BRUSSELS — It takes an 80-year-old German diplomat to put the Jensen Huang spy scandal in proper historical context. 

Asked about allegations that the Nvidia CEO was lured into an intimate relationship with a Chinese spy, Wolfgang Ischinger, Germany’s former ambassador to Washington, pointed to the past. “In World War I, you had female spies who used the honey trap to talk to a general, and maybe sleep with him and extract some secrets,” he told POLITICO.

As Europe’s 27 leaders meet for their first conversation on AI the day after Huang’s resignation, his fall from grace is a reminder that it’s American tech bros — not European military men — who now hold the secrets that can decide the fate of the world.

Friday’s roundtable of European leaders in Dublin must reckon with the reality that the pursuit of AI superiority remains a two-horse race between the U.S. and China — and that Nvidia’s leadership drama doesn’t change the fact that a single American company’s chips will continue to drive the artificial intelligence boom.

With the meeting meant to help forge a new direction for Europe on AI, leaders will mostly be in listening mode, with presentations by Arthur Mensch, CEO of French AI company Mistral, and computer scientist Geoffrey Hinton, considered one of the “godfathers of AI” and an advocate for more stringent safety measures. 

“We’re not going to pretend that we are experts,” an EU official said.

The session comes as Europe struggles to find its place in global AI. The rise of advanced U.S.-built proprietary models and open-weight artificial intelligence progress in China has created a new range of national security issues — and while the EU has proven to be a leader in regulating the technology, its own would-be frontier AI labs continue to play catch-up.

Although this week’s events are a reminder to European politicians that they are spectators in the AI race, an emerging split between top industry leaders in Europe is also dampening momentum on a bloc-wide effort to take on the competition.

“Switching dependency from California to France doesn’t solve Europe’s sovereignty problem, right?” Uljan Sharka, CEO of leading Italian company Domyn, said in an interview.

Sovereignty

Sovereignty and security are the main arguments for Europe not to abandon efforts to develop its own frontier AI models. In the past year, EU officials and companies have struggled to access the latest models from OpenAI and Anthropic, which drive much of the current risks around the technology. 

“How can Europe become a leading and sovereign player in strategic areas of the AI value chain?” European Council President António Costa asked EU leaders in his invitation letter. 

Yet in classic European fashion, officials disagreed ahead of the meeting on how to catch up across many layers of the AI supply chain, from chips and cloud services to cutting-edge models.

“We cannot abandon the frontier ambition. It’s an issue of our sovereignty, it’s an issue of our competitiveness,” Malgorzata Nikowska, head of unit at the European Commission’s Artificial Intelligence Office, said at an event last week. 

It was a moment of clarity in a frantic debate about whether Europe should still chase the frontier in AI, which would require ploughing billions of euros into compute power — or focus on deriving value by applying the technology in key industries. 

Mistral’s Mensch has been a key voice in that debate, warning that giving up the frontier would be “a fundamental mistake” in an essay titled “Build or Surrender.” 

Two European PMs — Finland’s Petteri Orpo and Estonian Kristen Michal — came out swinging ahead of Friday’s meeting, saying “Europe can compete at the frontier” and that the bloc’s ambition “must go beyond adoption of AI,” in a letter exclusively seen by POLITICO.

These statements, and Mensch’s presence at the table Friday, suggest EU leaders will commit to continuing to build competitive frontier models. But how to get there is a far more vexed question — and European industry executives are offering different answers. 

Mistral has taken the route of building large open models. Its latest Large 4 model crossed the 1-trillion-parameter mark, allowing it to rival some of the world’s best open models such as from China’s Moonshot AI. 

To get there, the French company relied on President Emmanuel Macron as its best-placed salesman, and multi-billion euro funding rounds valuing the company at over €21 billion. 

But Mistral’s European rivals don’t necessarily agree with Mistral’s approach of training large models through vast amounts of compute power. 

Sharka — whose company was chosen to lead an EU-backed consortium to develop an open-source AI model that can cover the 24 EU languages by next year — argues Europe should not follow the example of OpenAI and Anthropic to use ever-more compute power for ever-larger models.

Instead, Sharka argued, it should aim to optimize the training runs of the models. “To build the European frontier, we need to be original,” he said. “I think we need to shift away from the concept of copy-pasting like unfortunately some players are pushing.” 

He argued that the typical AI scaling laws, which claim that a model performs better if it is larger and has more compute power, have “plateaued.” 

