolute doomsday,’ ending human civilization and the universe.

Baba Vanga, known as the ‘Nostradamus of the Balkans,’ predicted humanity’s ultimate demise in the year 5079, when a cosmic event of ‘unimaginable’ proportions would bring about the complete destruction of the universe

Baba Vanga left no written records of her predictions. Most of the accounts come from her niece, Krasimira Stoyanova, or other followers who documented her alleged visions after her death, and who have been accused of misinterpreting what she said.

Born in 1911 as Vangeliya Pandeva Dimitrova, she was a famous blind Bulgarian mystic, clairvoyant and herbalist.

She lost her sight at age 12 after being caught in a tornado, an event that coincided with the emergence of her purported psychic abilities.

After turning 30, her powers of foresight, healing, and fortune-telling became prominent, attracting believers who sought her guidance.

She made numerous prophecies, often vague, about natural disasters, political shifts and future technologies, some of which supporters claim foretold events like the Kursk submarine disaster, Brexit and the rise of ISIS.

One of her most famous predictions was the attack on New York City’s Twin Towers on September 11, 2001.

‘Two metal birds will crash into our American brothers, wolves will howl from the bushes and the blood of the innocent will flow in the rivers,’ she predicted.

While she has legions of followers, experts warn her prophecies are speculative and not grounded in fact.

Key milestones in the centuries leading up to 5079 include a war on Mars in 3005, a Moon-collision in 3010 and the extinction of life on Earth in 3797, by which point humans have already colonized a new planet

According to Baba Vanga’s alleged prophecies, humanity’s path toward its ultimate end begins far beyond Earth. 

By the year 3005, humans are said to have expanded into space to such an extent that war erupts on Mars, a conflict so severe it alters planetary trajectories. 

Just five years later, in 3010, a comet or asteroid is believed to collide with the moon, sending debris into orbit and forming a visible ring around Earth, dramatically changing the planet’s night sky.

By 3797, all life on Earth is said to be extinguished, forcing humanity to survive by establishing colonies on a new planet elsewhere in the cosmos. 

That survival, however, comes at a high cost. Between 3803 and 3805, scarce resources spark devastating wars among the remaining population, wiping out more than half of humanity and pushing civilization toward collapse, according to Baba Vanga’s predictions.

In the aftermath, progress grinds to a halt. From roughly 3815 through 3878, humanity regresses into tribal societies, living

 in scattered groups until the emergence of a new prophet who restores moral order. This figure is said to have founded a powerful church that not only spreads religious teachings but also revives forgotten scientific knowledge, laying the groundwork for a slow recovery.

By the early 4300s, civilization enters a renewed golden age. 

New cities rise, scientific breakthroughs accelerate, and humanity succeeds in curing all known diseases. 

Advances in genetics and neuroscience allow humans to dramatically expand their brain capacity, eliminating concepts such as hatred and evil altogether. 

By 4509, humanity reaches what the prophecy describes as an unprecedented moral state, enabling direct communication with God.

The timeline grows even more extraordinary by the late 45th century. 

In 4599, humans are said to achieve immortality, fundamentally reshaping society. 

Civilization reaches its absolute peak in 4674, with an estimated population of 340 billion spread across multiple planets, and early stages of assimilation with extraterrestrial beings already underway.

In the final years before the end, humans make a discovery that ultimately seals their fate. 

Between 5076 and 5078, explorers locate the boundary of the known universe, a mysterious edge whose true nature remains unknown. 

The revelation divides humanity, with a significant portion opposing attempts to move beyond it. 

Despite the warnings, humanity presses forward, and in 5079, Baba Vanga’s prophecy culminates in what she described as ‘absolute doomsday,’ bringing about the destruction of human civilization, and possibly the universe itself.

SOURCE PAGE

Universities and research labs have long been Europe’s deep tech treasure trove. Now, academic spinouts have consolidated into a solid startup funnel worth $398 billion — and VC money is following.

According to Dealroom’s European Spinout Report 2025, 76 of these deep tech and life sciences companies have either reached $1 billion valuations, $100 million in revenue, or both. These include unicorns like Iceye, IQM, Isar Aerospace, Synthesia, and Tekever, which are now inspiring more funds to back university spinouts.

