TheChek News Flow

Artificial intelligence group has added fingerprint scans and hired military experts to protect important data

OpenAI has overhauled its security operations to protect its intellectual property from corporate espionage, following claims of having been targeted by Chinese rivals.

The changes in recent months include stricter controls on sensitive information and enhanced vetting of staff, according to several people close to the $300bn artificial intelligence company.

The San Francisco-based start-up has been bolstering its security efforts since last year, but the clampdown was accelerated after Chinese AI start-up DeepSeek released a rival model in January.

OpenAI claimed that DeepSeek had improperly copied the California-based company’s models, using a technique known as “distillation”, to release a rival AI system. It has since added security measures to guard against these tactics.

DeepSeek has not commented on the claims.

The episode “prompted OpenAI to be much more rigorous”, said one person close to its security team, who added that the company, led by Sam Altman, had been “aggressively” expanding its security personnel and practices, including cyber security teams.

A global AI arms race has led to greater concerns about attempts to steal the technology, which could threaten economic and national security. US authorities warned tech start-ups last year that foreign adversaries, including China, had increased efforts to acquire their sensitive data.

OpenAI insiders said the start-up had been implementing stricter policies in its San Francisco offices since last summer to restrict staff access to crucial information about technologies such as its algorithms and new products.

The policies — known as information “tenting” — significantly reduced the number of people who could access the novel algorithms being developed, insiders said.

For example, when OpenAI was developing its new o1 model last year, codenamed “Strawberry” internally, staff working on the project were told to check that other employees were also part of the “Strawberry tent” before discussing it in communal office spaces.

The strict approach made work difficult for some staff. “It got very tight — you either had everything or nothing,” one person said. They added that over time “more people are being read in on the things they need to be, without being read in on others”.

The company now keeps a lot of its proprietary technology in isolated environments, meaning computer systems are kept offline and separate from other networks, according to people familiar with the practices. It also had biometric checks in its offices, where individuals could only access certain rooms by scanning their fingerprints, they added.

In order to protect model weights — parameters that influence how a model responds to prompts — OpenAI adopts a “deny-by-default egress policy”, meaning nothing is allowed to connect to the internet unless explicitly approved.

OpenAI had also increased physical security at its data centres, the people said. It was one of a number of Silicon Valley companies that stepped up their screening of staff and potential recruits because of an increased threat of Chinese espionage, the Financial Times reported last year.

Washington and Beijing are locked in a growing strategic competition, with the US imposing export controls to make it harder for China to obtain and develop cutting-edge technologies. However, concerns have also been raised about a rise in xenophobia at US tech companies given the prevalence of skilled workers of Asian descent.  

OpenAI hired Dane Stuckey last October as its new chief information security officer from the same role at Palantir, the data intelligence group known for its extensive military and government work.

Stuckey works alongside Matt Knight, OpenAI’s vice-president of security products. Knight has been developing ways to use OpenAI’s large language models to improve its defences against cyber attacks, according to a person with knowledge of the matter.

Retired US army general Paul Nakasone was appointed to OpenAI’s board last year to help oversee its defences against cyber security threats.

OpenAI said it was investing heavily in its security privacy programs, as it wants to lead the industry. The changes were not made in response to any particular incident, it added.

Why driverless vehicles just can’t quit humans

Regulators need to ask more questions about the people in the shadows

“There’s nobody in the truck,” Sterling Anderson, co-founder of autonomous truck company Aurora, said in a podcast interview last year. “We’re not Wizard of Oz-ing this thing.” Anderson was referring to the company’s plans to begin a commercial delivery service using driverless trucks between Dallas and Houston in Texas. What he meant, I think, was this: our technology is not a parlour trick. Unlike in the Wizard of Oz, there won’t be a human hidden behind the curtain.

In May this year, Aurora announced its commercial driverless trucking service had officially begun. But a few weeks later, the company made another announcement: its truck manufacturing partner PACCAR “requested we have a person in the driver’s seat, because of certain prototype parts in their base vehicle platform” and “after much consideration, we respected their request and are moving the observer, who had been riding in the back of some of our trips, from the back seat to the front seat”.

Aurora insisted this wasn’t necessary to operate the truck safely, and that the observer would not operate the vehicle. Still, it was clearly a blow to its ambition to have “nobody in the truck”. Aside from the fear that it might look like they are “Wizard of Oz-ing this thing”, the investment case for driverless trucks doesn’t look so good if you need to pay someone to sit in each one.

