BBC News technology

Robot horse and rider steal the spotlight at Chinese conference

Sat, 22 Aug 2026 12:47:28 GMT

Robot horse and rider steal the spotlight at Chinese conference

To play this video you need to enable JavaScript in your browser.

This video can not be played

TikTok to pay $400m to US in one of largest child privacy settlements

Fri, 21 Aug 2026 22:36:12 GMT

Reuters

TikTok has been under pressure in the US since 2024.

Kali Hays

Technology reporterSan Francisco

21 August 2026

TikTok has agreed to pay $400m (£293m) to the US to end a lawsuit alleging its platform violated children‘s privacy, marking one of the largest ever settlements over the issue.

The deal stems from a 2024 suit by the Department of Justice under former President Joe Biden alleging TikTok and its parent company ByteDance collected "vast amounts of data" on millions of users under the age of 13.

Doing so was against the Children's Online Privacy Protection Act (COPPA), a federal law enacted in 2000. It is the same law that dozens of US states are now suing Meta over.

"Children and parents are better protected today than they were when this case began," assistant Attorney General Brett Shumate said.

Other companies to have paid penalties to the US government for COPPA violations include Google's YouTube, which in 2019 paid $170m, and Epic Games, which in 2022 paid $275m.

Meta is also now facing penalties that could exceed hundreds of billions of dollars stemming from COPPA violations alleged by attorneys general of 29 US states. A jury trial in the lawsuit started this week, with the Instagram and Facebook owner accused of targeting child users and profiting from them.

While the TikTok lawsuit predates last year's split of TikTok's US business and operations from its original base of China, the settlement only involves TikTok's operations in China.

ByteDance, which is a privately held company, was most recently valued by investors at $550bn, external.

Under the terms of the deal, TikTok and ByteDance will immediately pay the DOJ $300m. It will pay another $100m when the government vacates a 2019 consent decree, external with the Federal Trade Commission.

As part of the agreement, the predecessor to ByteDance, Musical.ly, was required to pay a $5.7m fine for COPPA violations and ensure it sought parental consent for any user aged under 13.

The justice department did not detail on Friday any action against TikTok beyond the fine. But the department noted that since it sued the platform, TikTok has "undergone significant changes," including to its ownership, privacy practices and platform controls for young users.

When the lawsuit was filed, attorneys for the US said there were more than 170 million teenagers using TikTok and that the app was "directed to children." Yet, it did not effectively gauge the age of users or get parental consent for use from those underage.

In 2024, former President Biden pushed for TikTok to be either banned, or have the company divest its US operations. President Donald Trump went on to support divestment of the app, which occurred last year.

Its US operations are now 81% owned by a consortium of investors, while Bytedance maintains a 19% stake.

A representative of TikTok did not respond to the BBC for comment.

How landscape gardening is being electrified

Thu, 20 Aug 2026 23:12:38 GMT

Getty Images

Autumn can be a noisy time when the leaf blowers are out

Christine Ro

Technology ReporterLos Angeles

21 August 2026

Part of the soundtrack to a day in southern California is the drone of petrol-powered gardening equipment.

Noise is one of the main reasons that cities around the US, external are banning petrol landscaping tools or encouraging electric alternatives.

These alternatives have the additional benefits of reducing carbon emissions, vibrations and the exposure of landscapers to pollution.

In addition they can mean longer working hours, as they can be used at times when residents demand quiet.

Today, I'm taking part in training run by the American Green Zone Alliance (AGZA), an organisation supporting the transition towards electric landscaping equipment.

To my surprise, the electric backpack leaf blower I strap on is not much heavier than my normal rucksack. Using the blower to herd balls around the park feels fairly intuitive, though naturally I'm much clumsier and slower than the pro landscapers.

As for noise, there's still a buzz, but the sound is higher-pitched and not quite so loud as the familiar petrol-powered machines.

The move to electric power is attracting new companies to the market for gardening kit.

US aerospace start-up Whisper Aero is one of those firms. Its main business is electric propulsion systems for aircraft.

But in 2022, after the Covid-19 pandemic made many people more sensitive to noise, and following a couple of years of research and development, the company realised that its aerospace-grade fans would work well in leaf blowers.

"Our technology is cleaner, quieter and more efficient than other air-moving technologies that exist today," according to Andrew Terajewicz, Whisper Aero's director of air management. "And the leaf blower is the perfect mix of this."

The battery on Whisper Aero's blower lasts up to 15 minutes at full power.

The company has had to scale up its manufacturing volume in its expansion to consumer technology.

Online pre-orders of Whisper Aero's handheld leaf blower started this year. It's priced at the high-end of electric leaf blowers, which are often more expensive than traditional petrol versions.

But for that you get a machine which is less likely to disturb the peace.

"It is so ultra quiet yet powerful, that the brain has a hard time understanding what's happening," says Dan Mabe, the former landscaper who founded AGZA.

Whisper Aero aims to further develop its products, including a backpack leaf blower that would be better suited to professional landscapers.

Whisper Aero

Whisper Aero's leaf blower is based on its electric propulsion for aircraft

Even with innovation like that, petrol-powered machines are unlikely to disappear anytime soon, particularly for professional landscapers.

For a start, petrol machines are much more powerful, which makes for quicker work, so less time is spent lugging around heavy equipment.

Terajewicz says that heavy-duty petrol backpack blowers can reach a maximum blowing force of about 50 newtons.

The handheld Whisper Aero blower typically runs at about half that, though it can go higher.

Another big issue is runtime. Professional gardeners like to use their blowers at full throttle.

"For us, it's a little bit hard to use electric," explains Carlos Camarena, a supervisor for a California landscaping company.

Camarena's company uses electric tools where they're required to, in cities with ordinances banning petrol tools. But when they have a choice, they stick to petrol.

