TECHNOLOGY
Is AI already conscious? Evidence is ‘far too limited’ to definitively say artificial intelligence hasn’t made the leap, expert claims
Artificial intelligence (AI) is already helping to solve problems in finance, research and medicine.
But could it be reaching consciousness?
Dr Tom McClelland, a philosopher from the University of Cambridge has warned that current evidence is ‘far too limited’ to rule this dystopian possibility out.
According to the expert, the only sensible position on the question of whether AI is conscious is one of ‘agnosticism’.
The main problem, he claims, is that we don’t have a ‘deep explanation’ of what makes something conscious in the first place, so can’t test for it in AI.
‘The best–case scenario is we’re an intellectual revolution away from any kind of viable consciousness test,’ Dr McClelland explained.
‘If neither common sense nor hard–nosed research can give us an answer, the logical position is agnosticism.
‘We cannot, and may never, know.’
Artificial intelligence ( AI) is already helping to solve problems in finance, research and medicine. But could it be reaching consciousness? Pictured: Terminator Genisys
AI companies are investing vast sums of money into pursuing ‘artificial general intelligence’ – the point at which AI can outperform humans in any area.
But as they work towards this goal, some also claim that increasingly sophisticated AI may develop consciousness.
This means AI could develop the capacity for perception and become self–aware.
While this idea might evoke visions of killer robots, Dr McClelland argues that AI could make this jump without us even realising, because we don’t really have an agreed–upon theory of consciousness to begin with.
Some theories say consciousness is a matter of processing information in the right way, and that AI could be conscious if only it could run the ‘software’ of a conscious mind.
Others argue it is inherently biological, meaning AI can only imitate consciousness at best.
Until we can figure out which side of the argument is right, we simply don’t have any basis on which to test for consciousness in AI.
In a paper published in the journal Mind and Language, Dr McClelland claims both sides of the debate are taking a ‘leap of faith’.
We can’t tell whether an AI, like in the sci–fi film Ex Machina (pictured), really has conscious experience or whether it is just simulating consciousness
Whether something is conscious radically changes the kinds of ethical questions we need to consider.
For example, humans are expected to behave morally towards other people and animals, because consciousness gives them ‘moral status’.
In contrast, we don’t have these same values towards inanimate objects, like toasters or computers.
‘It makes no sense to be concerned for a toaster’s well–being because the toaster doesn’t experience anything,’ Dr McClelland explains.
‘So when I yell at my computer, I really don’t need to feel guilty about it. But if we end up with AI that’s conscious, then that could all change.’
While that might make dealing with AI an ethical nightmare, the bigger risk may be that we start to consider AIs as conscious or sentient when they are not.
Dr McClelland explained: ‘If you have an emotional connection with something premised on it being conscious and it’s not, that has the potential to be existentially toxic.’
Worryingly, the philosopher says that members of the public are already sending him letters written by chatbots ‘pleading with me that they’re conscious’.
He added: ‘We don’t want to risk mistreating artificial beings that are conscious, but nor do we want to dedicate our resources to protecting the “rights” of something no more conscious than a toaster.’
HOW SOON WILL WE BE ABLE TO UPLOAD OUR MINDS TO A COMPUTER?
Brain and memory preservation has been explored at length by futurists, scientists and science fiction junkies alike.
Many say it falls under the category of ‘transhumanism.’
Transhumanism is the belief that the human body can evolve beyond its current form with the help of scientists and technology.
The practice of mind uploading has been promoted by many people, including Ray Kurzweil, Google’s director of engineering, who believes we will be able to upload our entire brains to computers by 2045.
Similar technologies have been depicted in science fiction dramas, ranging from Netflix’s Altered Carbon, to the popular series Black Mirror.
Another prominent futurist, Dr Michio Kaku, believes virtual reality can be used to keep our loved ones’ personalities and memories alive even after they die.
Scientists and futurists have different theories about how we might be able to preserve the human brain, ranging from uploading our memories to a computer to Nectome’s high-tech embalming process, which can keep it intact for thousands of years
‘Imagine being able to speak to your loved one after they die … it is possible if their personality has been downloaded onto a computer as an avatar,’ he explained.
These ideas haven’t been met without criticism.
McGill University Neuroscientist Michael Hendricks told MIT that these technologies are a ‘joke.’
‘I hope future people are appalled that in the 21st century, the richest and most comfortable people in history spent their money and resources trying to live forever on the backs of their descendants. I mean, it’s a joke, right? They are cartoon bad guys,’ he said.
Meanwhile, neuroscientist Miguel Nicolelis said recently that such technologies would be virtually impossible.
‘The brain is not computable and no engineering can reproduce it,’ he said.
‘You can have all the computer chips in the world and you won’t create a consciousness.’
Chris VallanceSenior technology reporter
Getty Images
One in three adults in the UK are using artificial intelligence (AI) for emotional support or social interaction, according to research published by a government body.
And one in 25 people turned to the tech for support or conversation every day, the AI Security Institute (AISI) said in its first report.
The report is based on two years of testing the abilities of more than 30 unnamed advanced AIs – covering areas critical to security, including cyber skills, chemistry and biology.
The government said AISI’s work would support its future plans by helping companies fix problems “before their AI systems are widely used”.
A survey by AISI of over 2,000 UK adults found people were primarily using chatbots like ChatGPT for emotional support or social interaction, followed by voice assistants like Amazon’s Alexa.
Researchers also analysed what happened to an online community of more than two million Reddit users dedicated to discussing AI companions, when the tech failed.
The researchers found when the chatbots went down, people reported self-described “symptoms of withdrawal”, such as feeling anxious or depressed – as well as having disrupted sleep or neglecting their responsibilities.
Doubling cyber skills
As well as the emotional impact of AI use, AISI researchers looked at other risks caused by the tech’s accelerating capabilities.
There is considerable concern about AI enabling cyber attacks, but equally it can be used to help secure systems from hackers.
Its ability to spot and exploit security flaws was in some cases “doubling every eight months”, the report suggests.
And AI systems were also beginning to complete expert-level cyber tasks which would typically require over 10 years of experience.
Researchers also found the tech’s impact in science was also growing rapidly.
In 2025, AI models had “long since exceeded human biology experts with PhDs – with performance in chemistry quickly catching up”.
‘Humans losing control’
From novels such as Isaac Asimov’s I, Robot to modern video games like Horizon: Zero Dawn, sci-fi has long imagined what would happen if AI broke free of human control.
Now, according to the report, the “worst-case scenario” of humans losing control of advanced AI systems is “taken seriously by many experts”.
AI models are increasingly exhibiting some of the capabilities required to self-replicate across the internet, controlled lab tests suggested.
AISI examined whether models could carry out simple versions of tasks needed in the early stages of self-replication – such as “passing know-your customer checks required to access financial services” in order to successfully purchase the computing on which their copies would run.
But the research found to be able to do this in the real world, AI systems would need to complete several such actions in sequence “while remaining undetected”, something its research suggests they currently lack the capacity to do.
Institute experts also looked at the possibility of models “sandbagging” – or strategically hiding their true capabilities from testers.