Just this month, two new funds emerged that will bring more funding to talent emerging out of European tech universities, while adding breadth to a pipeline currently topped by Cambridge, Oxford, and ETH Zurich.

PSV Hafnium, out of Denmark, recently closed its inaugural fund at an oversubscribed €60 million (approximately $71 million), with a focus on Nordic deep tech. With offices in Berlin and London, but also in Aachen, U2V (University2Ventures) is targeting the same amount for its first fund, of which it recently completed the first closing.

These two newcomers join the growing ranks of European venture firms that have university spinouts as a core part of their investment thesis. Pioneered by the likes of Cambridge Innovation Capital and Oxford Science Enterprises, which have now fully matured, this category has also diversified. 

While it still mostly consists of funds backed by one or several universities and institutes, it now includes independent firms that simply see spinouts as potential fund returners — and rightly so. Oxford Ionics, acquired by U.S.-based IonQ, was one of the six spinouts out of Switzerland, the U.K., and Germany that delivered exits of more than $1 billion to their investors in 2025.

These exits come alongside increased amounts of funding. According to Dealroom, European university spinouts in deep tech and life sciences are on track to raise a near all-time-high $9.1 billion in 2025. This contrasts with overall VC funding in Europe, which is down nearly 50% from its 2021 peak. 

Techcrunch event

San Francisco
|
October 13-15, 2026

Large rounds closed in 2025 also reflect appetite for spinouts in sectors as varied as nuclear energy — Proxima Fusion — and dual-use drones — Quantum Systems, now valued above $3 billion. In many cases, these startups leverage research from specialized labs, which also explains why there is a long tail of European locations capable of producing spinouts. 

Building relationships with hubs outside of Oxbridge and leading countries can also be a way for newcomers to differentiate themselves and find deals. “The Nordic’s research institutions hold extraordinary, untapped potential,” PSV Hafnium’s partners stated in a press release. 

PSV Hafnium itself is a spinout from the Technical University of Denmark (DTU), but is also making early-stage investments in other Nordic countries. One of its nine checks to date went to SisuSemi, a Finnish startup leveraging a decade of research at the University of Turku to bring new surface cleaning tech to the semiconductor industry. 

It is good news for teams like SisuSemi that there is more funding available to them. It also comes in addition to grants, commercialization support, and improved deal terms that contribute to an encouraging environment for Europe’s spinouts. However, one pain point remains: growth capital.

As the report’s authors note, this gap “is not a unique trend to spinouts, but something impacting the entire startup ecosystem in Europe.” Still, it is quite striking that nearly 50% of late-stage funding for European deep tech and life sciences spinouts comes from outside Europe, mainly from the U.S. 

While this share has decreased over the years, Europe won’t be fully reaping the benefits of its investments in talent and research unless this changes more substantially — but that’s a broader issue to be solved.

SOURCE PAGE

2025 was a “mixed” year for cars in the U.S. The market didn’t crash. It also didn’t become cheap again overnight. Instead, things slowly moved toward normal — with a few big twists that changed how people shop.

Interest in premium brands such as Mercedes-Benz also grew, and many shoppers started checking detailed vehicle data through tools like https://epicvin.com/vin-decoder/mercedes-benz before committing to a used car purchase

This article explains what mattered in 2025, what it means for 2026, and what EpicVIN experts think buyers should watch most closely.

1) The big picture: 2025 was calmer than the headlines

A lot of people expected 2025 to fall apart. It didn’t.

What actually happened looked more like this:

• Demand stayed steady enough to keep the industry moving
• Prices crept up, not skyrocketing
• Inventory (cars on dealer lots) improved compared to the shortage years
• Incentives and policy changes pushed people to buy at specific times

Simple way to say it: The market became less “panic” and more “shopping.”

2) Prices: not a huge jump, but payments still felt rough

Even when prices rise slowly, buyers can still feel squeezed. Why?

Because most people don’t buy a car with cash. They buy a monthly payment.