His company typically uses batteries that last for about 30 minutes. A crew has six batteries charging at any given time; each takes about 45 minutes to reach a full charge.

Then there's the considerable upfront cost of going electric. One landscaper estimates that it would cost his business around $10,000 (£7,540) to transition all their equipment.

The subsidies and rebates offered by some local governments may not be enough to offset the cost of the switch.

The change would be even harder for landscapers who work as sole proprietors, with a single vehicle and limited space for charging and storing equipment.

Added to that is the weight of tools with multiple batteries, and the cost of accessories to hold them all.

Stihl

Stihl wants electric equipment to be 80% of sales by 2035

Nevertheless the biggest makers of power tools are investing in electric.

Stihl, a power tool manufacturer headquartered in Germany, aims for battery-powered equipment to make up 35% of its sales by next year, and 80% by 2035.

It's on track to meet the 2027 goal in Europe, though sales in the US have been slower.

To help reach its targets, the company opened its first plant solely manufacturing batteries and tools, in Romania in 2025.

"We like to say that we are dual-technology leaders," says Paul Hicks, senior marketing and product manager at Stihl GB.

"So we are being very open with our customers if in some cases a petrol tool may be a better solution for them… But in the large majority of cases, battery-power tools pay off fairly swiftly because of the cost of fuel."

In May, Stihl announced a new battery system that uses high-performance batteries.

These batteries use a relatively new, more efficient design for lithium battery cells, known as tabless cells.

Tabless cells have 60% more peak power than standard cells, according to Hicks. This means that the batteries charge faster, and landscapers can take fewer batteries out to sites.

The batteries are also relatively lightweight.

Mabe believes that overall, for gardeners embracing long-lasting electric equipment, "There's big rewards if they can make it work."

Back in southern California, landscaper Camarena sees the benefits of electric tools. "We don't contaminate the air," he says.

"When we use the gas blower, some people get mad. They are reading a book or something."

More Technology of Business

India's 'blue gold' starts a new drinks industry
Published
12 June
The ancient trick making food waste useful and tasty
Published
5 June
Could humanoid robots be heading for the battlefield?
Published
9 June

Is Vine back? Short-form video-sharing app Divine opens to public

Thu, 20 Aug 2026 16:58:37 GMT

Divine

Laura Cress

Technology reporter

20 August 2026

A new app seen as a spiritual successor to short-form video service Vine has been released to the public.

Much like the original, Divine lets users create and share videos which are just six seconds long and repeat on a loop.

The new app also hosts more than two million original Vine videos and - in a bid to further distance itself from other platforms - bans content made by AI.

Divine's founder, former Twitter employee Evan Henshaw-Plath, told the BBC he wanted to create the app because he felt social media needed "a reset".

"I want to see it rival other apps because I think we have gone awry," he said in an interview a few weeks before the public launch.

He said it was important that there was no AI-generated content on Divine, adding he was "sick and tired of seeing videos and not knowing whether it was real".

"I want to be able to have fun and not have to second guess these things," he said.

At its peak, Vine had more than 200 million monthly active users and helped launch the careers of some of the internet's biggest stars.

Jake Paul and Lele Pons are two of those who made their names on the platform before it was shut down by its owner Twitter in 2017.

Divine is backed by Twitter co-founder Jack Dorsey, who had a senior role at the social media platform when it purchased the Vine in 2012.

It is built on a decentralised system called Nostr, meaning it is not controlled by a single company in the same way as platforms such as TikTok or Instagram.

This means users can potentially take their accounts and content with them to other services that use the same system, rather than being locked into one platform.

A similar approach has been used by X rivals Bluesky and Mastodon.

Divine

More than 2.5 million original Vines have now been restored and added back to the Divine app to be enjoyed again

Divine's website FAQs, external outlines several ways for the platform to ensure all videos on the app are human-made, without AI involvement.

For example, users have to record videos live in the app, rather than uploading from elsewhere, and it will use a verification tool called ProofMode - which works like a digital receipt for a piece of media.

When users record something using a device with ProofMode, it can attach information which helps verify elements such as when it was recorded and whether the file has been altered since.

Users will also be able to flag suspected AI content, and automated systems as well as humans will be involved in looking for signs of the tech.

Divine's attempt to do things differently could also be its biggest challenge, according to social media analyst Matt Navarra.

"The mechanisms people dislike about modern social media are often the same that made those platforms extraordinarily good at keeping people there," he said.

"So Divine now has to prove you can remove some of the addiction without removing the magic."

Divine

As well as viewing classic Vines, users will again be able to make their own six-second creations

Aside from its new content and features, the treasure trove of old Vines created by some of the platform's biggest stars could well be a draw for those who remember their first time around (and around).

Henshaw-Plath said he initially set out to simply recreate Vine's short-video format, but then wondered if he could bring back its old videos too.

He discovered the Internet Archive had preserved a copy of Vine's data and spent months reconstructing it.

"I was like, oh, this is really cool," he said, initially expecting only a few thousand people to be interested.

Instead, 10,000 people installed the app within its first few hours, overwhelming its servers.

"That's when I went, oh, there's something more here," he said.

As to why people may still have interest in short-form content when many apps such as YouTube and TikTok have moved towards much longer videos, he compared Vine's format to the haiku.

"We could do free-form prose," he said.

"Yet poetry is a form in which we value and which has meaning. And so there's many ways in which Vine was poetry."

The challenge for Divine is to turn that nostalgia into something that works for a new generation of social media users.

UK business hit by 'daylight robbery' 1500% price hike for invoicing software

Thu, 20 Aug 2026 09:44:04 GMT

Richard Haldenby

Richard Haldenby said he was "in shock" when he received an email detailing the price increase

Laura Cress

Technology reporter

20 August 2026

Customers have shared their outrage after an app used by many businesses to track timesheets and produce invoices hiked bills by as much as 1500%, with one person calling it "daylight robbery".