Aurora isn’t the only autonomous vehicle company that hasn’t quite been able to quit humans. When Tesla launched its robotaxi service last month in Austin, Texas, the cars had human “safety monitors” in the passenger seats. Even the more established self-driving taxi services, which don’t have anyone inside the car, still have humans behind the scenes. In China, Baidu’s robotaxis launched with “remote human operators” who could take control of the cars if necessary. Waymo, in contrast, doesn’t have “remote drivers”, but it does have “human fleet response agents”. Confused Waymos remain in control but can ask these humans for advice.

If humans are such poor drivers (as many self-driving car companies allege), why can’t supposedly superior machines cope without them? Because machines and humans are good at different things. Machines don’t get tired, bored, drunk or distracted, but they struggle with real-world “edge cases” that require contextual awareness and intuition, such as how best to navigate a blockage on the road, or what a construction worker waving his arms around is trying to tell you. On top of that, every safe system should have a backstop in case of technical problems.

In that sense, the autonomous vehicle companies should be applauded for keeping humans around. It doesn’t mean they’re trying to pull off a parlour trick. But it does mean we should know much more about how these human roles actually work.

That’s because systems that rely on a combination of machines and humans can suffer from all sorts of well-documented problems. Humans asked to be safety monitors might suffer from “automation complacency”, which is the human tendency to lose concentration while supervising autonomous systems. Remote drivers might struggle with technical issues like connection problems and poor latency. Then there are questions of liability: if a human in a support centre somewhere gives bad advice to an autonomous vehicle that leads to an accident, who is to blame? The technology company? The employer? The individual? What if that human isn’t in the same state, or even the same country?

Bryant Walker Smith, an associate professor of law at the University of South Carolina, told me the onus should be on the self-driving companies to explain exactly what they’re doing and why they think it is safe. “Regulators absolutely should interrogate every piece of that.”

Yet so far, many of these human roles have remained in the shadows. When I sent a list of basic questions to Waymo, Tesla and Aurora, only Waymo responded. The company declined to say how many people worked in its fleet response team, but it did say they were employed by Cognizant, an IT company, that they required drivers’ licences and that they were “seated in Arizona, Michigan, and in an offshore location”. When I asked about lines of accountability, the company said that “to the extent a Waymo vehicle was involved in a collision that caused property damage or injury, Waymo would be responsible for the liability imposed on it by law”.

It shouldn’t be viewed as a problem that self-driving cars still need support from humans behind the scenes. But nor should those roles be hidden away. It’s time for regulators to pull back the curtain.

Brussels reworks plans to protect direct income for farmers in new budget after pressure from industry

Brussels is planning to ringfence hundreds of billions of euros of direct subsidies to farmers despite efforts to redirect its budget towards other objectives, such as boosting Europe’s defence and economic competitiveness.

The European Commission plans to merge several funding streams into a single envelope per country in its multiannual budget due in mid-July, giving capitals more leeway on how to spend the money. But most agricultural subsidies will be handed over in a separate, protected package, according to three officials with knowledge of the plans.

Under pressure from the EU’s powerful farming lobby, the commission has acknowledged that many small farms would go under if they lost the direct income support paid under the EU’s Common Agricultural Policy (CAP), the officials said.

“Farmers must have their direct payments,” one EU official said.

The CAP makes up around one-third of total EU spending in the current budget cycle. It is one of the EU’s oldest policies, first established in 1962, and was designed to protect the bloc’s food security and ensure farmers were paid a fair wage.

EU officials have said that overall the CAP budget is likely to fall compared with the current €386bn fund, over the seven-year period.

Income support for farmers currently amounts to €291bn of the fund, or around three-quarters.

The CAP has been mired in controversy over the amount of money going to large landowners. Farmers took to the streets in protest last year, burning hay and blocking roads with tractors — in part over the complex bureaucracy involved in accessing the funds.

The remaining quarter of the CAP is currently paid to farmers if they fulfil certain environmental criteria, such as preserving hedgerows and allowing land to lie fallow. After the protests, however, the commission cut back on the requirements that farmers must fulfil in order to access the funds.

Environmental groups have already complained that the commission’s efforts to simplify the CAP have weakened protections for natural ecosystems.

Célia Nyssens-James, policy manager for agriculture and food systems at the European Environmental Bureau, said that only protecting the CAP direct payments was “not acceptable” and that they were “broken subsidies” in need of a “genuine overhaul”.

The much-anticipated plan for the new EU budget, which will run from 2028 and is mostly funded by member state contributions, has caused deep divisions in the EU executive, as commissioners fight to preserve pots of money for the departments they are responsible for.

The proposal had been to combine funding streams, such as regional funds, social spending and agricultural subsidies, into one payment dubbed “national and regional partnerships”.