“There’s only one way, and it’s not the Mistral way to be clear,” he said, adding that there’s a “lot of space for optimization.” 

Mistral did not respond to a request for comment.

The Nvidia problem

Sharka was joined in his assessment by Aldo Podestà, CEO of Swiss AI model provider Giotto.ai., who argued there are ways to be more efficient about “pre-training,” typically the most compute-intensive part of building an AI model.

“We have the opportunity to be at the frontier without having that massive pre-training cost,” Podestà said. “My bet would be to have one very strong common open-source pre-training exercise” — after which different companies can tweak the outcome to their needs.

Under any scenario, Europe will still rely on the U.S. for the hardware components that power its AI models — especially Nvidia’s AI chips — since the continent has little design or production capacity of its own. 

Podestà said the bloc shouldn’t worry merely about building its own models.

“We need to open up a serious discussion about the hardware dependency we have,” he said. “Today it’s a monopoly for a single company.”

ANTALYA, Turkey — “Anything is possible if you believe,” croons Turkish pop megastar Tarkan in his theme song for this year’s global COP31 climate conference.

Attendees appear to have taken his words to heart.

The Antalya summit marks the first climate talks since the United Nations told the world global warming would soon exceed 1.5C, effectively burying the totemic Paris Agreement target that has steered COP negotiations for more than a decade.

Yet leaders here are straining to keep the goal on life support.

“We are on course to overshoot the 1.5 degree target, and every fraction of a degree matters. So we must double down,” said European Commission President Ursula von der Leyen in her speech to fellow leaders on Wednesday.

“Let’s show that the fight for 1.5C is not over,” Hilda Heine, president of the Marshall Islands told leaders.

Even before the talks began, Danish Climate Minister Samira Nawa told POLITICO her country’s priority at COP31 was “to push for keeping the 1.5C target in the heart of global climate ambitions.”

Other leaders and ministers, reviving slogans from COPs past, have exhorted their colleagues to “keep 1.5 alive” or ensure the target remains “within reach.”

But scientists say that dream is over.

The U.N. last month said that, even if countries dramatically slash planet-warming emissions starting right now, the best outcome would be a rise in global temperatures of 1.8 degrees. If they also deploy vast amounts of carbon removal tech to clean up the atmosphere, there’s a chance of getting back to 1.5 degrees after surpassing it — in theory.

In reality, though, greenhouse gas emissions are still rising and the carbon-hoovering technologies needed to achieve that reversal are unproven at scale.

The risk

The collective ambition of national climate plans, already insufficient before the U.N.’s announcement, is even weaker than projected last year. Countries that have been among the strongest proponents of meeting the 1.5-degree target are watering down efforts to achieve it.

But at the same time, many countries (and climate scientists) fear that publicly dumping the rhetoric around 1.5C will open the door to weakening global ambitions even further. Highlighting other possible targets, such as the 1.8C mentioned by the U.N., would be a “slippery slope,” warned one European climate negotiator.

“I do not support the political messaging of keeping 1.5C alive — but on the other hand, I don’t want to lose it,” said Johan Rockström, a leading climate scientist and director of the Potsdam Institute for Climate Impact Research.

“Politicians, when they say that, may naively think that we somehow still are on this linear, incremental journey that can simply take us to safety, which is not the case,” he added. “On the other hand, if we scientifically criticize it, the risk is that they do indeed say: ‘OK, let’s just drop it and aim for 2 degrees instead,’ which would be a big mistake.”

Beyond symbolism

For island nations already losing land and livelihoods to rising oceans, the 1.5 degree target is not merely symbolic. It represents their very survival.

“Our response to the threat of overshoot must be overwhelming, or the world we know is over,” said Heine, the Marshall Islands president. “Overshoot means waves eating away our shores, flooding that destroys towns, schools, hospitals. It destroys our land and ocean.”

The temperature target was not chosen arbitrarily. It represents the level of warming scientists say the planet can endure before the heat trapped in the atmosphere unleashes exponentially greater dangers.

In the Paris accord, countries vowed to limit warming to below 2 degrees and “ideally” 1.5 degrees. The second target has become so dominant that the International Court of Justice last year declared it the primary goal.

A group of developed and highly vulnerable nations, known as the High Ambition Coalition, have called for a credible response to the overshoot as one key COP31 outcome.