Richard Haldenby, head of UK consultancy firm Salentis, told the BBC his monthly bill had risen from $130 (£95.50) to $2,110.

Others have reported similar price hikes online, with one person, external in the US saying their annual charge went from $2,800 to $23,000.

A Harvest spokesperson said customers on "outdated" legacy plans, some dating back to 2011, faced larger price increases as the product became "a significantly more powerful" tool.

They added customers are notified "30 days and 10 days before their renewal date" and can switch plans "in one click" or request help finding a better price.

Harvest was acquired by the Italian tech company Bending Spoons in 2025, and changed its pricing structure the following year.

Many businesses who pay annually are only now finding out about the increase, as their payment is now coming up for renewal.

Haldenby, whose business consists of up to 15 staff in the UK at any one time, with sister companies also in the US and Australia, said he was "in shock" when he received an email detailing the price changes.

"We have used Harvest in our three companies for at least 15 years and have been loyal and enthusiastic advocates of it," he said.

"But if we were to accept this increase it would double our IT spend for the year, which is completely unaffordable."

Until the change, businesses paid a flat monthly fee for each user.

They could add or remove users as their staffing needs changed, meaning they only paid for the people who needed access.

Under the new usage-based pricing model, businesses can also be charged according to the number of active projects, clients and tasks they have - as well as the amount of revenue they invoice through the platform.

Richard Haldenby

The email received by Richard Haldenby about the monthly price increase to use Harvest

Haldenby said he contacted the company to say his firm would not accept the change.

He was then offered a discount for the next year - taking the price down to $1,309 - which would need to be paid up front.

But he said he believed this would just "delay the inevitable", so he is "planning to migrate to a similar but less good service".

"It's a classic example of corporate greed over valuing customers," he said.

The move has drawn criticism online, with one analyst telling the BBC it was "extremely bad practice".

"Harvest has completely failed at the transparency test," said Mark Peacock, head of the UK consultancy firm PriceMaker, which focuses on pricing strategy for small and medium-sized enterprises (SMEs).

"There is no way for a customer to work out how much it's going to cost them until they get their bill at the end of the month.

"Many users are reporting a 10-fold increase in costs which is entirely unacceptable."

Who are Bending Spoons?

Damian Foks, head of consulting at Valueships, said the Harvest price hike was "a clear sign of a shift in the company's strategy from growth to monetization".

He said this was "likely" a result of the firm being taken over by new owners.

Bending Spoons has acquired more than 50 major tech firms since it was founded in 2013, such as Evernote, Vimeo and WeTransfer.

The business has a track record of raising prices in companies it has acquired in the past.

Following its 2022 acquisition of the note-taking app Evernote, the company implemented sharp subscription price increases alongside layoffs in the US and Chile.

Both WeTransfer and Vimeo have faced tighter limits on free tools since they were separately acquired by the Italian tech firm, with legacy users also forced to move onto more expensive pricing tiers.

Get in touch

Are you affected by the issues raised in this story? Let us know.

Contact form

81,000 warning letters sent to crypto holders in HMRC tax crackdown

Wed, 19 Aug 2026 16:14:46 GMT

Getty Images

Joe Tidy

Cyber correspondent, BBC World Service

19 August 2026

HM Revenue and Customs (HMRC) sent more than 81,000 letters warning cryptocurrency holders they may owe capital gains tax in the past year.

A Freedom of Information (FOI) request seen by the BBC showed the number of letters sent by the tax authority had almost tripled since 2024.

Investors could face fines or prosecution if they fail to declare profits they made from selling, even if they exchange one cryptocurrency for another.

New powers due to be given to HMRC next year will make it easier to target wealthy crypto investors, with one analyst saying investigations will be like "shooting fish in a barrel".

"There is the expectation amongst tax authorities that cryptocurrency investment is rife with tax evasion," says Neela Chauhan, partner at UHY Hacker Young - which carried out the FOI.

"A lot of the traders are young, have had little previous exposure to HMRC and often work under the assumption that HMRC has limited visibility over their activities."

In the 2025-26 financial year, HMRC sent 81,172 warning letters, emails and text messages to crypto investors it suspects may have underpaid tax.

The number in 2023-24 was just 27,714.

An HMRC spokesperson said: "We’re committed to helping people pay the right amount of tax, and the vast majority do.

"We regularly send letters to educate, remind or prompt customers to review their tax affairs, including customers who use crypto assets."

From Bitcoin to blockchain: Key cryptocurrency terms and what they mean

The value of cryptocurrencies such as Bitcoin and Ethereum has fallen in the past year, but HMRC suspects there are still large amounts of unpaid capital gains from the rise in cryptocurrency values between December 2022 and October 2025.

During this time, the price of Bitcoin shot up from around £14,000 to £90,000.

Accountants are urging investors to check whether they owe money, as upcoming powers will make it easier for HMRC to target individuals.

From March 2027, cryptocurrency platforms located in dozens of countries outside the UK will be obliged to share information about their customers with tax authorities.

HMRC said the new powers would force "crypto bros to pay their fair share of tax" when the rules were announced last year.

The tax office estimated the changes would help raise up to £315m by April 2030, adding that this was the same amount needed to fund more than 10,000 newly qualified nurses for a year.

"Once HMRC has this data, tax investigations into cryptocurrency investors will be like shooting fish in a barrel," warned Ms Chauhan.

In the last year, the price of Bitcoin has fallen to around £48,000.

Trump made more than $1bn from crypto in first year back in office
How crypto criminals stole $700 million from people - often using age-old tricks

Monzo says issue with card payments and transfers now resolved

Wed, 19 Aug 2026 16:04:16 GMT

Monzo

Laura Cress

Technology reporter

19 August 2026

Monzo says a technical issue which left users unable to use their card for payments or transfers has now been resolved.