There has also been pushback from current beneficiaries, including farmers and regions that fear their funding streams will be lost in favour of new priorities like defence spending.

“The proposal to centralise EU funding into a single fund by the European Commission risks dissolving the CAP into a broader framework with less focus, fewer guarantees and no shared vision,” said a petition signed by more than 3,100 farming groups, including dairy multinational Arla Foods and the European Landowners Organisation.

The commission’s Romanian executive vice-president Roxana Mînzatu has, for example, been pressing to maintain social spending. “The next budget needs to have a strong social dimension, as strong as the competitiveness and security dimension,” she told the Financial Times.

A majority of EU countries, alongside regional and local governments, have called to preserve regional development funds — or cohesion policy — which make up over one-third of the current EU joint budget.

“Only a distinct and robust budget and a region-based allocation methodology” can ensure that the next budget “will deliver long-term unity, competitiveness and convergence across EU regions”, wrote 14 countries including Italy, Spain and Poland.

The commission declined to comment on the budget plans.

EU pushes to fill Ukraine’s $19bn budget gap next year

Brussels explores options including frontloading G7 loans to ease fiscal strains in Kyiv

Brussels is urgently exploring ways to cover a shortfall of up to $19bn in Ukraine’s budget next year, as Kyiv grapples with diminishing US support and receding prospects for a ceasefire with Russia.

The European Commission is discussing options with EU member countries, including channelling military support to Ukraine as off-budget grants, frontloading loans from an existing $50bn G7 support scheme for Kyiv, and further leveraging Russian state assets immobilised in the EU, according to multiple people familiar with the discussions.

Ukraine’s projected budget gap next year is not yet covered by external financing.

“There’s growing concern about next year and many stakeholders that were banking on a ceasefire deal this year [to ease Ukraine’s fiscal strains] are having to recalculate their outlays and realising that there’s a [financing] hole whichever way they to try to slice it,” said a senior EU official involved in discussions with Kyiv.

The commission has already had to adjust outlays from Ukraine-related funding streams during 2025, officials told the Financial Times, in response to the extended conflict and lack of confidence in an imminent ceasefire with Moscow.

The urgency in Brussels to rustle up fresh funding comes ahead of a summit focused on funding Ukraine’s reconstruction needs in Rome later this week, which European Commission president Ursula von der Leyen will attend.

The IMF estimates that Ukraine’s financing needs for next year are covered, but that is premised on the war ending this year or in the middle of 2026 — a scenario which Ukraine and the EU do not share.

IMF managing director Kristalina Georgieva said last month that the Fund “will assess whether this financing gap is going to increase and will require further financing on the external aid”.

The goal is to ensure that Kyiv’s needs are covered well in advance of the winter, especially given the uncertain prospects for further US military support, said one EU diplomat.

One proposal, shared by Kyiv with G7 countries and under consideration by the European Commission, is to channel military support to Ukraine as bilateral grants that would be accounted for separately as an “off-budget external transfer”, while at the same time counting towards national defence spending targets.

This would serve the double goal of contributing to Nato pledges to increase national defence spending to 5 per cent of GDP while providing support to Ukraine. “Instead of duplicating capabilities, European allies could co-finance Ukrainian forces — treating it as a service Ukraine provides to enhance continental security,” Kyiv wrote in a paper shared with G7 allies and seen by the Financial Times.

The commission was set to discuss this and other options with EU finance ministers on Monday evening, two people familiar with the matter said.

“Clearly the military support for Ukraine that member states are giving are not only funds for the defence of Ukraine but for the defence of Europe, and some of that of course will count as defence spending,” said one senior EU diplomat.

Another option is to anticipate disbursements from an existing $50bn G7 scheme that issues loans to Kyiv on the back of profits from Russian state assets immobilised in the West.

Without a ceasefire to boost Ukraine’s domestic economy, Kyiv expects a shortfall of at least $8bn for 2026 even if some of the pledged amounts can be brought forward, from partners including the EU, Japan and US. If that does not happen, then the financing gap could reach $19bn.

A further option could be to extract more value from the Russian immobilised assets, by reinvesting them in riskier asset classes — while devising a way to share responsibility for potentially greater financial losses that would not make Belgium, where most of the assets are held, solely liable for them.

“We are exploring those options including the possibility to further leverage the use of Russia’s immobilised assets,” EU economy commissioner Valdis Dombrovskis said last month.

One year on from Labour lifting a de facto ban, few large applications have been made

A surge in onshore wind turbines in England has yet to materialise one year on from Labour’s lifting of a de facto ban on new projects. 

Planning authorities have not received any applications for new large scale projects since restrictions were eased last July, according to Financial Times analysis of government data. 