“There is no way we can lose that anchor, out of our moral responsibility toward the people who have contributed the least and are the most impacted,” said Christiana Figueres, who headed the U.N. climate body when the Paris Agreement was adopted.

Words vs. action

But promises have not aligned with reality.

“Leaders, and in particular those from developed countries, cannot keep expanding fossil fuel production while claiming to be committed to 1.5 degrees,” U.N. Secretary-General António Guterres told world leaders in his opening speech on Wednesday.

Yet immediately after his speech, leaders did just that.

Australia, COP31 co-hosts with Turkey, is expanding investments in coal and natural gas, even as its Climate and Energy Minister Chris Bowen urges leaders in Antalya to act with “urgency, solidarity and resolve.”

“With every minute that we fail to achieve the targets set out in the Paris Agreement, the risks and costs we face continue to grow,” said Turkish President Recep Tayyip Erdoğan. Two days earlier, he had celebrated the discovery of two new oil fields.

“The science is clear. Breaching 1.5 isn’t where we down tools and give up,” Katie White, the U.K.’s climate minister, said in response to the U.N. findings. White promised COP31 was the place to double down on action — but a day later, her own department approved a major natural gas project in the North Sea.

Meanwhile the United States, the largest polluter in world history, has withdrawn from the Paris Agreement and is no longer part of COP talks. Donald Trump’s abandonment of efforts to address global warming has left a large gap in the world’s collective efforts to cut emissions.

There’s always next year

Tina Stege, the Marshall Islands’ climate envoy, acknowledged the difficulty of the challenge. But she was unbending.

“That’s not a reason to not do everything that we absolutely can,” she told reporters. “Every country that showed up at this talk to do the work needs to do that work and more.”

Doing more means ramping up clean electrification and alternative fuels, say experts.

It also means dramatically curbing short-term climate pollutants, such as methane. That’s emerging as a possible response to the overshoot declaration in the final agreement from this year’s negotiations, according to the European diplomat cited above.

“If we cut methane by nearly half by 2030, we could avoid almost 0.3 degrees of warming by the 2040s and save 250,000 lives every year,” said Mia Mottley, the prime minister of Barbados, on Wednesday. “This buys us the time to get back to the 1.5 degrees.”

The politics are tricky. Even as technological advances have allowed for better detection of the potent pollutant, the U.S. and Europe have weakened regulations aimed at curbing methane emissions.

Mottley was adamant nonetheless that flagging global ambitions were no reason to give up.

“I call for a global movement: 1.5 billion voices for 1.5 degrees,” she said. “We need to do this before we meet at COP32 next year. … Make 1.5 impossible to ignore on the planet Earth.”

MADRID — Trailing in the polls ahead of Nov. 29 national elections, Spain’s ruling Socialist Party is turning to TikTok. Not to use it, but to try to ban it. 

Following the publication on Monday of a Spanish investigation accusing TikTok of allowing child sexual abuse material, the youth ministry announced it would try to block the platform.

On Wednesday the prosecutor’s office opened a preliminary inquiry, prompting a triumphant post on X by Prime Minister Pedro Sánchez. “Great news,” he wrote. “The days of impunity on social media are over. No platform is above the law, and much less above the rights of our children.”

The bombshell report by civil society group Maldita has shocked Spain and given the government a chance to show it’s tough on Big Tech just ahead of a momentous election, with polls showing the right firmly ahead.

Over the past 72 hours the government has pulled every available legal lever against TikTok, including lodging cases with both the Data Protection Agency and the public prosecutor’s office.

The reality that EU legal thresholds for shutting off open platforms are relatively high isn’t slowing the public onslaught. In a TV interview aired Monday after Maldita’s investigation was released, Sánchez cast his government as a bulwark against Big Tech. “If we have to suspend a social media platform in Spain, have no doubt that we’ll do so,” he said.

TikTok says the company moved quickly to identify and remove content after the Maldita report was published. “CSAM (child sexual abuse material) is categorically prohibited on our platform, and we deploy advanced detection technologies alongside a dedicated Child Safety Team to aggressively identify and eliminate it,” a TikTok spokesperson said. “As bad actors continuously adapt their tactics, we work tirelessly to improve our safeguards.”

Sánchez has every reason to turn the conversation toward protecting children and holding foreign tech companies accountable after sagging in the polls following last month’s train crash in Toledo that killed 69 people. The Socialist leader explained this week that his government was pushing to enforce the “criminal liability of the leaders of these platforms.”

