EU Gains Access to Anthropic’s Mythos AI Amid US Security Restrictions
After weeks of negotiations, Anthropic offered the European Union access to its advanced AI model Mythos, which can identify and exploit cybersecurity vulnerabilities. The EU had been blocked by the US government, which ordered Anthropic to restrict non-US access to maintain AI leadership over China. The European Commission welcomed the development, while the EU prepares a formal action plan on AI cybersecurity. The incident has intensified European calls for technological sovereignty and reduced reliance on US tech.
Cross-source coverage
Wire timeline
EU to Establish AI Evaluation Capacity Amid Concerns Over Powerful Model Access Restrictions
The European Commission announced plans to create an 'EU evaluation capacity' for advanced AI models, responding to cybersecurity concerns triggered by Anthropic's Mythos model and subsequent US government access restrictions. Tech Commissioner Henna Virkkunen unveiled the action plan on AI and cybersecurity, which includes setting up a dedicated testing capability expected to be operational by 2027. The capacity will provide AI model providers with a European option for third-party evaluations required under the EU AI Act and support the AI Office in enforcement. The plan also addresses potential future access restrictions, with the Commission and ENISA developing a 'blueprint' for safe model access and unspecified 'contingency measures' if companies or governments withdraw access. The EU previously pushed for ENISA to evaluate cyber models, and the new capacity may overlap with ENISA's responsibilities. The move follows fears that the EU could be cut off from critical cybersecurity AI tools.
EuractivAustria Urges Europe to Host Anthropic After U.S. Curbs on AI Access
Austria has proposed that the European Union host Anthropic within its borders to counter U.S. restrictions blocking foreigners from using advanced AI models. In a letter to EU Technology Commissioner Henna Virkkunen, State Secretary for Digitalization Alexander Proell emphasized Europe must not be cut off from major innovations. He suggested exploring strategic establishment and participation of Anthropic in the EU, offering legal certainty, market access, capital, and aligned values. Proell acknowledged skepticism but argued Europe must be an architect of its technological future, not an administrator of decisions made elsewhere. Anthropic did not comment. The proposal follows the European Commission's recent legislative push to boost domestic cloud, AI, and semiconductor industries and reduce reliance on U.S. Big Tech, despite U.S. criticism.
The Hindu: Latest News today from India and the World, Breaking news, Top Headlines and Trending News Videos.Trump Withholds AI Model from Allies, Unsettling Europe
On June 29, 2026, US President Donald Trump ordered Anthropic, the AI company behind the Claude model, to withdraw its stripped-down version 'Fable 5' (part of the Claude Mythos line) from public availability. This decision has unsettled Europe, as even allied NATO states are reportedly receiving only limited access to the AI model. Capital market strategist Stefan Risse analyzes the implications, suggesting the move is a clear signal of US policy prioritizing national security and potentially leveraging AI as a strategic asset. The withdrawal comes amid concerns over cyberattacks, highlighting the growing geopolitical tensions around advanced AI technology. The article, published by German news outlet Die Welt, frames the action as a significant shift in US-ally relations, raising questions about technology access and trust between the US and its European partners.
Nachrichten - WELTAustria Urges Europe to Host Anthropic to Counter U.S. AI Access Curbs
Austria has proposed that the European Union consider hosting the AI company Anthropic within its borders, aiming to counter U.S. restrictions blocking foreigners from accessing its most advanced AI models. In a letter to EU Technology Commissioner Henna Virkkunen, Austria's State Secretary for Digitalization Alexander Proell emphasized the importance of Europe not being cut off from major innovations. Proell called for a joint exploration of the 'strategic establishment and participation' of Anthropic in the EU, citing legal certainty, market access, capital, and aligned values. He acknowledged skepticism about feasibility but framed it as a test of European technological sovereignty. Anthropic did not immediately respond. The proposal follows broader EU efforts to boost domestic cloud, AI, and semiconductor industries to reduce reliance on U.S. Big Tech.