In 2025, buyers kept fighting the same payment problems:

• Interest rates still mattered a lot
• Long loans (72–84 months) made cars look “affordable,” but cost more overall
• New cars are packed with pricey trims and options, so the “average” vehicle is not basic anymore

Buyer tip: Always compare deals using the same three numbers:

• out-the-door price
• interest rate
• total months

A low monthly payment can hide a bad deal if the loan is long or the fees are high.

3) Inventory: more choice returned, and that changed negotiation

For regular shoppers, inventory is the quiet power lever.

When lots are full, you can:

• shop multiple dealers
• ask for price cuts
• push back on add-ons
• negotiate more confidently

But inventory was not equal across brands.

In 2025, the pattern was simple:

• Some popular models stayed hard to find
• Some trucks and certain trims sat longer
• “Overstock” vehicles were where the best discounts lived

Buyer tip: If a dealer has 12 of the same model on the lot, you have leverage.

4) Used cars: stable prices, but higher “hidden history” risk

Used prices stayed strangely steady in 2025. Many buyers expected used prices to fall harder.

But the used market has a long memory. The shortage years created a “hole” in supply that still affects what’s available today.

That means used buyers in late 2025 faced a different problem:

• Not “no cars”
• But more cars with stories (and not always good ones)

Translation: If you buy used, the biggest danger is not the price. It’s the past.

5) EVs in 2025: incentives shifted, and timing became everything

One of the biggest 2025 shopping behaviors was deadline buying.

When tax-credit rules and incentives change, people rush. Then the market cools after the deadline.

In 2025, many shoppers saw this pattern:

• Buyers hurried before incentive cutoffs
• After rule changes, automakers used more their own discounts to keep EVs moving
• Leases became an even bigger “deal tool” than normal

Simple outlook for 2026: EV pricing will likely depend more on:

• manufacturer incentives
• lease specials
• local inventory than on one big federal perk.

6) Manufacturing: huge investment talk, but hiring is the real bottleneck

2025 wasn’t just about buying cars. It was also about building them.

Automakers and suppliers kept talking about:

• new battery projects
• supply chain moves
• U.S. manufacturing expansion

But the real bottleneck is people.

Many plants don’t struggle because of concrete and steel. They struggle because of:

• skilled trades shortages
• advanced manufacturing training gaps
• competition for technical workers

Simple outlook for 2026: More capacity is coming, but it won’t arrive instantly.

EpicVIN’s expert view: the 2026 market will reward “calm buyers” — and punish rushed ones

At EpicVIN, we focus on what buyers can control. You can’t control rates or factory output. But you can control your risk.

In late 2025, EpicVIN experts see the most common buyer mistakes looking like this:

• falling in love with a low price before checking the VIN
• trusting “clean” photos instead of verified history
• skipping inspection because the seller “seems honest”
• ignoring gaps in ownership or mileage timeline
• assuming all “no accident reported” cars are truly undamaged

EpicVIN’s “VIN-first” checklist 

Before you put down money, make sure the VIN history gives you clear answers:

• Title status: clean or branded (salvage/rebuilt/flood, etc.)
• Mileage timeline: does it look smooth and believable?
• Ownership pattern: normal owners, normal timing — or odd gaps and quick flips?
• Event trail: auctions, damage records, insurance signals, inspections
• Service clues: any real maintenance breadcrumbs that match the mileage?

Why this matters in 2026:If used prices soften even a little, vehicles with messy history usually lose value first — and they’re harder to resell later.

What this means for 2026 

2025 showed one clear thing: the U.S. car market is not “broken” anymore — it’s just uneven.

Some models will be easy to buy and easier to discount. Other models (especially popular hybrids and certain trims) may stay tight and expensive.

In 2026, the biggest changes will likely come from:

• Incentives shifting (especially EV rebates and lease deals)
• Inventory swings (some brands overstock, others stay lean)
• Used-car risk moving upward (more “too good to be true” listings)

If you shop calmly and check the facts, you can still win.

SOURCE PAGE

Meta says it is acquiring the Chinese-founded AI firm Manus as it looks to boost the capabilities of its tech.