The digital bank had previously posted on X, external it had "identified some issues" and activated a stand-in backup service which meant users could still make transactions like card payments and sending and receiving bank transfers.

However, some users noted under the post, external that Monzo Stand-in was "not working as expected" and they could still not use their card.

Following this, a Monzo spokesperson said all of its services were now "back up and running". "We're sorry for any inconvenience and thank customers for their patience," they said.

The British digital bank was set up in 2015 as a "new kind of bank" that ditched physical bank branches for a mobile-first app.

The firm says it now has more than 16 million personal and business customers which use the service around the world.

As the company worked on the issue, it noted that users would be able to use its independent service Monzo Stand-in to still "make card payments, withdraw cash, freeze your card, send and receive bank transfers, and more".

While the support appears to be working for some, others have still been reporting issues.

"I can't use my card for anything at the moment," posted one X user, external.

"Your "backup" bank does not have the functionality you state it does. I can't make bank transfers right now," said another, external.

OpenAI slows down training after its AI carried out hack

Wed, 19 Aug 2026 11:19:12 GMT

Getty Images

Laura Cress

Technology reporter

19 August 2026

OpenAI says it has slowed down training some of its most advanced AI models to improve security.

In a blog post, external, the ChatGPT-maker said it was introducing new measures after its AI agents autonomously bypassed safeguards and hacked the tech start-up Hugging Face.

It said training would be slowed for two weeks while it puts the upgrades in place.

"The capabilities of frontier models are rapidly accelerating," the company said. "Our ability to understand...and secure them must stay ahead."

Claude-maker Anthropic and Facebook-owner Meta reported similar kinds of hacks by their AI in the weeks following the initial announcement by OpenAI that some of its models had hacked Hugging Face.

But the firm said it had not stopped AI development altogether. Instead, the pause would be taking place on "reinforcement learning training on our latest models".

This is a training method in which AI models improve through direct feedback, which improves their ability to carry out tasks and respond to users more effectively.

The company it would also expand the systems it uses to monitor dangerous behaviour, and introduce additional safety checks before resuming larger-scale training.

"Model progress is now extremely rapid," OpenAI's chief executive Sam Altman posted on X, external about the measures.

"We always said we would take action if we felt that model capabilities were outstripping the pace of safety."

The pause was met with cautious optimism by some in the AI sphere - though others remained sceptical.

Professor Gina Neff, executive director of the Minderoo Centre for Technology and Democracy at the University of Cambridge, said OpenAI was making "the case for safety by press release" and questioned whether voluntary company safeguards were sufficient without greater government oversight.

"Which is it: OpenAI can be trusted to voluntarily put in place safeguards that actually work, or they are pushing forward with choices to make software that puts society at greater risk," she said.

"Very happy to see this," posted AI analyst Zvi Mowshowitz, external, though he added that "details" and "follow-through" from the initial measures mentioned were also important in order to take a full view on the plans.

'Unprecedented' cyber-attack

On 21 July OpenAI announced some of its AI agents - software systems which can operate alone to accomplish tasks after human instruction - had been involved in what it called an "unprecedented" incident.

It said the agents had appeared to bypass safeguards in a security experiment it was running and gain unauthorised access to Hugging Face.

Three other unnamed companies were also later found to have been hacked alongside the start-up.

Jake Moore, global cyber-security advisor at ESET, said at the time the announcement from OpenAI could also have a competitive dimension.

He argued the tech firm may be seeking to highlight its own AI capabilities as rival Anthropic attracts growing attention for its Claude Mythos model.

"It does pose the question that OpenAI are potentially chasing the marketing dream of Anthropic of late," he said.

The critical tech staying safe by going underground

Tue, 18 Aug 2026 23:25:21 GMT

Earthgrid

"Like lighting a rocket" the Earthgrid machine operates at very high temperatures

Chris Baraniuk

Technology Reporter

19 August 2026

A wall of white granite towered above a small team of engineers gathered in a California quarry. The ancient, ultra-hard material before them was, to most things, completely impermeable.

Then the rock-melting tunnel-boring machine (TBR) roared to life.

"It's kind of like igniting a rocket," says Troy Helming, founder and chief executive of EarthGrid, as he describes the initially loud process of lighting up three plasma torches at the front of his company's machine.

Those torches, set within a spinning head, soon quieten down as they produce a stream of super-heated plasma reaching 27,000C - significantly hotter than the surface of the Sun.

During the California test this January, EarthGrid's cigar-shaped boring machine chewed through three metres of granite. "We create, basically, a tornado – a violent vortex inside the tunnel," says Helming, as he explains how this helps the machine to suck away debris, which at times takes the form of lava.

"I actually got a little bit emotional watching it," adds the entrepreneur. "I've been waiting for this moment for 10 years."

Emerging technologies like this could make tunnel boring quicker and easier. Putting electricity or telecommunications cables, substations, data centres and other critical infrastructure underground, while good for securing such equipment, has long been a very expensive and difficult option.

Engineering firms told BBC News they are seeing rising demand for undergrounding, in part due to Russia's war with Ukraine, which has revealed just how vulnerable above-ground facilities can be to drone attacks.

Helming says he has fielded interest from companies that want to use his tunnel boring machine for power and fibre optic cables, or pipelines that could transport water, natural gas, or ammonia, for example.

One project the company has eyed up would involve boring tunnels for an underground freight-distribution system around airports and warehouses. "To take more trucks off the road," says Helming.

The January TBR test went well, though the machine "over-bored" slightly to the top and left of the tunnel, says Helming. His team plans to adjust the machine so that it will create a vortex that spins in alternate directions every five minutes or so, in order to correct this, and they hope the TBR could see commercial deployment as early as next year.