Still, the points Sánchez is looking to score by targeting Big Tech may be limited because his main opponent also slammed TikTok for failing to act.

Alberto Núñez Feijóo, leader of the conservative People’s Party and current favorite to be Spain’s next prime minister, met on Tuesday with Maldita’s team. “It’s shocking, and we can’t look the other way,” he wrote on X. “We have a duty to protect children, and this is one of my top priorities for families.” 

Borja Sémper, a key PP lawmaker who attended the meeting between Feijóo and the investigative team, said: “What’s happening on TikTok is unacceptable. We’re not going to let this go.”

The Socialist Party has positioned itself as Spain’s toughest voice against the platforms, and Sánchez has made clashes with “techno-oligarchs” a key campaign calling card. 

A spokesperson for Spain’s digital ministry pointed to the PP and the far-right Vox as partly responsible for the inaction over TikTok, given that they voted against the government’s designation of the National Markets and Competition Commission as the regulator of digital services. By rejecting the motion, the spokesperson said, they deprived the state of “the tools to protect our children.”

Earlier this week, Youth Minister Sira Rego asked the country’s data protection watchdog to suspend TikTok. “If it were up to us, TikTok would have been shut down from the very moment we learned of the investigation,” said a ministry official, who was granted anonymity to discuss the issue. “But we don’t have the authority to do so.” 

The same official said the Ministry of Youth and Children had met with Spain’s TikTok team but hadn’t received a satisfactory response. “We asked them to voluntarily suspend their services in Spain,” the official said, adding the platform had only agreed to “look into it” without providing a timeline.

Bans are hard

The only way for the Spanish government to force a suspension of the service is to get a court order. The ministry has written to the State Attorney General’s Office to open an investigation that could lead to a suspension of operations, the official said. 

In 2024, Spanish judge Santiago Pedraz ordered Telegram to be blocked for three days before reversing his decision, ruling that the measure wasn’t proportionate. The precedent is there, and shows that although a ban is possible, it is also unlikely.

The current Spanish push isn’t the first time a European government facing tough elections has floated suspending an online platform in response to illegal content.

Just last month, the French government touted the suspension of social media amid widespread protests, but a constitutional court partly struck it down, saying it wasn’t in line with French law. Back in 2025 the Albanian government banned TikTok for just over a year, citing concerns over violence among children, although opposition parties accused the administration of trying to stifle criticism ahead of general elections.

But the constitutions of EU countries and bloc-wide rules don’t easily allow such measures. Laws to regulate big online platforms like TikTok are harmonized across the EU, which significantly limits what Madrid can do alone. 

The EU’s privacy rulebook, the General Data Protection Regulation, is enforced by national regulators. Ireland’s data protection authority oversees TikTok, which means the Youth Ministry’s request to Spain’s privacy regulator, the AEPD, is unlikely to go far. 

The AEPD cannot impose a permanent TikTok ban in Spain, but it could trigger what’s known as an “urgency procedure” under the GDPR. 

The emergency measure allows a privacy regulator to impose legally binding measures domestically that can last up to three months, where there is “urgent need to act” to protect people’s privacy rights.

The AEPD said that Ireland’s privacy regulator is the main authority responsible for TikTok’s compliance with privacy rules and said it will “share all available information with the Irish authority” and other European privacy regulators. 

The regulator said it has asked Maldita for a list of links to alleged unlawful processing of minors’ data, and asked TikTok what it has done to take down the reported content. 

TikTok’s wider content moderation practices are regulated by the Digital Services Act, which gives the European Commission oversight powers over big platforms. Brussels already has several probes open against TikTok and has said in preliminary findings that the Chinese-owned platform breaches the rules by failing to protect minors and by designing an addictive service, although it hasn’t issued any fines.

BRUSSELS—The European Union should have a plan to respond to serious artificial intelligence incidents, two EU countries said ahead of a key Friday meeting of national leaders focused on AI.

In a letter seen by POLITICO, Finland and Estonia say the bloc “can still compete at the frontier” of AI by mobilizing capital and building up compute power, but this should go hand in hand with “governing it responsibly.”

The two countries, considered among Europe’s digital leaders, call for “EU-level arrangements for responding to serious AI incidents.”

AI safety and security are on Friday’s agenda of the first-ever European Council meeting devoted to the revolutionary new technology.