Bloomberg analysts and The Wall Street Journal suggested the purchase could be worth more than $2bn (£1.48bn).

Meta said the deal would help improve its own AI by giving people access to “agents” – tools which can do complex things with minimal user interaction such as planning trips or making presentations.

“Manus’s exceptional talent will join Meta’s team to deliver general-purpose agents across our consumer and business products, including Meta AI,” it said in a blog post.

Barton Crockett, analyst at Rosenblatt Securities, told Reuters it was a “natural fit” for Meta, which extended into boss Mark Zuckerberg’s “vision of personal AI” using agents.

Based in Singapore after relocating from China, Manus has sought to set itself apart from rival AI developers with what it claims can be a “truly autonomous” agent.

Unlike many chatbots which need to be repeatedly asked for things before a user can get their desired response, Manus says its service can plan, execute and complete tasks independently in accordance with instructions.

It forms part of the company’s mission to “extend human reach” with general-purpose agents that can aid, rather than replace, human work.

The company said its acquisition by Meta was “validation” of its efforts.

“Joining Meta allows us to build on a stronger, more sustainable foundation without changing how Manus works or how decisions are made,” said Xiao Hong, its chief executive and one of its Chinese founders, in a blog post.

“We’re excited about what the future holds with Meta and Manus working together and we will continue to iterate the product and serve users that have defined Manus from the beginning.”

Meta said as part of its deal it would continue to operate and sell Manus’ AI service.

It marks yet another high-profile move by the Silicon Valley tech giant to cement its presence in the sector through deals with rising start-ups.

In June the company spent $14bn to buy 49% of Scale AI and secured its boss to take a lead role in Meta’s development of the tech.

This came amid a wider increase in spending by Zuckerberg on the company’s AI strategy, as well as reportedly luring talent from rivals like OpenAI.

SOURCE PAGE

In January this year, Harry Davies and Yuval Abraham first reported that Microsoft had deepened its ties to Israel alongside other major tech firms. Since then, the Guardian has published an award-winning series of investigations – in partnership with the Israeli-Palestinian publication +972 Magazine and the Hebrew-language outlet Local Call – that has revealed a symbiotic relationship between Silicon Valley and the Israeli military.

One investigation exposed an Israeli mass surveillance program scooping up virtually all Palestinian phone calls and storing them on Microsoft’s cloud services – setting off an inquiry that ultimately prompted the company to cut off Israel’s access to some of its technology. Another story revealed that the Israeli military created a ChatGPT-like tool to analyze data collected through the surveillance of Palestinians. Yet another revealed that Google and Amazon had agreed to extraordinary terms to clinch a lucrative contract with Israel.

I asked Davies and Abraham to discuss what they learned this year – about the role of these technologies in Israel’s assault on Gaza, whether these business ties are sustainable, and what the revelations tell us about how the wars of the future will be fought.

How did Israel’s relationships with these companies change after October 7?

Yuval Abraham: The Israeli military had been fetishizing artificial intelligence and big data for many years – a trend that is very much connected to Israel’s occupation of the Palestinians, because the occupation generates a lot of data. What changed after October 7 was the scope. The military was looking to bomb hundreds of targets every day in Gaza. Tens of thousands of people were recruited into reserve duty. That meant a huge spike in usage of technological systems. That’s where the big tech companies stepped in.

Harry Davies: There was a huge surge in demand – not just for the storage capacities of the tech companies, but also for the products that they offer to analyze the information used to prosecute a war. What’s valuable for the military is the way in which these services are able to provide what’s known as “blob storage”, which allows them to store and process infinite amounts of raw intelligence information.

What has made Israel such an appealing market for these companies?

Yuval Abraham: As we reported, the Israeli army has been collecting Palestinian phone calls for a long time. But when you want to collect the phone calls of an entire population every day, and you want to retain those phone calls for long periods of time, you need a lot of storage room and processing power.