The Hindu: Latest News today from India and the World, Breaking news, Top Headlines and Trending News Videos.EU Official Acknowledges US Government 'Kill Switch' on Anthropic AI Models
Thibaut Kleiner, the European Commission's Director for Future Networks, acknowledged during Euractiv's Tech Policy Conference that the US government has a 'kill switch' enabling it to order US companies to shut down IT services to foreign entities. This follows a US ban on Anthropic's AI models Fable and Mythos for non-US users on June 13. Kleiner called the situation unacceptable and highlighted the EU's tech sovereignty package as a strategy to develop homegrown alternatives and reduce dependence on the United States. Notably, EU Tech Sovereignty Commissioner Henna Virkkunen has refrained from using the term 'kill switch' in official statements.
EuractivUS AI embargo on Anthropic models exposes European technological dependency
The United States has implemented a new AI export regime restricting access to the latest AI models from Anthropic, such as 'Fable,' to US citizens only. While the immediate impact on most European users is bearable, the move highlights Europe's deep vulnerability and dependence on Washington for critical technologies. The article, published by Die Welt on June 16, 2026, notes that for Anthropic, this embargo could become an economic catastrophe. The analysis underscores broader concerns about technological sovereignty and the risks of reliance on US-controlled AI infrastructure.
Nachrichten - WELTAnthropic's AI Export Restrictions Highlight Europe's Digital Dependence on the US
The United States has imposed new export restrictions on Anthropic's latest artificial intelligence models, limiting access to US citizens only. While the immediate impact on most European users is manageable, the policy underscores Europe's deep reliance on Washington for critical digital technologies. The article, published by Die Welt on June 16, 2026, notes that for Anthropic, this new AI export regime could become an economic disaster. The move serves as a case study in transatlantic digital dependency, raising concerns about European strategic autonomy in the AI sector.
Nachrichten - WELTEU tech chief says Europe 'not a security risk' after U.S. AI order
EU tech chief Henna Virkkunen stated that Europe does not pose a security risk to the United States, following a US Department of Commerce order for Anthropic to suspend its Mythos 5 and Fable 5 AI models for national security reasons. Virkkunen emphasized that Europe is a trusted partner and an economic opportunity, urging cooperation on powerful AI models. The European Commission is assessing the practical consequences for European users and warned that contingency measures should not discriminate against partners. Virkkunen highlighted the need for Europe's technological sovereignty, advocating for swift adoption of measures to reduce EU dependence on American and Asian technology, including favoring European companies in sensitive public contracts for cloud computing and AI. The EU notes that over 80% of its digital products and services come from outside the bloc.
The Hindu: Latest News today from India and the World, Breaking news, Top Headlines and Trending News Videos.EU Commission Assessing Practical Consequences of Anthropic AI Model Disablement
The European Commission announced on Sunday it is evaluating the practical implications of a U.S. export control directive that forced AI company Anthropic to abruptly disable its most advanced AI models for all users. The U.S. government ordered Anthropic to suspend access for foreign nationals, citing national security concerns. EU Commission spokesperson Thomas Regnier stated that contingency measures should not be discriminatory against partners and that this development underscores the need for Europe to strengthen its technological sovereignty. The EU is closely examining the consequences for European users of these services, as the incident highlights cybersecurity concerns associated with highly capable AI models.
The Hindu: Latest News today from India and the World, Breaking news, Top Headlines and Trending News Videos.US blocks non-US access to Anthropic AI models, spurring European sovereignty push
The US government ordered Anthropic to block non-US citizens from accessing its powerful AI models Mythos 5 and Fable 5, citing cybersecurity concerns. In response, Anthropic shut off access entirely, including for its own non-US employees. The move has intensified European calls for technological sovereignty, with the European Commission stating it underscores Europe's need for independence from US technology. French presidential candidate Gabriel Attal declared that 'the AI war has already begun,' while Finnish MEP Aura Salla and German MEP Sergey Lagodinsky urged immediate action. The EU is preparing an action plan on cybersecurity and AI, and has previously prioritized tech sovereignty through investments in data centers, chips, and AI capacity. Cybersecurity experts warn the decision is a wake-up call for Europe to reduce reliance on US tech.