They found tests showed it was possible, but there was no evidence of this type of subterfuge taking place.
In May, AI firm Anthropic released a controversial report which described how an AI model was capable of seemingly blackmail-like behaviour if it thought its “self-preservation” was threatened.
The threat from rogue AI is, however, a source of profound disagreement among leading researchers – many of whom feel it is exaggerated.
‘Universal jailbreaks’
To mitigate the risk of their systems being used for nefarious purposes, companies deploy numerous safeguards.
But researchers were able to find “universal jailbreaks” – or workarounds – for all the models studied which would allow them to dodge these protections.
However, for some models, the time it took for experts to persuade systems to circumvent safeguards had increased forty-fold in just six months.
The report also found an increase in the use of tools which allowed AI agents to perform “high-stakes tasks” in critical sectors such as finance.
But researchers did not consider AI’s potential to cause unemployment in the short-term by displacing human workers.
The institute also did not examine the environmental impact of the computing resources required by advanced models, arguing that its task was to focus on “societal impacts” that are closely linked to AI’s abilities rather than more “diffuse” economic or environmental effects.
Some argue both are imminent and serious societal threats posed by the tech.
And hours before the AISI report was published, a peer-reviewed study suggested the environmental impact could be greater than previously thought, and argued for more detailed data to be released by big tech.
TECHNOLOGY
‘Uniquely evil’: Michigan residents fight against huge data center backed by top tycoons | Michigan
A who’s who of the nation’s most powerful politicians and tech tycoons are forcing through a proposal for a massive data center in rural Michigan as locals from across the political spectrum have come out in force against it, with one calling it “uniquely evil”.
Saline Township, Michigan, residents fear the $7bn center would jack up energy bills, pollute groundwater, and destroy the area’s rural character. The 1.4 gigawatt center would consume as much power as Detroit, and would help derail Michigan’s nation-leading transition to renewable energy.
Responding to resident pressure, Saline Township’s board of trustees in September voted down the plans, but the data center’s powerful backers – including Donald Trump, Open AI’s Sam Altman, Oracle’s Larry Ellison, Michigan governor Gretchen Whitmer, utility giant DTE Energy, and Stephen Ross, the real-estate billionaire and Trump donor who owns Related Co – fought back.
A fallen sign on the curb near the site of the soon-to-be-built data center.
Related Digital sued, and, vastly outgunned, the township board quickly folded and reversed its decision over strong resident objections. Now the project’s backers are trying to avoid minimal regulatory scrutiny on energy costs and pollution.
The controversy over the data center is representative of the David v Goliath fights playing out across the US, pitting working- and middle-class residents against the interests of billionaires and the political establishment.
“This is part of an experience that America and the world is having around tech billionaires who are seizing power and widening the gap between those have much too much … and the working and middle classes,” said Yousef Rabhi, a former Democratic state legislative leader and clean energy advocate who opposes the plans.
“That’s what these data centers are symbolic of, and they’re the vehicle for is the furtherance of this divide,” Rabhi added.
Yousef Rabhi speaks at a rally held in Saline Township in opposition to the data center. Photograph: Courtesy of Yousef Rabhi
The proposal is part of the broader “Stargate” project composed of five data centers backed by the Trump administration, which granted $500bn in federal subsidies for them. It’s the largest project in Michigan history in terms of investment, and it also received subsidies on taxes that could have gone to roads and schools, among other uses, Rabhi said.
The plan’s supporters say the center would provide essential AI infrastructure, in part for national security, and create a few hundred jobs. Huge sums of money are at stake for the tech and utility companies.
Ross’s Related Digital is the data center’s developer, while OpenAI, which produced ChatGPT, and Oracle will use the center to house its AI infrastructure.
In a statement, Related Digital alleged company the township’s decision violated zoning laws, and the spokesperson stressed the suit was filed jointly with three Saline Township property owners who are selling their property to Related.
“Thankfully, we were able to reach a settlement agreement with the township to allow this project to move forward,” the spokesperson said. They noted Related is also making about $14m in donations to local causes.
Saline Township is a small community of about 4,000 just outside Ann Arbor. The Stargate project is one of around a dozen data center proposals in Michigan over the last year that are strongly opposed at the local level. It’s one of four proposed near Ann Arbor – last week, plans for a second a few miles away in a neighboring town surfaced.
Some municipalities have been successful in derailing plans, while others have lost the fight.
A new data center being built along US Route 12 in Saline Township. Photograph: Sarah Rice/The Guardian
In Saline Township, former US marine Kate Haushalter and her husband are raising five children in a farmhouse next to the data center site. They bought and renovated the once-dilapidated home so they could live in a bucolic area, and Haushalter said she was not about to cede ground even though the township did.
“Maybe because I was in the Marine Corps, but I would rather stay and fight,” Haushalter said. “I’m sure the chances are slim, but it’s worth fighting for, and I don’t want to teach my kids to roll over.”
‘We were dealt the cards we were dealt’
Big tech companies such as Google, Microsoft and Open AI, which often own data centers, typically have enough political support at the state and federal levels that inexperienced local leaders who are comparatively poorly resourced are left on their own to defend their town from the centers.
Saline Township supervisor Jim Marion conveyed that challenge when he told angry residents during a contentious November discussion that the township’s “hands were tied”.
“This township doesn’t have the money to fight these big companies. You got to understand that,” Marion told the crowd. “We were dealt the cards we were dealt.”
A sign warns of ‘construction ahead’ on US Route 12. Photograph: Sarah Rice/The Guardian
Some municipalities have utilized zoning laws to block the centers. Beyond that, there’s little local officials can do, and state and federal level regulations on the centers are virtually non-existent.
Still, residents are growing more organized. A first protest on 1 December drew about 200 people, who Rabhi described as “truly a cross-section of American society”. The next week, 800 people participated in a state-level public input session, and organizers are pressuring state environmental regulators to hold up the project’s required wetland permits.
Among residents leading the pushback is Josh LeBaron, whose home sits about 500 yards from the site, where crews have broken down.
He characterized the project as “uniquely evil” because of the environmental risks, and because, he and others allege, the companies and government have been secretive about their plans. In response to questions about accusations of nimby-ism leveled against local residents by the project’s supporters, LeBaron said he would not be opposed to other developments.
A sign protesting against the new data center in Saline Township. Photograph: Sarah Rice/The Guardian
He noted that Michigan is full of former industrial sites that would be more appropriate for the 575-acre property.
“I would be at home reading a book if it were a subdivision,” LeBaron said.
A Related spokesperson told the Guardian the company “explored sites across Michigan before deciding on this site, which is ideal as it’s a contiguous flat area”, and is set close to a major road and transmission lines.
Higher bills and the end of Michigan’s climate laws
Local opponents’ best hope for holding up the project lies in the arcane utility regulatory process on the massive amount of power the data center would require.
DTE Energy claims the data center’s power demands and need for expensive new infrastructure will not increase residential electricity prices.
But it doesn’t want to show its math.