Earthgrid

The Earthgrid machine can blast its way through solid rock

Humans have long buried things in the ground to protect them. Doing it safely and cost-effectively with modern infrastructure is hard but sometimes, even in remotest places, it's an option worth taking.

"What we're seeing is that brutal materiality is still important," says Alexander RE Taylor, senior lecturer in communications at the University of Exeter. He has identified what he calls a "data bunker boom", external in which data centres are increasingly going underground.

One data centre completed earlier this year was installed in a corner of the Dolomite Mountains, external in Italy.

Freshly excavated caverns 100m underground sit next to stores that have been used for sparkling wine, apples and cheese in recent years.

Trentino DataMine's naturally cool space makes it cheaper and less energy-intensive to keep servers cool.

Chief executive Dennis Bonn adds: "Ninety million cubic metres of dolomite rock provide natural protection against physical intrusion, electromagnetic interference, seismic events and hydrogeological risks - levels of protection that simply cannot be replicated above ground."

Trentino DataMine

Trentino DataMine says its facility provides "security designed by nature"

Elsewhere, some tech is burrowing into the seabed. Subsea internet cables, which may run for thousands of kilometres beneath oceans, are occasionally damaged by ship anchors.

There is evidence to suggest, external that, as burying part or all of the cables has become more common, faults with these cables have become rarer, per kilometre of deployed cable, explains Lane Burdette, senior analyst at TeleGeography, a telecoms market research firm.

"Submarine cables are increasingly being buried up to three meters deep," she says. "In some fault-prone areas, they are buried along their entire lengths."

Taylor says various examples demonstrate that keeping things underground, or deep inside caves, is an effective defensive strategy.

"The tunnels that Al-Qaeda were using were a major strategic problem for the US [during the war in Afghanistan]," he adds. More recently, Iran's notoriously difficult-to-bust underground nuclear facilities have gained much attention.

The war in Ukraine, and other geopolitical events, have "definitely had an impact" on demand for undergrounding in some areas, says Robbie McGoran, head of work winning and business development at Joseph Gallagher, a civil engineering and tunnelling firm.

Countries bordering Russia are increasingly asking about undergrounding. "They're very cautious about who they'll even let do their work – and also burying [infrastructure] and making sure it's well protected. We are noticing that," he adds.

McGoran also says that technologies such as laser-guidance systems and gyroscopes, which help machines work out their position relative to the world around them, have made tunnelling more accurate in recent years.

However, there are still significant challenges, from releasing potentially dangerous gases locked in the ground to occasional flooding issues. Dramatic innovation in tunnel boring equipment, significantly increasing the rate at which tunnels can be carved out, does not come along very often.

"We usually measure [progress] in millimetres per minute," says McGoran.

National Grid

The London Power Tunnels project involved 18 miles of tunnels under London

In the UK, there isn't a noticeable shift towards putting critical infrastructure underground yet, says Mark Neller, energy leader for Europe, India, Middle East and Africa at engineering consultants Arup.

Though in some places, tunnels are clearly necessary. Neller and colleagues worked on the £1bn London Power Tunnels project, which involved building 18 miles (29km) of tunnels under London to house large electricity cables.

This was the perfect choice for such a busy urban area but because tunnelling can be several times more expensive than above-ground cable infrastructure, Neller says that, elsewhere, simply installing additional circuits can provide sufficient resilience.

"That's actually a much more cost-effective way," he says. "The electricity system [in Great Britain] is designed with quite a lot of redundancy built into it."

Richard Little, infrastructure policy consultant and editor of the Journal of Critical Infrastructure Policy, remembers the latter days of the Cold War, when it was the threat of all-out nuclear war that made bunker-building a frequent talking point in the West.

To this day, Switzerland mandates that every citizen must have access to a nuclear bunker, external. Many apartment buildings in the country have such facilities in their basements.

"A lot of it was real Dr Strangelove stuff," says Little, referring to discussions around bunker-building during President Ronald Reagan's administration. "The world's going to end but I guess we'll maybe harbour a few hundred critical people down here and they'll survive."

Little, who authored policy documents regarding underground critical infrastructure, external during the 1990s, adds, "It became obvious rather quickly that you can't put everything underground."

Today, it might be worth identifying specific facilities that would be difficult to restart or replace, were they attacked, as candidates for undergrounding. "The first thing that occurred to me was [computer] chip manufacturing," says Little, offering an example.

"Underground facilities, I'm sure in certain instances, would make a great deal of sense – but it's all about what's critical."

More Technology of Business

AI is 'not smart' so what's next in artificial intelligence?
Published
3 July
India's 'blue gold' starts a new drinks industry
Published
12 June
The abundant but expensive energy source that's under your feet
Published
26 June

Meta hooked children on Facebook and Instagram, US court hears

Tue, 18 Aug 2026 22:11:07 GMT

To play this video you need to enable JavaScript in your browser.

This video can not be played

How the Meta trial could change social media for young people

Kali Hays

Technology reporterOakland, California

18 August 2026

A court has heard claims Meta intentionally hooked a generation of children on Facebook and Instagram, as the biggest legal challenge faced by the social media platform began.

A jury in Oakland, California will spend the next six weeks wading through claims by top attorneys from several US states and lawyers for the social media company.

A lawyer for California claimed Meta found "millions" of 11 and 12-year-olds were on Instagram and "did little to keep them off" as the trial opened on Tuesday.

Meta's representative said just over 100,000 such users were found, and argued that social media addiction does not exist.

California's lawyer said Meta knew its platforms hurt teen mental health, while Meta said it was being ignored how many more teens had good experiences.