Concerns about whether Europe is prepared to respond to AI-powered threats have been picking up steam since a series of incidents this summer in which AI agents went rogue and hacked into other companies. Recent cyberattacks and revelations of coding vulnerabilities have also raised concerns that security systems must be better equipped for the arrival of artificial intelligence.

Hungary pushed for a network of national contact points in case AI model providers lose control of their models, in an October paper seen by POLITICO. The country also pushed for a team of “pre-vetted, security-cleared” independent evaluators who can be ahead of incidents.

The letter from Finland and Estonia called for investment in European frontier AI models “as an economic priority and as a foundation for Europe’s influence on the future of AI.”

The countries warn that Europe requires “independent capacity to evaluate the most advanced models and their safety and security risks.”

New York is preparing to enforce its sweeping AI safety law, requiring major developers like OpenAI and Anthropic to register, disclose safety practices, and report critical incidents or face fines.

BRUSSELS — You know that customer-service bot that won’t let you cancel your subscription? The EU wants to kill it.  

With a new law presented Wednesday, the European Commission will take aim at the endless array of practices websites have used over the past 20 years to make more money from consumers.  

The Digital Fairness Act is the European response to a wider regulatory trend that has seen consumer authorities curb irksome and exploitative practices. These include dynamic pricing that changes ticket prices based on demand, cumbersome processes for canceling subscriptions and the near-complete surrender of customer service to automated systems.  

According to a draft of the DFA obtained by POLITICO, the Commission will address several aspects of the design, service and terms under which businesses interact with consumers online. 

“Europe should not accept a digital economy where profits depend on exploiting people’s vulnerabilities,” Spanish socialist lawmaker Laura Ballarín Cereza, a key European lawmaker on tech legislation, said in a statement.

Industry leaders have pushed back on this potential new rulebook ever since the EU executive began considering a revision of consumer law in 2022. They say current consumer protection and digital services laws already cover these practices. “Where existing rules already cover a practice, it should not be regulated or enforced a second time,” said Ben Brake, director general of tech lobby DOT Europe.  

Consumer groups say existing rules, some dating back to 2005, are not clear enough and that many unfair design practices are still rampant.  

The Commission would have the power to impose fines of up to 6% of global annual revenue for companies that violate the new law. Those penalties are similar to fines levied under the Digital Services Act, the EU’s platform rules meant to stop Big Tech platforms from provoking systemic, society-wide harms through everything from election interference to keeping minors away from violent content. The new law instead would apply to any business offering services to EU consumers online. 

Dark patterns

The DFA would ban deceptive design known as “dark patterns,” which manipulate consumers through the use of such tactics as bright-colored buttons placed prominently to get customers to buy something. It also aims to ban personalized pricing that targets vulnerable consumers, including minors.  

The practice of dynamic pricing is also targeted in the law. That’s when a seller adjusts the price based on conditions such as demand or a user’s location, and it is commonly used in the sale of concert and airline tickets. The proposed law requires transparency when customers are buying products whose prices are subject to change.  

The DFA also covers consumer rights around “agentic commerce,” when artificial intelligence agents are used to find and complete purchases. 

Four lawmakers from the conservative, socialist and liberal groups said the requirement doesn’t go far enough and that the DFA should completely ban dynamic pricing for cultural and sporting events, according to a letter obtained by POLITICO. 

Sellers offering subscriptions will have to provide an easy way to cancel, reminiscent of New York State’s “click to cancel” rules. The new EU rules also require that a human is available for after-sales customer support.  

The rules also address, for the first time, specific aspects of online gaming, including bans on video games that allow kids to pay to succeed and progress in the game.  

The DFA is also trying to make its mark on what is known as “addictive design” of online platforms. This has already been the target of enforcement action in Europe and litigation in the U.S. in cases involving minors. But the new consumer rules would apply to adults as well, requiring online platforms to offer an opt-out from addictive features like autoplay.  

Spain’s Ballarín Cereza thinks this is not enough, saying users should be opted out of these features by default. “Why should consumers carry the burden when companies deliberately design these addictive mechanisms? We must shift from consumer responsibility to corporate responsibility: safety and fairness must be the default, not an option,” she said. 

PARIS — French Minister of Foreign Affairs Jean-Noël Barrot said Iran was behind the recent series of “unprecedented” cyberattacks on refrigerators in several major supermarket chains in the country last month.