The fourteen applications submitted between the lifting of the ban and April this year are all for small projects involving one turbine, amounting to a total combined capacity of around nine megawatts — enough for around 9,000 homes. 

Twelve projects totalling 600MW are at an earlier stage of development in England and have yet to submit planning applications, according to data from industry group Renewable UK.

That compares with the government’s aim to almost double Britain’s onshore wind capacity by 2030, from around 16 gigawatts to around 27-29 gigawatts, suggesting the bulk of that will need to come from Scotland and Wales.

Developers need to go through lengthy processes before they can apply for planning permission, such as identifying sites and gathering environmental data.

“It can easily be 2-5 years before you are in a position to submit [an application for] planning consent, where it will show up on the database,” said Lisa Christie, director of public and regulatory affairs in the UK for Swedish developer Vattenfall and a member of the UK’s onshore wind task force.

“We certainly see [England] as an exciting market and we are looking at quite a lot of sites, including in the south of England.”

However, the slow progress so far highlights the scale of the challenge as the government tries to develop turbines in England to boost overall clean power supplies and balance out the distribution of turbines across Britain.

Development in recent years has mostly been in Scotland, after David Cameron’s Conservative government put in place highly restrictive planning rules in England in 2015 following opposition from rural communities.

In 2023, Michael Gove, then communities secretary, eased these rules but legal experts said at the time it did not go far enough to encourage developers’ interest. 

Planning applications for Scotland, Wales and Northern Ireland show 68 were submitted since Labour took power last July, with an additional 600 filed during England’s de facto ban.

In total there are 528 onshore wind schemes in the pipeline in Scotland, Wales and Northern Ireland — either under construction or awaiting a planning decision — with a total planned capacity of 17.6GW.

Labour went much further days after it won the general election last July, removing rules which meant any local opposition was enough to block a project. 

It has since taken further steps to improve the chances of onshore wind projects getting through the planning process in England, and last week set out plans to try to tackle problems such as relationships with local communities and concerns about interference with military and aviation radar. 

Responding to the planning application figures, a spokesperson for the Department of Energy said it had laid “strong foundations to reverse a near decade of decline in the industry” but it would “take time for these changes to filter through the planning system”.

RenewableUK’s head of onshore wind delivery James Robottom said: “The de facto ban on onshore wind was in place for nine years, so naturally it takes time to develop new projects from a standing start.”

From satellite navigation to spy probes, anti-sat weapons and reusable space planes, Chinese technology is advancing at a fast pace

The new space race is heating up, with the United States warning that its major rival China is narrowing the gap as it makes rapid technology gains.

While China says its ambitious plans remain peaceful and that it rejects the weaponisation of space, some of the technologies it has developed in recent years also have military uses. Here are some of them.

The Chinese navigation satellite system provides positioning, navigation and timing services worldwide.

Its network of 60 satellites has been in full global operation since 2020, with the final backup satellites launched in 2024.

BeiDou – a symbol of China’s growing tech self-sufficiency – aims to challenge the dominance of GPS, the global positioning system run by the US military, especially in Belt and Road Initiative countries.

Besides its civilian applications, BeiDou provides navigation and positioning services to all branches of the Chinese military with even higher precision in the Asia-Pacific than GPS offers the US military.

That enables independent guidance for Chinese missiles, bombs and other precision munitions, with high accuracy. The People’s Liberation Army can also use BeiDou’s short messaging function to communicate.

Besides its civilian applications, BeiDou provides navigation and positioning services to all branches of the Chinese military with even higher precision in the Asia-Pacific than GPS offers the US military.

That enables independent guidance for Chinese missiles, bombs and other precision munitions, with high accuracy. The People’s Liberation Army can also use BeiDou’s short messaging function to communicate.

At least two low-Earth orbit communication satellite mega-constellations are currently in development in China as alternatives to SpaceX’s Starlink.

The Qianfan, or Thousand Sails, constellation – which is also known as the G60 Starlink – is a 14,000-satellite broadband network being developed by Shanghai Spacecom Satellite Technology. About 90 satellites have been launched since last year and the target date for completion is 2030.

Meanwhile the Guowang, or National Network, being developed by China Satellite Network Group aims to have two sub-constellations with a total of 13,000 satellites. It began at the end of 2024 and as of June, 34 satellites had been launched into low-Earth orbit.

Just as Starlink has provided Ukrainian forces with a satellite internet service during the war with Russia, these Chinese networks could give China’s military resilient connectivity during a conflict.

Their high-bandwidth and low-latency connections could enhance command and control, situational awareness and joint all-domain operations – especially when it comes to supporting unmanned equipment in the era of drone warfare.