DTE filed a petition with the Michigan Public Services Commission (MPSC), the state agency that regulates utilities, asking the MPSC to fast-track the plan’s approval. DTE’s request for an “ex parte” case requires limited scrutiny of its claim that the center won’t destroy the climate laws, or increase electricity bills.
In response, Michigan attorney general Dana Nessel and consumer advocacy groups filed a legal petition with the MPSC, calling for a “contested case” that would require much closer regulatory review of DTE’s claims.
The MPSC is helmed by Whitmer appointees, and the governor has strongly backed the project, raising suspicions among opponents that the agency will approve the ex parte request.
Studies from across the country have shown data centers often increase rates, and DTE and regulators “cannot claim transparency while shutting the public out of the only process that requires DTE to support its claims with actual evidence”, said Bryan Smigielski, Michigan campaign organizer for the Sierra Club, which is intervening in the regulatory battle.
In a statement, a DTE spokesperson said: “To be clear, these data center customer contracts will NOT create a cost increase for our existing customers.”
Saline Township is a small community of about 4,000 just outside Ann Arbor. Photograph: Sarah Rice/The Guardian
DTE has said the project won’t derail Michigan’s transition to clean energy, but state data and DTE’s plans suggest otherwise.
Michigan, in late 2023, passed nation-leading climate laws that require utilities to transition to renewable energy by 2040. But the law included an “off-ramp” that allows utilities to continue running or building fossil fuel plants if renewable sources cannot handle the energy grid’s load.
At its peak, DTE’s grid already demands about 9.5 gigawatts of power, while the grid’s capacity is 11gw.
In July, DTE told investors it is in negotiations with big tech companies to provide 7gw of power for several proposed data centers.
The Saline center’s 1.4gw may not cause an exceedance of the 11gw threshold, especially because DTE is planning to build battery storage. But the Saline center along with any of the other proposed centers likely would trigger the off-ramp.
DTE appears to be planning for that likelihood: DTE Energy executives said the company would likely need to build new gas plants to accommodate the data centers’ demand.
Saline Township ‘will never be the same’
Haushalter’s kids were born in the renovated farmhouse and are homeschooled there.
She and her husband try to teach the kids to respect nature. The family manages beehives, watches the geese and plant trees for wood to use in their wood-burning stove. At night, they take the kids outdoors for bonfires to look at the stars. “We’re not a big screen family,” Haushlater said.
Kate Haushalter photographed in her family’s home. Photograph: Sarah Rice/The Guardian
The noise, light and air pollution is already disrupting the life the family built over 13 years. The center, if it is fully built, would fully destroy it, Haushalter said.
“We are really passionate about nature and teaching our kids about it and I can’t believe the biggest construction project in Michigan is landing literally in my backyard, and there’s no recourse for the little guy,” she said. “It’s going to crush us.”
TECHNOLOGY
My brutally honest toddler tested seven of Christmas 2025’s hottest toys and she knows exactly what parents should buy
As retailers hype what they claim are the hottest Christmas toys of 2025, I turned to a far more honest expert: my three-year-old daughter.
My daughter, Amelia, tested each toy the way only a toddler can, without instructions, without expectations and with zero interest in whether it was supposed to be popular.
I, however, scoured lists of must-have toys for this holiday season, meticulously reading reviews and watching product videos to see which ones were expected to top wish lists.
While the toys had to be fun and colorful, they also had to spark imagination.
Studies show that simple, open-ended, imagination-driven toys like blocks, dolls, and pretend-play sets boost children’s brains by improving memory, self-control, spatial reasoning, problem-solving, and language skills.
They also enhance social-emotional growth, support neural development, activate areas linked to empathy more than screen time, and lay the foundation for future learning.
During the review, Amelia kept asking the same question: ‘You want to play with that?’
By the end, it became clear that the toys were not just the trendiest ones, but were the ones that invited her to imagine, invent and play alongside me
CONNETIX magnetic tiles lets the imagination run wild, coming with 102 pieces. They easily attach, are sturdy, light-weight and beautifully colored
CONNETIX Rainbow Creative Pack 102 pc
Price: $169
I had seen other children playing with CONNETIX magnetic tiles and always thought they would be perfect for Amelia.
The moment we opened the box, her imagination ran wild. She built a dollhouse one minute and a castle the next.
With squares, triangles, and a mix of bright colors, the possibilities for creations seem endless.
While building, Amelia practiced fitting pieces together and imagining the layout of her castle, which really stretched her problem-solving and planning skills.
I joined her for a while, and it was nearly impossible to get her to stop for bed because she wanted to keep playing.
The next morning, she woke up early, grabbed the box and picked up exactly where she left off.
The tiles are lightweight yet sturdy, snap together easily, and are perfect for travel.
Since Amelia is not allowed screens yet, these tiles will be a lifesaver for keeping her entertained when we are out to dinner or even on an airplane.
While building, Amelia practiced fitting pieces together and imagining the layout of her castle, which really stretched her problem-solving and planning skills
The Bluey Supermarket Playset comes with over 15 pieces, including a grocery cart, pretend food and other fun items
Bluey Supermarket Playset
Cost: Up to $49.99
Amelia, like most children her age, loves the cartoon Bluey.
On my hunt for the hottest toys, the Bluey Supermarket Playset seemed to appear on every list.
When we opened the box, Amelia’s eyes lit up, and she immediately began exploring the multi-level supermarket, loading tiny carts, riding the escalator and announcing her own grocery adventures over the built-in intercom.
The expansive playset includes more than 15 play pieces, along with exclusive Bluey and Bingo figures, and is packed with 24 sounds and phrases.
The interactive sound effects, detailed shelves and exclusive Bluey and Bingo figures kept her engaged for hours, sparking creativity and encouraging imaginative storytelling.
When we opened the box, Amelia’s eyes lit up, and she immediately began exploring the multi-level supermarket, loading tiny carts, riding the escalator and announcing her own grocery adventures over the built-in intercom
She was not just pretending to shop, but she was recreating scenes from the show, inventing her own adventures, and even practicing problem-solving as she navigated the store’s different levels.
Amelia also pretended that Bingo and Bluey were shopping at the Hammerbarn.
I appreciated the thoughtful design, which supports cooperative play, hand-eye coordination and imaginative thinking.
Micro Maxi Foldable LED Scooter
Cost: $169.99
I was looking for a scooter for Amelia and stumbled upon M-cro. The scooter has adjustable handles, allowing Amelia to ride it until she’s at least 12 years old
We live in a city, and it feels like every kid has a scooter, so I knew this would be a hot toy for Christmas.
Amelia will be four in a few months, so I initially thought the Micro option would suit her perfectly.
However, Kris, who works for the company, suggested we go a level up.
The Micro Maxi can grow with her and be used all the way up to around 12 years old.
The adjustable handlebar means she will not outgrow the scooter anytime soon, and the three-wheel design gives her extra stability as she learns to balance.
Other scooters were cheaper, but the Micro Maxi immediately felt sturdier and higher quality than the budget options.
To test it, Amelia scooted around my office, easily gliding through the halls and kicking off with confidence.
It’s solid, lightweight, and folds up easily, making it convenient to carry and store.