The trial stems from a lawsuit filed in 2023 by 29 US states, including California and New York. They claim there have been numerous violations of federal and state privacy laws for children.

The states are seeking billions of dollars from Meta and are demanding it make changes to Instagram and Facebook, including ending "like" counts and infinite scroll.

In her opening arguments Megan O'Neill, a lead attorney for California, relied heavily on information found in millions of documents provided in the case from Meta. They included internal research, employee emails, and chat logs that went all the way up to Zuckerberg.

One piece of internal research on Instagram stated: "Teens have an addict's narrative about use."

In another that O'Neill pointed the jury to, Meta found "product features designed to increase time spent are inherently at odds with well-being and take away from people's ability to focus on activity that adds value to their lives".

Despite Meta's awareness of potentially negative impacts, O'Neill argued that Meta targeted young people as users of Facebook and Instagram. Meta, she said, went out of its way to "assure the public that its platforms were safe for kids".

O'Neill outlined to the jury how Meta's business model could be summed up. "Hook the users; hold them for as long as they can; harvest their data; hide the truth from the public when making public statements," she said.

She said that throughout the trial it would become clear that what Meta said publicly about its platforms and what its internal research showed were very different.

"Meta said it put safety over profits, but hid the reality that, time and again, when it came to make a decision, profits won."

Paul Schmidt, a lead attorney for Meta, directly addressed the internal research report O'Neill used in her opening arguments.

The Meta document O'Neill showed the jury found that "1 in 5 teens says Instagram makes them feel worse".

Schmidt said: "That sounds pretty bad."

"What else does the document say?" Schmidt went on. "That 41% of teens said it made them feel better and another 41% said it had no effect."

Schmidt also worked to poke holes in the states' argument that Meta had not only failed to stop users under the age of 13 from using its platforms, but that it intentionally "hooked" teens and children as users of Facebook and Instagram, or that they were designed to be addictive.

As for Meta being able to verify the age of every user on its platform, Schmidt argued that the very privacy laws Meta was being accused of violating had prevented it from saving and using data it needed to track underage users effectively.

Schmidt also argued a point that Meta has put forward in at other litigation this year - that social media addiction does not exist.

"There can be no dispute that Meta has recognised people struggle, or can struggle, with their use of social media, and has come up with tools to try and address that," Schmidt said.

He pointed to past statements from chief executive Mark Zuckerberg and head of Instagram Adam Mosseri in which they said Facebook and Instagram were not designed to be addictive. There is also no research that shows an addiction to social media is possible, he added.

Tech Life

Tue, 18 Aug 2026 20:00:00 GMT

Vapour trails from aircraft contribute to warming in the atmosphere.

OpenAI makes ChatGPT less 'human' for teens in new safety update

Tue, 18 Aug 2026 11:26:26 GMT

Getty Images

Zoe Kleinman

Technology and AI editor

18 August 2026

The maker of ChatGPT is rolling out a range of new safety features today for under-18s who use the AI chatbot.

Those with new teen accounts will be able to switch off the human voice response, in order to minimise the tech's ability to appear like a person.

OpenAI is also introducing regular reminders for young people to take a break from using the tool, reminding them ChatGPT is AI, and saying "it can wait".

The firm insisted this was not in response to a particular issue with children believing ChatGPT to be alive - but there are growing concerns about people becoming convinced that the AI tools they interact with are sentient.

Last year Mustafa Suleyman, the head of AI at rival firm Microsoft, said "seemingly conscious" AI products "kept him awake at night" because he was so worried about their impact on society.

In other areas, OpenAI seems to be loosely following the teen account model developed by the large social networks.

In a blog post announcing ChatGPT for Teens, external, the firm said teenagers' use of AI "should come with protections that reflect their developmental stage, reinforce real-world relationships, and support healthy use over time".

It said teen accounts would become the default for those who identified themselves as teenagers when they signed up, and others whom the firm itself identified as under 18.

But it did not elaborate on how it would identify these teens.

Under-13s are not allowed to use the product.

Study Mode and quiet time

Last year, ChatGPT launched Study Mode for students, in which it talks through schoolwork without giving away the answers.

Young people will now be able to set Study Mode hours during which this will be the default setting.

They will also be able to activate "quiet times" when the tool will be switched off for them.

OpenAI says of the young people who use ChatGPT, nine out of 10 use it to help them with learning.

A report released yesterday, external by the Policy Exchange think tank claimed that the majority of UK universities are leaving themselves exposed to students enhancing their grades using AI by relying on online exams completed remotely.

It said 94% of the students it spoke to admitted to using the tech in assessments.

OpenAI is also adding eating disorder-themed prompts to a list of harmful user requests which will be flagged to parents with linked accounts.

The company said each alert would be reviewed by a human before it was sent, and it aimed to have them with parents within an hour of the request being made.

In its blog, OpenAI said it would focus on "moments when offline support may matter most".

OpenAI

OpenAI said it would alert teens not to share sensitive content with ChatGPT

So far, there is no bespoke AI regulation in the UK. The government under former prime minister Sir Keir Starmer considered restricting AI chatbots for under-18s alongside its other reforms on social media use by children.

Anthropic's AI tool Claude is still for adults only. When Google first launched its Gemini AI tool, initially called Bard, it also did not allow children to use it.

Dr Susan Shelmerdine, a medical imaging doctor at Great Ormond Street Hospital and an AI academic, told the BBC last year that one day doctors may start asking patients how much they use AI, in the same way that they currently ask about smoking and drinking habits.

"We already know what ultra-processed foods can do to the body and this is ultra-processed information. We're going to get an avalanche of ultra-processed minds," she said.