China has developed a number of satellites with what is known as ISR – or intelligence, surveillance and reconnaissance – capabilities.

They include the remote-sensing Yaogan satellites that have optical reconnaissance, synthetic-aperture radar and electronic intelligence sensors and are said to be primarily for military reconnaissance use.

Nasa describes the experimental communication technology network Tongxin Jishu Shiyan, or TJS – which is in geostationary orbit – as military satellites that provide early warning and signals intelligence for the PLA.

Another example is the Gaofen, or High Resolution, satellites that use optical, multispectral, radar and radio frequency for Earth observation. Officially they are for agricultural, disaster, resource and environmental monitoring, but given their capabilities they are also believed to have military uses too.

The US Space Force has claimed that, by the end of last year, more than 510 of the 1,060 Chinese satellites in orbit were capable of ISR and could be used by the PLA to monitor US and allied forces in the Pacific region.

China is also developing an unmanned reusable spacecraft – similar to the American X-37B Orbital Test Vehicle – that can orbit the Earth for hundreds of days then return. The name of this secretive space plane is still classified but many believe it could be the Shenlong, or Divine Dragon – a project listed under the country’s hi-tech development plan since the 1980s.

The reusable experimental spacecraft has already had three known missions. It made its maiden flight in 2020, when it was delivered into orbit by a launch vehicle, flew for two days and landed in western Xinjiang province.

It spent nine months in orbit during a second mission in 2022, then more than eight months in orbit on a third mission from 2023. Each time, unidentified objects were released before its re-entry.

The spacecraft could be used as an experimental platform for space technologies, including power systems and the durable materials needed for hypersonics, and it could also be suitable for surveillance.

It could also have other military uses. Since it would be able to deploy payloads, it could be used to launch small satellites or release destructive weapons targeted at other satellites or even the ground. And if it can retrieve payloads like the X-37B, it could also be used to capture adversarial space assets.

China is developing another reusable space plane called the Tengyun, or Cloud Rider, designed to separate in two at an altitude of 30-40km (18-24 miles), with the first stage returning to the ground while the second stage continues its ascent to low-Earth orbit. Its first flight is planned for 2030.

China’s Shijian and TJS satellites have been observed performing “unusual, large and rapid” orbital manoeuvres that US officials have described as “stalking” – closely approaching, tracking and monitoring American satellites apparently to collect intelligence or potentially disrupt operations.

Among those targeted were US Geosynchronous Space Situational Awareness Programme satellites and Silent Barker observation satellites, according to the US Space Force.

It said earlier this year that three Shiyan-24C and two Shijian-6 05A/B satellites had carried out synchronised controlled moves in low-Earth orbit in what one Space Force official called “dogfighting in space”.

These manoeuvres – rendezvous and proximity operations, docking and capture – are needed for peaceful missions such as removing orbital debris, but they could also be used in attacks against high-value military satellites.

Dealing with space debris – from monitoring to early warning and emergency response – has been listed as a key task by Beijing. The China National Space Administration has sent several satellites, mainly Shijian, to experiment with orbital maintenance and cleaning up space debris.

Several methods have been tested to clear space junk, including scooping it up with robotic arms and nets, and obliterating it with lasers.

But these technologies could also be used for anti-satellite warfare, to take down probes that are critical for positioning, navigation and timing, for command, control and communications, as well as missile early warning and other vital military functions.

The US Space Force raised concern over the Shijian-17 satellite launched in 2016, which it said had made some “unusual manoeuvres” in geostationary orbit and varied its position in relation to other satellites. The Space Force also noted that its robotic arm could be used “for grappling other satellites”.

Advertisement

These concerns escalated when the Shijian-21, launched in 2021, towed a defunct BeiDou navigation satellite from geostationary orbit to a higher “graveyard” orbit in 2022.

The Shijian-21 used up much of its propellant during that operation, and China now appears to be about to attempt the first satellite-to-satellite refuelling and servicing mission in high orbit – closely watched by the US.

This year, the Shijian-25 satellite was launched for the mission, which would mark major technological progress and the potential to significantly extend the operational lifespan of space assets.

Anti-satellite weapons

China conducted its first successful direct ascent anti-satellite missile test in 2007. A land-based ballistic missile directly intercepted and destroyed a defunct Fengyun-1C weather satellite in low-Earth orbit, generating a large amount of debris and drawing international criticism.

In a 2013 test, a ballistic object was launched to an altitude of 30,000km, suggesting it could be capable of shooting down satellites in geostationary orbit (35,786km).

Ground-based laser weapons and powerful orbital- and ground-based jamming are among other anti-satellite capabilities the PLA is pursuing.