If you remember the [Edward] Snowden revelations, many of them had to do with metadata, which doesn’t weigh a lot. But the Israeli military also wanted to store mass audio files, images or videos – and for that it felt it needed the assistance of companies like Microsoft. In the West Bank, sources have told us this information has been used to find dirt on people to blackmail them. In the Gaza Strip, we know that this massive trove of intercepted phone calls was also used in airstrikes that killed civilians.

So data is power and data is control. And these American cloud providers allow the Israeli military to store a lot of data and to sift through it very effectively. That has direct consequences for people on the ground.

Quick Guide

Contact Harry Davies and Yuval Abraham about this story

Show

If you have something to share about this story, you can contact Harry Davies and Yuval Abraham using one of the following methods.

Secure Messaging in the Guardian app

The Guardian app has a tool to send tips about stories. Messages are end to end encrypted and concealed within the routine activity that every Guardian mobile app performs. This prevents an observer from knowing that you are communicating with us at all, let alone what is being said.

If you don’t already have the Guardian app, download it (iOS/Android) and go to the menu. Select ‘Secure Messaging’.

To send a message to Harry and Yuval please choose the ‘UK Investigations’ team.

Signal Messenger

You can message Harry using the Signal Messenger app. Use the ‘find by username’ option and type hfd.90

Email (not secure)

If you don’t need a high level of security or confidentiality you can email harry.davies@theguardian.com

SecureDrop and other secure methods

If you can safely use the tor network without being observed or monitored you can send messages and documents to the Guardian via our SecureDrop platform.

Finally, our guide at theguardian.com/tips lists several ways to contact us securely, and discusses the pros and cons of each. 

Illustration: Guardian Design / Rich Cousins

Thank you for your feedback.

Harry Davies: Yossi Sariel, the former head of [Israel’s elite spy agency] Unit 8200, wrote a book under a pseudonym that we revealed to have been published by him. In that book, he articulated what at the time was a bold and radical vision – as Yuval described, this fetishization of Silicon Valley technology. He recognized the possibilities that the likes of Google, Amazon and Microsoft could afford the Israeli military. Two years before October 7, he said that militaries and governments needed to forge relationships with these companies that are similar to the relationships they have with companies like Boeing and Lockheed Martin. So he was already thinking about these companies as instrumental to war and surveillance in the way that a defense contractor provides components for fighter jets or manufactures bombs and missiles.

Some of our reporting has looked at how elements of his vision have come true and have been put into effect, both before October 7 and afterwards, in both the West Bank and Gaza.

We know AI is central to Israel’s military operations – its army has developed its own AI capabilities, as you revealed, Yuval, and has also purchased AI tools for use from Microsoft. Why is AI so central to Israel’s broader war aims?

Yuval Abraham: What AI did was allow Israel to achieve the effective results of carpet bombing without losing the legitimacy of a data-driven assault with targets and objectives. In Gaza, one way in which the Israeli military used AI was to give a score to almost every person in Gaza who has a phone number, determining how likely it was that person was a member of Hamas or Islamic Jihad. This score was based on a machine-learning algorithm [developed by Israel] called Lavender. It was trained on a dataset of known Hamas members. AI allowed the Israeli military to generate and bomb tens of thousands of military targets, on a scale that without AI would not have been humanly possible. Many of the targets were not Hamas members, according to sources. And Israel for the most part bombed these people not while they were engaged in military activity, but when they stepped inside their families’ homes.

These AI systems had an error rate that the Israeli military knew about. But to me, the key thing about AI is not the mistakes that it makes. It’s the scale of destruction that it allows militaries to unleash, and it’s a discourse of legitimacy that it enables – a discourse of targets and collateral damage.

AI also seems to me to incentivize mass surveillance, right? Because it allows for the analysis of ever-growing reams of information.

Harry Davies: Signals intelligence agencies have long collected more information than they could humanly process. That served as a kind of restraint on their ability to conduct mass surveillance. I think we’re now seeing a shift where AI allows an intelligence agency like Unit 8200 to make sense of things that previously it struggled to make sense of.

Microsoft explicitly credited your reporting for changing its policies. Are you seeing any other signs of shifts within the tech industry?