“Thanks to the outstanding work of the government departments, I am able to confirm that responsibility for this attack lies with the Iranian Revolutionary Guards,” Barrot told reporters Monday at a G7 meeting in Paris. “We will impose sanctions at European level against those responsible.”

Barrot said that 700 food distribution sites in France were impacted in the cyberattack. French prosecutors said shortly after that they were opening an investigation into the hacks.

Though Barrot was only referring to the incidents reported in France, grocery stores across Europe and even French blood banks have been targeted by hackers in recent days. Cyber experts had previously raised concerns that the attacks may have been part of Russia’s ongoing hybrid warfare against Europe, though the Kremlin has denied any involvement.

In a statement, a spokesperson for France’s foreign affairs ministry said the French investigation found that Iran’s cyber unit, Shahid Kaveh, carried out the destructive attacks. Shahid Kaveh, which is linked to the Islamic Revolutionary Guard Corps, carries out offensive cyber operations.

The French investigation will focus on the unit’s head, the Paris prosecutor’s office said in a statement, adding that about 20 fires had been started because of the cyberattack.

“The attackers took remote control of the equipment via web interfaces to trigger abnormal defrost cycles of unlimited duration, while disabling the security alerts on the sensors that were supposed to activate once a certain temperature was exceeded,” it added.

“We just heard now that [the cyberattacks] were attributed to Iran,” European Council President Kaja Kallas said after Barrot’s statement, adding that the matter would be discussed at next week’s European foreign affairs council. If the attribution is confirmed, she said, “[Europe’s] response should be strong.”

The European Commission is also looking into these cyberattacks, which EU Agriculture and Food Commissioner Christophe Hansen described as a “deliberate attempt to disrupt our food supply chain.”

Iran’s foreign affairs ministry and its embassy in Paris did not immediately respond to a request for comment. 

This story has been updated.

BRUSSELS — At least 15 European flights were targeted by bomb threats Thursday, following seven similar incidents earlier in the week in which authorities found no explosives.

Irish airline Aer Lingus reported that one of its flights from Dublin to Rome was diverted to an airport in France following a security threat.

The office of French Prime Minister Sébastien Lecornu said that the plane landed safely in Évreux, northwest of Paris, and that the 157 people on board were taken to safety.

According to the BBC, two other flights were diverted Thursday following similar alerts: a Wizz Air plane from Milan to Warsaw, which was diverted to Vienna; and a Ryanair flight from London to Sofia, which was diverted to Plovdiv, Bulgaria.

Finnish carrier Finnair said that two of its aircraft — both departing from Helsinki and bound for Rome and Vienna — faced “faced unfounded or hoax bomb threats” and were subject to police checks after landing. The plane bound for Rome was also escorted by Italian fighter jets.

“Unfortunately, air traffic is sometimes subject to threats, as we have seen this week, when also other airlines in Europe have faced hoax bomb threats,” said spokesperson Päivyt Tallqvist.

Another bomb threat affected a flight operated by LOT from Warsaw to Brussels Airport on Thursday morning. Airport spokesperson Ihsane Chioua Lekhli told Belgian newspaper Het Nieuwsblad that police carried out searches and no further security measures were deemed necessary.

LOT said that another aircraft scheduled to fly from Warsaw to Paris “was stopped while taxiing and directed to a designated parking stand” before departure.

“The aircraft and all baggage underwent a security screening,” said the airline, which added that no hazardous items were found and that the flight departed with a replacement aircraft.

Italian news agency ANSA reported that two flights bound for Milan Malpensa Airport — an Aegean Airlines flight from Thessaloniki and a Ryanair flight from Porto — were also searched following bomb threats.

Two other Ryanair flights bound for Sofia — one from Copenhagen and the other from Madrid — were also reportedly targeted by bomb threats.

Czech police said on social media that four flights bound for Prague Airport were also threatened, but officers assessed “the severity of the threats as low.”

Ben Smith, CEO of Air France–KLM, said that the threats were happening “in a coordinated way.”

“I don’t think we’ve seen this before,” he said during a Thursday earnings call, but added that the airline group is not making any requests of governments. “We’re quite confident that we have in place what we need to ensure that it has minimal or no impact on our operations.”

EU Transport Commissioner Apostolos Tzitzikostas said: “It is too early to draw conclusions on whether the reported incidents are linked.”