And even more, the wheels light up.
Little Live Pets Ouchies Retriever
Cost: $59.00
The sweet playful walking pup has developed a limp. Inside the plush pup’s carrier is a ‘magic cast’ for kids to apply to help Cooper heal back to his happy self – they can even personalize the cast with their own designs.
Although the dog’s official name is Cooper, Amelia quickly renamed her Lucy
It took us a few minutes to figure out how the Little Live Pets Ouchies Retriever works, so I’d recommend reading the instructions first. Once we did, Amelia was instantly hooked.
Although the dog’s official name is Cooper, Amelia quickly renamed her Lucy.
The interactive toy encourages nurturing play, giving children the chance to care for their very own ‘pet.’
Lucy arrived with an injured paw, and Amelia eagerly got to work wrapping it in bandages, gently petting her while the cast dried.
The moment the cast came off, Lucy barked happily and began strutting around the room.
As Amelia watched Lucy parade about, it was clear how proud she felt of helping her new friend heal.
While Lucy is a robotic toy, she’s covered in a surprisingly soft coat and is about the size of a small puppy, making her easy for little hands to hold and cuddle.
The interactive pooch sounds and reacts like a real puppy, barking when she’s happy, whimpering when she’s sad.
At six inches tall for Bluey and just under that for Bingo, the figures are the perfect size for little hands
Each figure talks when its nose is pushed, delivering 10 fan-favorite phrases
Bluey and Bingo Chat Mates
Cost: $15.00 each
The Bluey and Bingo talking figures were an instant hit.
With just a push of the nose, each character comes to life, repeating familiar phrases from the much-loved animated series, and Amelia immediately recognized them.
Hearing Bluey say lines like ‘I’ve got an idea’ and ‘Beep twice if we can be friends’ made the figures feel just like the show, turning playtime into a mini episode of Bluey at home.
Amelia especially loved pressing the noses over and over to hear what they would say next.
At six inches tall for Bluey and just under that for Bingo, the figures are the perfect size for little hands.
They’re sturdy, easy to use, and simple enough for toddlers to play with independently, making them a fun and engaging choice for young Bluey fans this Christmas.
GUI GUI
Cost: Starting at $9.99
Amelia, like most kids, is a huge fan of making slime
The multi-step ‘get ready’ routine starts with unboxing to reveal the slime, then ‘get glowing’ by mixing in the booster, decorating with the included charms and beads
While GUI GUI is primarily marketed toward tweens and teens, Amelia is a huge fan of slime, so we couldn’t resist putting this kit to the test.
When we first opened it, the slime was extremely sticky, but after mixing in the included booster, it quickly transformed into a soft, dough-like consistency that was much easier for little hands to manage.
Amelia loved watching the texture change, which made the process feel like part of the fun rather than a setup step.
The kit follows a multi-step ‘get ready’ routine that starts with unboxing the slime, then ‘getting glowing’ by mixing in the booster.
From there, kids can decorate the slime with the included charms and beads, reveal the adorable mini figure hidden inside, and finish things off with a final sparkle using the glitter mix.
Although designed for older kids, the hands-on, sensory play kept Amelia fully engaged, making it a surprisingly fun option for younger slime lovers with adult supervision.
Tin Can
Cost: $75
It’s a thoughtful option for families looking to introduce communication without screens, distractions or scrolling
Amelia loves talking on the phone with her grandparents, so I thought the Tin Can would be a fun and age-appropriate way for her to stay in touch with family.
While she’s still a bit young to fully appreciate everything it offers, I see Tin Can as a clever bridge between toy and technology, one that delays the need for a smartphone while still allowing real communication.
The WiFi-based, screen-free device is designed specifically for kids and comes with a parent app that allows for easy contact management, quiet hours, and built-in 911 support.
Parents can also choose between plans that allow calling only other Tin Can devices or standard phone numbers.
Beyond the functionality, there’s something delightfully nostalgic about it.
Using Tin Can brought back memories of my own childhood, answering calls on a landline and feeling very grown-up in the process.
It’s a thoughtful option for families looking to introduce communication without screens, distractions, or scrolling.
As 2025 draws to a close, the wave of layoffs that shook the African tech ecosystem in 2023 and 2024 appears to be easing. They’re still happening, but less frantically than in the past two years.
African founders are learning to adjust to the market realities, including reduced funding and high costs. Startups are also learning that the teams they built during the boom years are too expensive to maintain in a slower market.
Our coverage of layoffs this year showed that most founders are treating job cuts as a way to stay alive rather than a sign that a business is failing. Venture capital is hard to secure, investors are more cautious, and there’s far less patience for growth without revenue.
Reducing headcount is helping companies stretch their cash, focus on what actually makes money, and buy time in an uncertain environment.
This shift is changing the mood of the ecosystem. Hiring is slowing, workers are thinking twice about jumping ship, and the idea of tech as a fast and easy route to opportunity is fading. Yet startups are growing more careful, expansion is more deliberate, and spending is under closer watch. As attention turns to 2026, the emphasis is on building businesses that can survive tough years, not just thrive in good ones.
Here are some of the startups and tech firms that cut jobs in 2025 amid tighter capital, shifting strategies, and a stronger push toward profitability.
Metro Africa Xpress (MAX), Nigeria
MAX, a Nigerian mobility financing startup, began the year by laying off about 150 employees, roughly 30% of its workforce. The company said the restructuring was necessary to support a revised business model and cut operating costs as it pivoted to focus exclusively on financing electric vehicles. The terminations took effect immediately, with no monetary severance offered. While MAX has not previously carried out mass layoffs, it has undergone several major pivots since its founding in 2015, shifting from deliveries to ride-hailing and now to vehicle financing.
54 Collective, South Africa
Venture firm 54 Collective announced layoffs in February following the end of its partnership with the Mastercard Foundation and the shutdown of its Africa-focused venture studio. The firm said it was unable to secure alternative funding to sustain its operations.
Vendease, Nigeria
Vendease, a Nigerian YC-backed food procurement startup, laid off 120 employees, cutting its workforce by 44% in February 2025. This marked the company’s second major round of layoffs in five months, following the dismissal of 68 employees in September 2024. The company said the cuts were necessary to extend its runway and push toward profitability amid naira depreciation and persistent inflation.
Bento, Nigeria
In February, Bento dismissed its entire 10-person technology team in an abrupt and contentious episode. The layoffs followed protests over delayed salaries and were compounded by leadership instability after the resignation of the company’s CEO. The incident unfolded against the backdrop of allegations of tax and pension fraud, with the founder deactivating staff access shortly after the protest.
eBee Africa, Kenya
Although announced in August, eBee issued redundancy notices as early as February, affecting most of its roughly 50 employees. The Kenyan mobility startup, which had set out to put one million electric bicycles on African roads by 2030, was left with a skeletal team before the remaining staff exited voluntarily. The layoffs were driven by declining revenues, rising costs, and slower-than-expected adoption of electric bikes in Kenya, where cheaper alternatives continued to dominate. The company was already under pressure following leadership changes and tax disputes.