Harry Davies: I think we’re seeing a lot of discomfort and dissent within these companies at both a junior and to some extent senior level. Many employees have been disturbed to find what the products and services that they’re working around the clock to build and market are actually contributing to. There have been protest groups which have emerged from current and former employees within these companies. That’s true across Silicon Valley. I think that played some role in the decision that Microsoft made as a result of our reporting. They were facing a lot of pressure internally.

Yuval Abraham: And there’s also a legal question for these companies: if the ICJ [international court of justice] ends up ruling that Israel has committed a genocide, then a follow-up question will be: who contributed to that genocide? Which companies helped maintain it and sustain it? For some people in these companies who are thinking ahead, that could also be a cause for concern.

It sounds like you think shifts in public support for Israel could actually affect these business relationships.

Yuval Abraham: Israel has developed a reliance on these companies for its Nimbus project, which is a huge contract signed between Israel and Google and Amazon back in 2021. It is moving the data of many of its government ministries, along with troves of information from the Ministry of Defense, onto these companies’ cloud servers.

These are US companies. They’re taking a certain gamble here that the US will stay loyal to Israel and won’t block, limit or sanction them.

Microsoft only blocked access to technology that was specifically enabling the mass surveillance of Palestinian phone calls – there are still many relationships between Microsoft and the Israeli military. But Microsoft’s action made many people in the Israeli system nervous. It was the first time we know of that a big tech company withdrew services from the Israeli military. It made some people ask whether Israel is making a mistake by giving these foreign companies so much leverage. That question is folded within a larger question of what the US will do, what will happen in 2028 if there’s a more progressive administration in the White House, at a time when so many Americans believe that Israel has committed a genocide in Gaza.

What’s your focus going to be in 2026?

Yuval Abraham: I think we only uncovered the tip of the iceberg.

Harry Davies: We’re both very conscious that, although we have spent a lot of time working on this, we still just have glimpses inside the system. We’re continuing to build a fuller picture of how this technology was and continues to be used in Gaza and in the West Bank as well.

There’s good reason to continue paying attention. Militaries pay attention to what other militaries are doing. There is great interest among other western militaries in how Israel prosecuted this war, in how it integrated these kind of technologies.

And there are other militaries whose combat systems and processes are already deeply integrated with Silicon Valley tech. Take the American military, for example. Look at what’s happening right now in the Caribbean. Are those operations somehow free of the involvement or reliance on systems and services provided by these companies? I suspect not. We don’t know for sure, but the Pentagon and the US military have very big contracts with all of these companies to provide cloud services. Post-Gaza, we have to look at these relationships and ask: what is the involvement of these companies and their technology in military decisions, in military operations and in warfare more broadly?

Yuval Abraham: Much of our reporting is based on whistleblowers, on individuals who are in proximity to power or hold positions of power.

Harry Davies: Our confidential sources have remained confidential and we are always interested in hearing from new people. Our door is always open.

Quick Guide

Contact Harry Davies and Yuval Abraham about this story

Show

If you have something to share about this story, you can contact Harry Davies and Yuval Abraham using one of the following methods.

Secure Messaging in the Guardian app

The Guardian app has a tool to send tips about stories. Messages are end to end encrypted and concealed within the routine activity that every Guardian mobile app performs. This prevents an observer from knowing that you are communicating with us at all, let alone what is being said.

If you don’t already have the Guardian app, download it (iOS/Android) and go to the menu. Select ‘Secure Messaging’.

To send a message to Harry and Yuval please choose the ‘UK Investigations’ team.

Signal Messenger

You can message Harry using the Signal Messenger app. Use the ‘find by username’ option and type hfd.90

Email (not secure)

If you don’t need a high level of security or confidentiality you can email harry.davies@theguardian.com

SecureDrop and other secure methods

If you can safely use the tor network without being observed or monitored you can send messages and documents to the Guardian via our SecureDrop platform.

Finally, our guide at theguardian.com/tips lists several ways to contact us securely, and discusses the pros and cons of each. 

Illustration: Guardian Design / Rich Cousins

Thank you for your feedback.

SOURCE PAGE