He added that the EU-level information sharing mechanism for aviation security has been activated. “We don’t have information yet whether this is a cyber attack,” he said.

EU Aviation Safety Agency spokesperson Janet Northcote said: “The immediate safety threat from individual hoax threats is low. However, the higher number of diversions places an additional strain on the system.”

Seven passenger aircraft were searched on Monday and Tuesday by authorities in Spain, Poland, Italy and Bulgaria following bomb threats. No explosives were found, but the alerts also led to the temporary closure of Warsaw Chopin Airport.

The risk of someone successfully smuggling a bomb aboard an airplane is low, said Jan-Arwed Richter, CEO of JACDEC, an aviation-safety analysis firm.

“Today there is hardly any thinkable way to get explosives aboard a commercial flight,” he said. “The last real bombing of a flight took place on Oct. 31, 2015 on a flight from Egypt to Russia because of security loopholes and personnel shortcomings that were rectified immediately.”

This article has been updated.

BRUSSELS — Prince Harry and Meghan Markle will visit the European Parliament next week to discuss online child safety, according to an email sent to lawmakers and seen by POLITICO.

Visiting Brussels for the first time as a couple, the Duke and Duchess of Sussex will on Tuesday join Parliament President Roberta Metsola in opening an exhibition honoring “the memory of children whose lives were lost in connection with online harms.”

Prince Harry will then discuss with Commissioner Dubravka Šuica, experts and lawmakers “how we can turn legislative commitments into effective safeguards for children, including through safer design, stronger enforcement, and clear accountability for technology companies,” the email reads.

The event is being hosted by Spanish center-right lawmaker Antonio López-Istúriz White, who is the vice president of the Parliament’s intergroup on children’s rights. In September, the European Commission proposed the EU Kids Act, which includes restrictions on social media use for minors.

“The decisions we take on AI, platform design and the enforcement of online safeguards will shape the digital environments in which children grow up. This responsibility cuts across parliamentary committees and policy areas; the experiences of children and families should inform and guide our work throughout,” Lopéz wrote in his invitation to other lawmakers.

OpenAI researchers tried to keep AI agents from communicating with each other by putting them in their own environments on the same computer, not connected to a network or to each other. But they figured out how to communicate with each other through a shared resource, the CPU cache. — Read the rest

The post OpenAI's isolated AI agents talked to each other through the CPU cache appeared first on Boing Boing.

A Crunchwrap Supreme and a bag of supermarket salad generally appeal to rather different notions of a good lunch. One offers the architectural pleasure of a crunchy tostada tucked inside a warm tortilla; the other, the satisfaction of having arranged to eat vegetables without chopping any. This summer, however, both came with a less welcome consideration: iceberg lettuce.

A massive cyclospora outbreak linked to shredded iceberg from Taylor Farms de Mexico prompted recalls extending from Taco Bell kitchens to supermarket produce aisles. Shoppers began trying to determine which seemingly unrelated groceries shared a supplier; someone even made a barcode-scanning app for precisely this purpose. (Such was the broader atmosphere of suspicion that an entirely unsupported rumor about the Trump administration’s promised bargain beef being horse meat eventually inspired online guides for distinguishing one animal from the other.)

It was, in short, Bad Lettuce Summer.

And according to a Gallup poll published October 8, Americans emerged from it considerably more cautious about what they eat. Nearly three-quarters now say they have avoided buying certain foods because of recalls or government advisories, up from 53% in 2024. This newfound vigilance crosses party lines: 81% of Democrats and 63% of Republicans report having taken such precautions.

What it did not produce was much agreement about the people responsible for keeping the lettuce safe.

In 2024, when Joe Biden occupied the White House, 74% of Democrats and 50% of Republicans expressed confidence in federal food-safety oversight. Two years and one change of administration later, those figures stand at 30% and 70%, respectively. 

A 24-point Democratic advantage has become a 40-point Republican one, a remarkable 64-point reversal.

Ahead of the midterms, the findings illustrate a curious feature of American political life: People can encounter the same dangers, take similar precautions and nevertheless arrive at radically different conclusions about the institutions responsible for protecting them. And if there were ever a summer when one might have expected a little consensus about the importance of competent food-safety oversight, surely it was this one.