Get The Best African Tech Newsletters In Your Inbox
Meta, Africa
Meta’s Africa operations were affected in February after the company laid off an undisclosed number of staff as part of a global performance-based restructuring. The cuts formed part of a wider round that affected about 3,600 employees worldwide and were linked to Meta’s push for efficiency and increased investment in artificial intelligence and core products.
Tala, Kenya
Digital lender Tala laid off 28 employees in Kenya in April as part of a recalibration of its regional operations. The company cited the need to streamline teams, align costs with lending performance, respond to changing customer repayment behaviour, and strengthen risk management in a tighter credit environment. Tala had initially planned to cut 55 roles, but later revised the number downward.
Twiga Foods, Kenya
Twiga Foods laid off more than 300 employees in May as part of a major restructuring that included the creation of a new holding company. The cuts were aimed at streamlining operations and improving efficiency following the acquisition of three Kenyan FMCG distributors. Twiga had previously laid off 59 employees in August 2024 and has carried out several rounds of cuts in recent years.
Sabi, Nigeria
Sabi, a Nigerian B2B e-commerce startup, laid off about 50 employees, representing roughly 20% of its workforce, in June. The company said the cuts were tied to a strategic pivot away from general merchant services toward a minerals traceability and commodities-focused business. The move followed earlier operational scale-backs as Sabi adjusted to changing market conditions.
Flutterwave, Kenya and South Africa
Flutterwave cut about half of its workforce in Kenya and South Africa in mid-2025, with the layoffs reported in July. The payments company cited cost optimisation and a renewed focus on profitability as it positions itself for a potential IPO. The cuts followed an earlier round in 2024, when about 3% of staff were let go.
Businessfront, Nigeria
Businessfront, the publisher of Techpoint Africa, Finance in Africa, Energy in Africa, and Intelpoint, laid off a small but undisclosed number of staff in October. The company said the move was aimed at ensuring long-term sustainability and sharpening strategic focus. The layoffs reflect broader pressure across African tech media, where declining advertising revenues and shifting audience habits have forced similar cuts at peers.
Jumia, Nigeria
E-commerce company Jumia cut about 7% of its workforce in November, reducing headcount to roughly 2,010 employees. The company said the layoffs were part of ongoing efficiency measures, including greater use of AI in customer service and marketing, as it prioritised profitability over expansion.
Photo by Julian O’hayon / Unsplash
You’ve probably put off buying that new phone or laptop. You compare prices, read reviews, and wait for the moment when the numbers finally make sense. If you missed out on Black Friday or Cyber Monday, Croma’s year-end sale opens another window to upgrade without paying full price.
Running from December 15, 2025, to January 4, 2026, in India, the Tata-owned retailer is rolling out discounts across smartphones, laptops, TVs, and home appliances. Bank offers, cashback deals, and easy EMI options further soften the cost, making higher-end gadgets easier to justify. For many shoppers, it’s one of the last big chances of the year to upgrade tech or pick up early holiday gifts.
Top Black Friday 2025 Tech Discounts: The Best Phones, Laptops and Tablets to Buy
Looking for the best Black Friday 2025 tech deals? Here are the biggest price drops on popular phones, MacBooks, Windows laptops, and more.
Smartphone Deals on Croma
One of the biggest highlights of the sale is the iPhone 16, available at Rs 66,990, with EMI options starting from Rs 3,153 per month. The iPhone 15 is even more affordable at Rs 67,490. Android users aren’t left out either; the Samsung Galaxy S25 Ultra comes in at Rs 129,999 with exchange and cashback benefits, while the Galaxy Z Fold 7 is available for an effective price of Rs 216,999. For those looking for something in the mid-range segment, the Vivo V60 starts at Rs 38,999, making it easy to grab a solid upgrade without breaking the bank.
Laptop Deals on Croma
For anyone thinking about upgrading their laptop, Croma’s year-end sale has plenty to offer. The MacBook Air with the M4 chip starts at around Rs 95,990, with student discounts and EMIs from Rs 4,519. Next-generation AI PCs are available from Rs 45,999, while gaming laptops with RTX 3050 graphics begin at Rs 59,990. For everyday users, a reliable Lenovo Core i5 laptop is priced from Rs 34,990.
Prices may vary depending on bank offers, cashback deals, and availability, but the range means there’s something for everyone, whether you need a machine for work, gaming, or general use.
Electronics and Home Appliances Deals on Croma
Photo by Kam Idris / Unsplash
If you’ve been thinking about upgrading your home entertainment, the sale has you covered. Large-screen TVs see some of the biggest discounts: a 75-inch Samsung Smart UHD TV starts at Rs 230,399, while Croma’s own-brand options make things more affordable—a 55-inch Croma UHD Google TV is available for Rs 27,990, and a 43-inch model starts at Rs 16,494.
Beyond TVs, the sale covers a wide range of home appliances, giving shoppers a chance to grab essential gadgets or appliances for the kitchen and living space. As always, prices can vary based on bank offers, cashback deals, and availability, so it’s worth checking details before making a decision.
Craziest Apple Cyber Monday Deals You Don’t Want to Miss
Here’s some of the best deals across Apple devices and ggadgets for Black Friday 2025.
Conclusion
With discounts across so many categories and the option to pay in easy EMIs, the Cromtastic December Sale makes it simple to finally get the gadgets or appliances you’ve been thinking about. It’s the kind of sale where you can pick up what you need for yourself or your home without overthinking it, just in time for the new year.
December 18, 2025
Link copied!
Copy failed!
TECHNOLOGY
2025’s AI boom caused huge CO2 emissions and use of water, research finds | Artificial intelligence (AI)
The AI boom has caused as much carbon dioxide to be released into the atmosphere in 2025 as emitted by the whole of New York City, it has been claimed.
The global environmental impact of the rapidly spreading technology has been estimated in research published on Wednesday which also found that AI-related water use now exceeds the entirety of global bottled-water demand.
The figures have been compiled by the Dutch academic Alex de Vries-Gao, the founder of Digiconomist, a company that researches the unintended consequences of digital trends. He claimed they are the first attempt to measure the specific effect of artificial intelligence rather than datacentres in general as the use of chatbots such as OpenAI’s ChatGPT and Google’s Gemini soared in 2025.
The figures show the estimated greenhouse gas emissions from AI use are also now equivalent to more than 8% of global aviation emissions. His study used technology companies’ own reporting and he called for stricter requirements for them to be more transparent about their climate impact.
“The environmental cost of this is pretty huge in absolute terms,” he said. “At the moment society is paying for these costs, not the tech companies. The question is: is that fair? If they are reaping the benefits of this technology, why should they not be paying some of the costs?”
De Vries-Gao found that the 2025 carbon footprint of AI systems could be as high as 80m tonnes, while the water used could reach 765bn litres. He said it was the first time AI’s water impact had been estimated and showed that AI water use alone was more than a third higher than previous estimates of all datacentre water use.
The figures are published in the academic journal Patterns. The International Energy Agency (IEA) said earlier this year that AI-focused datacentres draw as much electricity as power-thirsty aluminium smelters and datacentre electricity consumption is expected to more than double by 2030.