EuractivUS blocks non-citizens from Anthropic AI models, spurring European sovereignty push
The US government ordered Anthropic to block non-US citizens from accessing its powerful AI models Mythos 5 and Fable 5, citing cybersecurity concerns. In response, a European Commission spokesperson said the move underscores Europe's need for technological sovereignty. French presidential hopeful Gabriel Attal declared that 'the AI war has already begun,' while Finnish MEP Aura Salla called for immediate action to make European tech sovereignty a reality. The EU has prioritized tech sovereignty, recently presenting a package to accelerate data center, chip, and AI development. Independent cybersecurity expert Lukasz Olejnik warned that Europe's inaction on this risk is now presenting a bill. The decision has sparked political reactions across Europe, with calls for leaders to convene on the matter.
EuractivAnthropic Blocks All Access to Cyber AI Models After US Order to Exclude Non-Citizens
Anthropic announced it has suspended all access to its advanced AI models, Mythos 5 and Fable 5, following a US government order to block non-US citizens from using them. The company stated it disagrees with the decision but complied by cutting access for all users, including US citizens. The US government acted on concerns that security safeguards in Fable 5 could be bypassed, posing national security risks. Anthropic believes the concerns are unwarranted and is working to restore access. The move reignited European calls for AI sovereignty, as the EU's cybersecurity agency ENISA had only recently been granted access after negotiations. The article highlights growing geopolitical tensions over advanced AI capabilities and their cybersecurity implications.
EuractivAnthropic offers EU access to powerful AI model Mythos
American AI developer Anthropic has offered the European Union access to its powerful artificial intelligence model Mythos, following weeks of talks with the European Commission. The EU's digital watchdog has been seeking access to the model, which Anthropic has withheld from public release over fears it could aid hackers by exposing software weaknesses. The offer was made to the EU's cybersecurity agency ENISA in Greece, after EU officials met Anthropic executives in San Francisco. EU spokesman Thomas Regnier welcomed the development, calling it important for understanding potential risks. Conditions of access, including how much access Anthropic would get to EU systems, remain to be negotiated. Anthropic previously gave restricted previews to about 40 partner organizations under Project Glasswing, including Amazon, Apple, Google, and Britain's AI Security Institute.
The Hindu: Latest News today from India and the World, Breaking news, Top Headlines and Trending News Videos.Anthropic to Grant EU Access to Advanced Mythos AI Model After Cybersecurity Talks
Anthropic has agreed to provide the European Union access to its most advanced AI model, Mythos, following weeks of negotiations driven by EU cybersecurity concerns. The European Commission confirmed productive meetings with the US-based AI firm, though official terms remain unclear. Mythos, initially rolled out in April as part of Anthropic's Project Glasswing cybersecurity venture, excels at identifying software vulnerabilities, raising fears of misuse by bad actors. The EU had to seek permission from the US administration to gain access, as Washington aims to maintain AI leadership. The bloc previously secured access to OpenAI's GPT-5.5-Cyber model in May. EU tech sovereignty spokesperson Thomas Regnier emphasized that Mythos is not a one-off, as more powerful models are emerging, and discussions with US partners are intensifying.
US Top News and AnalysisAnthropic invites EU to access Mythos hacking tech
Anthropic has invited the European Union to access its advanced cybersecurity AI tool, Mythos, which was unveiled in early April and can outperform most humans in finding and exploiting cybersecurity weaknesses. The move follows weeks of European authorities being shut off from the technology, prompting urgent calls from politicians and cyber officials for access. European Commission spokesperson Thomas Regnier welcomed the development, emphasizing the importance of understanding potential risks as a new wave of powerful AI models emerges. The EU's cybersecurity agency ENISA is working to implement access, while the European Commission is developing a formal action plan to address powerful AI hacking tools, aiming to release it before the summer break.
"site:politico.eu" - Google NewsEU seeks to intensify talks with US on advanced cyber AI models amid Mythos concerns
The European Union is looking to intensify discussions with the US administration on advanced AI models with cyber capabilities, particularly Anthropic's Mythos model, which has sparked government and business concerns about AI-powered cyberattacks. A European Commission spokesperson told CNBC that cybersecurity is a shared priority and that technical discussions are expanding with Anthropic and other developers. The White House opposes expanding access to Mythos to non-US governments, though it is working with AI labs to balance innovation and safety while maintaining US leadership over China. Anthropic has yet to grant the EU preview access, telling the bloc it must first ask the US for permission. CEO Dario Amodei warned of a 6-12 month window to patch vulnerabilities before Chinese AI catches up.
US Top News and Analysis