The outbreak, the largest recorded cyclospora outbreak in U.S. history, sickened nearly 13,000 people across 21 states, hospitalizing 570 and killing two. The recall included lettuce served at Taco Bell and Marketside iceberg salad sold at Walmart. Taylor Farms maintained that its own branded salad kits were unaffected, a distinction that mattered scientifically, if not necessarily psychologically, to shoppers eyeing the refrigerated wall of greens.

The trouble with becoming an amateur food-safety inspector is that the job is never quite finished. You can establish that the blueberries in your refrigerator haven’t been recalled, but this tells you remarkably little about the blueberries you’ll purchase next Thursday. You can discover who owns the company that packaged your broccoli, though what to do with that information is another matter. Every new discovery yields a little more knowledge, if not necessarily a corresponding increase in certainty.

---

Want more great food writing and recipes? Sign up for Salon’s free food newsletter, The Bite.

---

There was a policy dimension to this uncertainty, too. In March 2025, the FDA announced plans to delay compliance with its Food Traceability Rule, which requires businesses handling higher-risk foods, including leafy greens, to maintain records that help investigators identify and remove contaminated products more quickly. Originally scheduled for January 2026, the compliance deadline was proposed to be extended to July 2028; Congress subsequently directed the FDA not to enforce the rule before then. The agency said businesses needed additional time to coordinate their systems.

The delay did not cause the summer’s outbreak. Still, there is something peculiar about postponing industry-wide traceability requirements while ordinary shoppers are left scanning their broccoli with smartphones, hoping to learn something useful about where it came from.

During Bad Lettuce Summer, even the official answers required some translation.

In July, the FDA announced that a lettuce sample had tested positive for cyclospora, then corrected the finding as a false positive. Taylor Farms emphasized the correction; federal investigators maintained that their broader epidemiological and traceback evidence still implicated the company’s shredded iceberg. Subsequent environmental testing strengthened that evidence, without conclusively establishing how the contamination occurred.

One could read the competing statements carefully and emerge with some more information, but not much more confidence about dinner.

Then came the political subplot. As the investigation intensified, Taylor Farms representatives met privately with White House officials. The company and its leadership had donated substantial sums to Republican-aligned political groups, including $1 million to the pro-Trump super PAC MAGA Inc. Rep. Robert Garcia, the ranking Democrat on the House Oversight Committee, subsequently sought answers about whether the company had attempted to influence the investigation. The White House denied compromising public safety, and no evidence established that political access changed the FDA’s findings.

Still, it is an extraordinary distance to travel from wondering whether the lettuce in your taco is safe to wondering who has the ear of the White House.

The Gallup numbers do not establish why Americans’ confidence has diverged so dramatically. Someone may distrust the industrial food system while enthusiastically supporting an administration promising to reform it. Someone else may have specific objections to that administration’s regulatory decisions. But the speed of the partisan reversal suggests that who runs Washington matters enormously to how Americans evaluate its institutions — though there is a certain irony to this in the age of Make America Healthy Again, which has elevated suspicion of the conventional food system into something approaching a civic virtue.

Its influence has extended all the way to the fast-food fryer: Steak ‘n Shake announced that it had “RFK’d” its French fries by replacing vegetable oil with beef tallow, a development Health and Human Services Secretary Robert F. Kennedy Jr. celebrated over burgers and fries during a televised appearance with Sean Hannity. The nation’s health secretary had become a verb, and the French fry a political statement.

It is, admittedly, easier to stage a photo opportunity around a basket of fries than a database of food-traceability records. Both involve questions about what Americans eat, but investigating foodborne illness requires a rather different kind of governmental attention.

Of course, Americans did not suddenly begin losing confidence in food-safety oversight under Trump. 

Gallup found that overall confidence has fallen from 85% in 2004 to just 47% today. And recalls themselves contain an uncomfortable ambiguity: They reveal that something dangerous entered the food supply, but also that someone detected a problem and warned the public. The same event can look like evidence of a system working or a system failing, depending on which part of the story one chooses to emphasize.

With the midterms approaching, that distinction matters. The lawmakers elected in November will have a say in how federal food-safety agencies are funded, scrutinized and held accountable when the next outbreak arrives. Yet Americans are approaching that election with increasingly different assessments of the institutions charged with protecting them, even as their own precautions suggest a broadly shared sense of vulnerability.

The next outbreak, unfortunately, will not wait for everyone to agree on whom to trust.

The post Food safety fears are bipartisan. Trust in oversight isn’t appeared first on Salon.com.