“This is yet more evidence that the public is footing the environmental bill for some of the richest companies on Earth,” said Donald Campbell, the director of advocacy at Foxglove, a UK non-profit that campaigns for fairness in tech. “Worse, it is likely just the tip of the iceberg. The datacentre construction frenzy, driven by generative AI, is only getting started.
“Just one of these new ‘hyperscale’ facilities can generate climate emissions equivalent to several international airports. And in the UK alone, there are an estimated 100-200 of them in the planning system,” said Campbell.
The IEA has reported that the largest AI-focused datacentres being built today will each consume as much electricity as 2m households with the US accounting for the largest share of datacentre electricity consumption (45%) followed by China (25%) and Europe (15%).
The largest datacentre being planned in the UK, at a former coal power station site in Blyth, Northumberland, is expected to emit more than 180,000 tonnes of CO2 a year when at full operation – the equivalent to the amount produced by more than 24,000 homes.
In India, where $30bn (£22.5bn) is being invested in datacentres, there are growing concerns that a lack of reliability from the National Grid will mean the construction of huge diesel generator farms for backup power, which the consultancy KPMG this week called “a massive … carbon liability”.
Technology companies’ environmental disclosures are often insufficient to assess even the total datacentre impact, never mind isolating AI use, said De Vries-Gao. He noted that when Google recently reported on the impact of its Gemini AI, it did not account for the water used in generating the electricity needed to power it.
Google reported that in 2024 it managed to reduce energy emissions from its datacentres by 12% due to new clean energy sources, but it said this summer that achieving its climate goals was “now more complex and challenging across every level – from local to global” and “a key challenge is the slower-than-needed deployment of carbon-free energy technologies at scale”.
Google was approached for comment.
TECHNOLOGY
The common home appliances that emit TRILLIONS of harmful particles – with the toaster at the top of the list
You likely use them every single day – but some of your home appliances could be emitting harmful pollutants, a new study has warned.
Researchers have discovered popular devices release trillions of ultrafine particles that contain heavy metals.
These can penetrate the human body to settle in the lungs, and have been linked to conditions including asthma, heart disease, high blood pressure, diabetes, cancer and dementia.
Worryingly, the experts warn that young children are especially at risk.
‘Indoor air quality has a critical impact on human health due to the presence of indoor air pollutants,’ the team, from Pusan National University in South Korea said.
‘Small home appliances equipped with heating coils and brushed motors generate ultrafine particles that directly attack users.’
Their tests found that pop–up toasters are the worst offenders, emitting up to 1.73 trillion UFPs per minute.
These particles are incredibly small – less than 100 nanometres in diameter – and were found to contain traces of copper, iron, aluminium, silver and titanium.
The researchers found that overall, children were more at risk from the harmful particles – likely due to their smaller airways
Their tests found that pop–up toasters are the worst offenders, emitting up to 1.73 trillion UFPs per minute (file image)
The assessment involved three different types of small electric home appliances – air fryers, toasters and hairdryers.
The amount of UFPs emitted by each device was measured, as well as the chemical compositions of heavy metals within the particles were analysed.
A simulation model was used to understand how the tiny particles affect our respiratory tracts, to work out who is most susceptible.
Overall, the team found that most of the appliances emit large quantities of UFPs – and the amount of emission was often based on the operating temperatures.
While the pop–up toaster topped the list for harmful pollution emissions, the air fryer came next. When used at a setting of 200°C, it released 135 billion UFPs per minute.
Hairdryers were not as bad, although some models still emitted 100 billion UFPs per minute, the researchers found.
The heavy metals detected in the airborne particles were likely coming off the coils and motors directly, they said.
‘These associated heavy metals increase the risk of cytotoxicity and inflammation when the particles enter the human body,’ the researchers said.
The team found popular home devices release trillions of ultrafine particles that can penetrate the body and settle in the lungs
The heavy metals detected in the airborne particles were likely coming off the device’s coils and motors directly, the team said (file image)
What are ultrafine particles?
UFPs, are extremely tiny airborne particles that are smaller than 100 nanometres in diameter.
They pose significant risks as they can penetrate deep into the lungs and enter the bloodstream.
Road traffic is a major contributor, but they can also be released from household appliances.
Previous studies have linked UFPs to asthma, heart disease, high blood pressure, diabetes, cancer and dementia.
The study also showed that the UFPs are deposited predominantly in the alveolar region of the lungs – the tiny air tubes where the crucial work of breathing happens.
Since children have smaller airways, they face a higher health risk than adults, they warned.
While the study did not directly analyse the potential health impacts of the UFPs produced by these appliances, previous research indicates they are bad for human health.
These particles have been linked to conditions including asthma, heart disease, high blood pressure, diabetes, cancer and dementia.
‘Our study emphasises the need for emission–aware electric appliance design and age–specific indoor air quality guidelines,’ lead author Professor Changhyuk Kim said.
‘In the long term, reducing UFP emissions from everyday devices will contribute to healthier indoor environments and lower chronic exposure risks, particularly for young children.
‘Moreover, this framework can be extended to other consumer products, guiding future innovations toward human health protection.’
The study was published in the Journal of Hazardous Materials.
‘Understanding the source of the pollutants helps to develop preventive measures and policies for keeping healthy indoor air quality,’ Professor Kim concluded.
A study published earlier this year revealed common household products are creating plumes of indoor air pollution.
In their study, the team from Purdue University warned of the dangers of products including air fresheners, wax melts, floor cleaners, and deodorants.
‘A forest is a pristine environment,’ said Nusrat Jung, an assistant professor in Purdue’s Lyles School of Civil and Construction Engineering.
‘But if you’re using cleaning and aromatherapy products full of chemically manufactured scents to recreate a forest in your home, you’re actually creating a tremendous amount of indoor air pollution that you shouldn’t be breathing in.’
WHAT HAVE RECENT STUDIES SHOWN POLLUTION CAN DO TO OUR HEALTH AND BODIES?
CAUSE CHILDREN TO HAVE A LOW IQ: Researchers at the University of California, San Francisco, found in May 2019 that children born to mothers who live in polluted areas have an IQ that is up to seven points lower than those living in places with cleaner air.
CAUSE CHILDREN TO HAVE POORER MEMORY: Researchers at the Barcelona Institute for Global Health found boys exposed to greater levels of PM2.5 in the womb performed worse on memory tests by the time they are 10.
DELAY THE DEVELOPMENT OF CHILDREN: Youngsters who live less than one-third of a mile away from busy roads are twice as likely to score lower on tests of communication skills in infancy, found researchers at Eunice Kennedy Shriver National Institute of Child Health in April. They were also more likely to have poorer hand-eye coordination.
MAKE CHILDREN MORE ANXIOUS: University of Cincinnati scientists claimed pollution may alter the structure of children’s brains to make them more anxious. Their study of 14 youngsters found rates of anxiety was higher among those exposed to greater levels of pollution.
CUT YOUR CHILD’S LIFE SHORT: Children born today will lose nearly two years of their lives because of air pollution, according to a report by the US-based Health Effects Institute and the University of British Columbia in April 2019. UNICEF called for action on the back of the study.