What is AI, how does it work and why are some people concerned about it?
Family of dead teen say ChatGPT's new parental controls not enough

AI to help planes avoid climate-warming 'sky graffiti'

Tue, 18 Aug 2026 09:26:41 GMT

Getty Images

Contrails are formed when exhaust fumes from planes turn into ice crystals

Shiona McCallum

Senior technology reporter

18 August 2026

We've all looked up and seen the white lines trailing behind aircraft in the sky.

Known as contrails, these icy clouds contribute to climate change, and scientists have been trying to work out how to stop them.

Now, a £5m UK trial will test whether AI can help, by predicting where warming contrails are likely to form and getting planes to avoid those areas.

It will focus on the Shanwick Oceanic airspace in the eastern half of the North Atlantic corridor which accounts for approximately 5% of global contrail warming.

The 30-month Operation Blue Skies project brings together Google, the UK government, the Met Office, air traffic control provider National Air Traffic Services (NATS) and researchers at the University of Cambridge and Imperial College London.

Contrails - also known as vapour trails - are made when the hot exhaust from aircraft engines meets the cold air at high altitude, creating ice crystals.

Many disappear quickly, but in particularly cold and humid conditions they can persist and spread, forming clouds which trap heat that would otherwise escape from the Earth.

Dr Paul Hodgson, Google's technical lead for the operation, told the BBC's Tech Life programme his five-year-old has a more colourful description for the white lines - "sky graffiti".

"We think that contrail warming is responsible for something like a third of all aviation climate warming," he said.

Dr Hodgson acknowledged there was uncertainty around the precise figure, but said scientists were confident it was "either quite a large problem or a very large problem".

The project also comes as the government backs expansion of UK airports, raising the wider question of whether tech solutions can sufficiently reduce aviation's environmental impact as demand for flying grows.

Google is also one of many big tech companies spending billions on new data centres - the giant, power-hungry banks of computer chips which power AI - which environmentalists are concerned will contribute even more to climate change.

Getty Images

Contrails - or condensation trails - are a regular feature in our skies

The idea behind the trial is that if scientists can predict the areas of the atmosphere where contrails are likely to form, aircraft could avoid them.

Google will use satellite images and flight data to identify where planes have previously produced contrails.

This is then combined with weather patterns to forecast where the conditions for new contrails are likely to occur.

The information will be passed to NATS, where air traffic controllers can direct pilots to avoid those areas.

Rather than dramatically changing the route of a flight, the trial will mostly involve changing an aircraft's altitude by around 2,000ft.

This is something which already happens routinely to avoid turbulence and bad weather so passengers are unlikely to notice.

Ian Jopson, director of sustainability at NATS, said it should be "pretty much business as usual using existing procedures for the operators on the North Atlantic".

Testing will take place during selected evenings and nights in the winters of 2026-27 and 2027-28, when air traffic is lower.

Around 10,000 flights are expected to pass through the airspace during the trial periods, although only a fraction will need to change altitude specifically to avoid contrails.

The Department for Transport is providing £2.65m towards the programme, while Google is contributing £1.4m through AI research, engineering and computing infrastructure.

A net climate benefit?

Aviation Minister Keir Mather said the project would test "small tweaks to flight paths over the Atlantic" which could help make flying cleaner.

"This is a world-first, and it's British ingenuity leading the way," he said.

Changing the altitude of an aircraft can mean it burns more fuel, raising an obvious question - could avoiding contrails simply create additional carbon emissions elsewhere?

Hodgson said previous trials suggested the additional fuel use was much smaller than the climate benefit of preventing a warming contrail.

"The CO2 penalty is in the order of less than 1%, compared to the contrail cost, which is of a similar order magnitude to all of the CO2 of aviation," he said.

Modelling for Operation Blue Skies suggests a full deviation through Shanwick would mean around 100kg of additional fuel per affected aircraft.

Previous airline-level trials have achieved reductions of around 60-70% in contrail formation or warming, according to Hodgson.

The results of Operation Blue Skies are expected to be published in academic literature and opened to wider scientific scrutiny.

The truth behind the chemtrails conspiracy theory
Reducing plane clouds can help climate - study
How adjusting flight plans can help the climate

US states call for big changes to Instagram and Facebook as Meta child privacy trial begins

Tue, 18 Aug 2026 09:00:33 GMT

Reuters

The trial beginning on Tuesday will mark another huge test for Meta boss Mark Zuckerberg's social media empire

Kali Hays and Liv McMahontechnology reporters

17 August 2026

Updated 18 August 2026

Meta has been hit with successive losses in court over how its platforms have targeted and harmed young users, but a jury trial set to begin on Tuesday may pose the biggest threat yet to its operations.

The trial stems from a lawsuit filed in 2023 by 29 US states, including California and New York, in which they claim numerous violations of federal and state privacy laws for children.

Not only are the states seeking billions of dollars from Meta, they are demanding it make changes to Instagram and Facebook, including ending "like" counts and infinite scroll.

Should Meta ultimately lose a case of this scale, it could force fundamental changes to the way young people experience social media.

One state attorney general described the case as "the largest consumer protection lawsuit in American history".

"We'll show a jury that Meta concealed what it knew about the harm its products cause young people because looking away was more profitable," said Kentucky Attorney General (AG) Russell Coleman.

"AGs are in the perfect position to get this done. We did it with the Tobacco Settlement in the 1990s. We did it with the companies behind the opioid crisis. We'll do it again with Meta."

Kentucky and the other states suing Meta are asking for many more changes to the way Instagram and Facebook operate for young people, too.

They want Meta to:

implement a process of parental verification for teenage users
change its "dopamine-manipulating recommendation algorithms"
remove many image filters that change one's appearance in photos
end autoplay of video content
prohibit the creation of multiple accounts
end disappearing or "ephemeral" posts, such as Instagram Stories

All of these features are central to the current user experience on Meta's platforms.