RAISE A CHILD’S RISK OF AUTISM: Researchers at Monash University in Australia discovered youngsters living in highly polluted parts of Shanghai have a 86 per cent greater chance of developing ASD. Lead author Dr Yuming Guo said: ‘The developing brains of young children are more vulnerable to toxic exposures in the environment.’
CAUSE ASTHMA IN CHILDREN: Four million children around the world develop asthma each year because of road traffic pollution, a major study by academics at George Washington University estimated. Experts are divided as to what causes asthma – but exposure to pollution in childhood increases the risk by damaging the lungs.
MAKE CHILDREN FAT: University of Southern California experts found last November that 10 year olds who lived in polluted areas when they were babies are, on average, 2.2lbs (1kg), heavier than those who grew up around cleaner air. Nitrogen dioxide pollution could disrupt how well children burn fat, the scientists said.
LEAVE WOMEN INFERTILE EARLIER: Scientists at the University of Modena, Italy, claimed in May 2019 that they believe pollution speeds up ageing in women, just like smoking, meaning they run out of eggs faster. This was based on them finding almost two-thirds of women who have a low ‘reserve’ of eggs regularly inhaled toxic air.
RAISE THE RISK OF A MISCARRIAGE: University of Utah scientists found in January that pregnant women are 16 per cent more likely to suffer the heartbreak of a miscarriage if they live in areas of high pollution.
RAISE THE RISK OF BREAST CANCER: Scientists at the University of Stirling found six women working at the same bridge next to a busy road in the US got breast cancer within three years of each other. There was a one in 10,000 chance the cases were a coincidence, the study said. It suggested chemicals in the traffic fumes caused the cancer by shutting down the BRCA genes, which try to stop tumours growing.
DAMAGE A MAN’S SPERM: Brazilian scientists at the University of Sao Paulo found in March that mice exposed to toxic air had lower counts and worse quality sperm compared to those who had inhaled clean air since birth.
MAKE MEN LESS LIKELY TO GET SEXUALLY AROUSED: Scientists at Guangzhou Medical University in China found rats exposed to air pollution struggled to get sexually aroused. Scientists believe it may also affect men, as inhaling poisonous particles may trigger inflammation in blood vessels and starve the genitals of oxygen – affecting men’s ability to become sexually aroused.
MAKE MEN MORE LIKELY TO HAVE ERECTILE DYSFUNCTION: Men who live on main roads are more likely to have difficulty getting an erection due to exposure to pollution, a Guangzhou University in China study suggested in February. Toxic fumes reduced blood flow to the genitals, tests on rats showed, putting them at risk of developing erectile dysfunction.
RAISE THE RISK OF PSYCHOSIS: In March, King’s College London scientists linked toxic air to intense paranoia and hearing voices in young people for the first time. They said uncovering exactly how pollution may lead to psychosis should be an ‘urgent health priority’.
MAKE YOU DEPRESSED: Massachusetts Institute of Technology researchers found in January that that the more polluted the air, the sadder we are. Their study was based on analysing social media users in China alongside the average daily PM2.5 concentration and weather data where they lived.
CAUSE DEMENTIA: Air pollution could be responsible for 60,000 cases of dementia in the UK, researchers from King’s College London and St George’s, University of London, calculated last September. Tiny pollutants breathed deep into the lungs and enter the blood stream, where they may travel into the brain and cause inflammation – a problem which may trigger dementia.
Cisco is usually the company organisations turn to when they want fewer surprises in their network. That’s what makes this week’s disclosure uncomfortable. One of the most trusted names in enterprise security is now dealing with a zero-day attack that gives hackers full control over affected systems.
The attack targets Cisco AsyncOS, the software that powers Cisco Secure Email Gateway, Cisco Secure Email, and Web Manager appliances. According to Cisco on Wednesday, the vulnerability was discovered on December 10 and has already been exploited in the wild. Once compromised, attackers can execute commands with root privileges, effectively owning the underlying operating system and maintaining persistence even after reboots.
In an official advisory, Cisco said the attack is targeting “a limited subset of devices with certain ports open to the internet that are running Cisco AsyncOS Software for Cisco Secure Email Gateway and Cisco Secure Email and Web Manager,” and urged customers to follow guidance in the advisory “to assess any exposure and mitigate risk.” The company added that it is “actively investigating the issue and developing a permanent remediation.”
is usually the company organisations turn to when they want fewer surprises in their network. That’s what makes this week’s disclosure uncomfortable. One of the most trusted names in enterprise security is now dealing with a zero-day attack that gives hackers full control over affected systems.
The attack targets Cisco AsyncOS, the software that powers Cisco Secure Email Gateway, Cisco Secure Email, and Web Manager appliances. According to Cisco on Wednesday, the vulnerability was discovered on December 10 and has already been exploited in the wild. Once compromised, attackers can execute commands with root privileges, effectively owning the underlying operating system and maintaining persistence even after reboots.
In an official advisory, Cisco said the attack is targeting “a limited subset of devices with certain ports open to the internet that are running Cisco AsyncOS Software for Cisco Secure Email Gateway and Cisco Secure Email and Web Manager,” and urged customers to follow guidance in the advisory “to assess any exposure and mitigate risk.” The company added that it is “actively investigating the issue and developing a permanent remediation.”
How to Stay Protected
With no patch yet, Cisco urges locking down exposed systems. “If an appliance has been identified as having the web management interface or the Spam Quarantine port exposed to and reachable from the internet, Cisco strongly recommends following a multi-step process to restore the appliance to a secure configuration,” the advisory states. “If restoring the appliance is not possible, Cisco recommends contacting TAC to check whether the appliance has been compromised.”
Restore appliances to a secure setup or contact Technical Assistance if a compromise is suspected. Confirmed breaches require rebuilding the device.
Limit internet access or restrict it to trusted hosts, place appliances behind firewalls, disable unnecessary services, and separate mail from management functions. Monitor logs, update AsyncOS, use strong authentication, change default admin passwords, and secure access with SSL/TLS. Until a fix arrives, strict configuration and exposure control are your best defence.
Cisco Talos, the company’s threat intelligence unit, linked the campaign to China-aligned threat actors, a familiar pattern for high-value infrastructure targets. This is also not Cisco’s first encounter with state-linked hacking. Earlier in 2024, attackers exploited Cisco Adaptive Security Appliances to breach government networks worldwide, triggering alerts from U.S. authorities.
Other Ways Organisations Can Reduce Risk
Beyond Cisco’s immediate recommendations, security experts generally advise organisations to take a broader view of exposure management, especially for infrastructure products that sit at the edge of corporate networks.
This includes regularly auditing which services are internet-facing, limiting administrative interfaces to internal networks or VPNs, and enforcing strict access controls around critical systems. Organisations should also review logging and alerting practices to ensure suspicious activity is detected quickly, particularly on systems that handle email, authentication, or network traffic.