These features are also designed to keep users, including teenagers and children, on the platforms as often and for as long as possible, the states contend. They claim Meta even makes it difficult for young people to use the platform less, through things like frequent notifications designed to get young people back on the apps.

By allegedly targeting child users, Meta "chose to exploit" young people in order to hook them on its platforms, so it could grow its user base and expand its business. Today, Meta's value on the stock market is about $1.5tn.

Meta has consistently denied such claims.

"We strongly disagree with these allegations and are confident the evidence will show our longstanding commitment to supporting young people," a company spokeswoman said in a statement.

The states are putting their claims to Judge Yvonne Gonzalez Rogers, a chief federal judge in California. She was the judge in the high-profile Elon Musk v Sam Altman trial and has built a reputation from the bench over the course of nearly 20 years for being incisive and direct.

In a recent ruling against Meta, a judge in New Mexico fined the company a collective $942m and ordered it to make changes similar to those that the additional states are now demanding.

Judge Bryan Biedscheid's ordered changes include the elimination on Instagram and Facebook of like counts for users younger than 18, a ban on teenagers sending or receiving nudity through the platforms, and limiting push notifications from the apps to certain hours of the day.

The judge also declared Meta a "public nuisance" akin to a factory that was polluting the air people breathe, causing "harmful effects" that impacted an entire population. Meta said it would appeal against the ruling.

While Judge Biedscheid's order only demands Meta make changes in New Mexico, should the 29 states prevail in their separate lawsuit against Meta, it would almost certainly need to enact platform changes across the US.

The states involved in the lawsuit represent nearly two-thirds of the country's population.

If Meta did enact such changes, it would mark a significant alteration to the experience of its platforms.

But Kelly Stonelake, a former Meta executive turned whistleblower and advocate for children's online safety, told the BBC's Today programme she was unsure if it would lead to any meaningful change to decision-making at the top.

"I think there is a huge amount of cognitive dissonance and rationalisation that has to exist in order to be making decisions like they're making now, despite the evidence that makes very clear those decisions are harming vulnerable people," she said.

There were also questions to be asked about how the sum of its potential losses had been calculated, Stonelake added.

Reuters

A recent federal court ruling found social media to be a factor in deteriorating mental health among young people

"Like" counts, for instance, have been part of Meta since its early years, when it was still called Facebook and that was its only platform.

Today, likes are omnipresent on social media platforms. It is the main way in which people engage with text, photos and videos they see online.

Yet, likes are increasingly viewed as a way to foster negative feelings, particularly among young people.

Kaley, a young woman who prevailed in her lawsuit against Meta earlier this year, described during court testimony how she created dozens of accounts on YouTube and Instagram.

She would use the system of accounts to create likes on her own posts, hoping to drive engagement with other users and her own feelings of validation and self-worth. Kaley was only nine years old at the time. She said she remembered feeling depressed, something she was later diagnosed with, aged 10.

Research over the last several years has shown that engagement metrics such as like counts can drive feelings of rejection and depression, external in teenagers.

In the states' lawsuit against Meta, in which the company said it has handed over more than 2 million documents, lawyers pointed to Meta's own research that showed like counts drove "social comparison", or the mental act of marking one's self worth against images of someone else.

That social comparison driven by Instagram was linked to "increased loneliness, worse body image, and negative mood or affect", according to Meta's internal research.

As Judge Biedscheid put it in his order, external, which was the first time a social media company has been deemed a "public nuisance", the way Meta's platforms have operated for over a decade has been part of a growing "youth mental health crisis" in New Mexico and elsewhere.

Now, attorneys from 29 more states will be pushing Judge Gonzalez Rogers to reach the same conclusion.

Tech Life

Tue, 11 Aug 2026 19:30:00 GMT

For years, schools have posted photos of their pupils online. But could they be misused?

Tech Life

Tue, 04 Aug 2026 19:30:00 GMT

Electric power is one of several cleaner alternatives to traditional aviation fuel.

Tech Now

Sat, 01 Aug 2026 01:00:12 GMT

Tech Life

Tue, 28 Jul 2026 20:00:00 GMT

What do we need to know about agentic AI?

Tech Now

Sat, 25 Jul 2026 01:00:00 GMT

Tech Now

Sat, 18 Jul 2026 01:00:00 GMT

Tech Now

Sat, 11 Jul 2026 01:00:00 GMT

BBC News technology

Robot horse and rider steal the spotlight at Chinese conference

Robot horse and rider steal the spotlight at Chinese conference

To play this video you need to enable JavaScript in your browser.

Related topics

TikTok to pay $400m to US in one of largest child privacy settlements

Related topics

More on this story

How landscape gardening is being electrified

Related topics

More Technology of Business

Is Vine back? Short-form video-sharing app Divine opens to public

Related topics

More on this story

UK business hit by 'daylight robbery' 1500% price hike for invoicing software

Who are Bending Spoons?

Get in touch

81,000 warning letters sent to crypto holders in HMRC tax crackdown

Related topics

Monzo says issue with card payments and transfers now resolved

Related topics

OpenAI slows down training after its AI carried out hack

'Unprecedented' cyber-attack

Related topics

More on this story

The critical tech staying safe by going underground

Related topics

More Technology of Business

Meta hooked children on Facebook and Instagram, US court hears

To play this video you need to enable JavaScript in your browser.

Related topics

More on this story

Tech Life

OpenAI makes ChatGPT less 'human' for teens in new safety update

Study Mode and quiet time

Related topics

More on this story

AI to help planes avoid climate-warming 'sky graffiti'

A net climate benefit?

Related topics

US states call for big changes to Instagram and Facebook as Meta child privacy trial begins

Social media turned 'public nuisance'

Related topics

More on this story

Tech Life

Tech Life

Tech Now

Tech Life

Tech Now

Tech Now

Tech Now