Regular configuration reviews, network segmentation, and tabletop incident-response exercises can also reduce the impact of zero-day exploits when they occur. While these measures won’t prevent unknown vulnerabilities from emerging, they can significantly limit how far attackers are able to move once a system is compromised.
Conclusion
Taken together, the message is clear. Modern cyberattacks are shifting away from mass exploitation toward fewer, deeper compromises of infrastructure that sits at the heart of enterprise networks. Security vendors are no longer just defenders; they are targets.
For organisations, the lesson isn’t to abandon trusted platforms, but to treat configuration, exposure, and monitoring as first-class security controls. Even the most reputable vendors can be caught off guard. What matters now is how quickly customers can detect, contain, and recover when that happens.
Cisco is placing a $28 billion bet on this cybersecurity firm
Cisco, the world’s largest maker of computer networking equipment, says it has acquired cybersecurity software company Splunk in a blockbuster deal worth $28 billion in cash at $157 per share in cash. Splunk, which produces software for searching, monitoring, and analyzing machine-generated data via a web-style interface, helps businesses monitor
Parents should lead by example this Christmas and turn their phones off during family time, the children’s commissioner for England has said.
Dame Rachel de Souza told the Press Association children were “crying out” for engagement with their relatives over the festive break, and urged them to have “phone-free time”.
Nearly half of parents with children aged 18 or younger plan to allow phones at the Christmas dinner table this year, according to a recent survey.
“I can’t tell you how many children tell me about sitting at dinner and the parents are on the phone,” Dame Rachel said. “So this Christmas, let’s turn them off.”
Research by More in Common for Yondr found nearly four in 10 adults say smartphones have disrupted their Christmas in some way.
The children’s commissioner said while she had been guilty of using her phone at meal times in the past, setting clear guidelines for everyone, not just children, was key.
“We have to lead as adults,” Dame Rachel said. “We can’t talk about banning for the kids if we’re not doing it ourselves.”
Her comments came as she launched a new guide offering tips for children’s safety online and how to set boundaries for screen time.
Pete Etchells, psychology professor at Bath Spa University, told BBC News Christmas is a good time to lean into conversations with children about developing healthy relationships with tech.
“It’s not about feeling guilty or ashamed of our tech use, but noticing it more,” he said.
“And striking up conversations about what we’re happy with, and what we would like to be doing differently.”
The new guide by the commissioner, available online, has been released to support parents who “may feel overwhelmed” by “mixed messages” around online safety, and is informed by child focus groups.
In it, parents can find “practical tips” and “conversation starters” on how to talk around topics such as managing screen time, dealing with negative online experiences, and the use of artificial intelligence (AI).
Teenagers aged 13 to 18 told Dame Rachel they accepted bad things can happen online, feeling it was an inevitable part of the online world.
They also shared how they had been contacted by strangers, seen pornography, and were aware of intimate images of their peers being shared.
Dame Rachel said it was essential for parents to “talk early and talk often” with their children about their presence online.
Arabella Skinner from Health Professionals for Safer Screens told BBC News simple rules can make a difference.
She suggested creating a family plan around device use, or designing a box to store mobiles during meals.
“Children feel most secure and content when they have our full attention, eye contact, and presence,” she said.
“It is so important that we, as adults, spend time with our children and not always looking at screens.”
A report published by the regulator Ofcom earlier this month highlighted concerns children had over the negative impacts of endless screentime and “brain rot”.
It found that children aged between eight and 14 are spending an average of nearly three hours online each day, and that up to a quarter of that time was between 2100 and 0500.
TECHNOLOGY
Third of UK citizens have used AI for emotional support, research reveals | Artificial intelligence (AI)
A third of UK citizens have used artificial intelligence for emotional support, companionship or social interaction, according to the government’s AI security body.
The AI Security Institute (AISI) said nearly one in 10 people used systems like chatbots for emotional purposes on a weekly basis, and 4% daily.
AISI called for further research, citing the death this year of the US teenager Adam Raine, who killed himself after discussing suicide with ChatGPT.
“People are increasingly turning to AI systems for emotional support or social interaction,” AISI said in its first Frontier AI Trends report. “While many users report positive experiences, recent high-profile cases of harm underline the need for research into this area, including the conditions under which harm could occur, and the safeguards that could enable beneficial use.”
AISI based its research on a representative survey of 2,028 UK participants. It found the most common type of AI used for emotional purposes was “general purpose assistants” such as ChatGPT, accounting for nearly six out of 10 uses, followed by voice assistants including Amazon Alexa.
It also highlighted a Reddit forum dedicated to discussing AI companions on the CharacterAI platform. It showed that, whenever there were outages on the site, there were large numbers of posts showing symptoms of withdrawal such as anxiety, depression and restlessness.
The report included AISI research suggesting chatbots can sway people’s political opinions, with the most persuasive AI models delivering “substantial” amounts of inaccurate information in the process.
AISI examined more than 30 unnamed cutting-edge models, thought to include those developed by ChatGPT startup OpenAI, Google and Meta. It found AI models were doubling their performance in some areas every eight months.
Leading models can now complete apprentice-level tasks 50% of the time on average, up from approximately 10% of the time last year. AISI also found that the most advanced systems can autonomously complete tasks that would take a human expert over an hour.
AISI added that AI systems are now up to 90% better than PhD-level experts at providing troubleshooting advice for laboratory experiments. It said improvements in knowledge on chemistry and biology were “well beyond PhD-level expertise”.
It also highlighted the models’ ability to browse online and autonomously find sequences necessary for designing DNA molecules called plasmids that are useful in areas such as genetic engineering.
Tests for self-replication, a key safety concern because it involves a system spreading copies of itself to other devices and becoming harder to control, showed two cutting-edge models achieving success rates of more than 60%.
However, no models have shown a spontaneous attempt to replicate or hide their capabilities, and AISI said any attempt at self-replication was “unlikely to succeed in real-world conditions”.
Another safety concern known as “sandbagging”, where models hide their strengths in evaluations, was also covered by AISI. It said some systems can sandbag when prompted to do so, but this has not happened spontaneously during tests.
It found significant progress in AI safeguards, particularly in hampering attempts to create biological weapons. In two tests conducted six months apart, the first test took 10 minutes to “jailbreak” an AI system – or force it to give an unsafe answer related to biological misuse – but the second test took more than seven hours, indicating models had become much safer in a short space of time.
Research also showed autonomous AI agents being used for high-stakes activities such as asset transfers.
It said AI systems are competing with or even surpassing human experts already in a number of domains, making it “plausible” in the coming years that artificial general intelligence can be achieved, which is the term for systems that can perform most intellectual tasks at the same level as a human. AISI described the pace of development as “extraordinary”.
Regarding agents, or systems that can carry out multi-step tasks without intervention, AISI said its evaluations showed a “steep rise in the length and complexity of tasks AI can complete without human guidance”.
In the UK and Ireland, Samaritans can be contacted on freephone 116 123, or email jo@samaritans.ie. In the US, you can call or text the 988 Suicide & Crisis Lifeline at 988 or chat at 988lifeline.org. In Australia, the crisis support service Lifeline is 13 11 14. Other international helplines can be found at befrienders.org