L'administration Trump assouplit les règles d'accès au modèle d'IA d'Anthropic

La Maison-Blanche a autorisé Anthropic à fournir son modèle avancé Claude Mythos 5 à plus de 100 organisations américaines. Cette décision fait suite à des semaines de négociations entre l'entreprise et les responsables gouvernementaux.

Le gouvernement américain a assoupli les restrictions préalables pesant sur le modèle d'IA le plus avancé d'Anthropic. Ce changement permet à l'entreprise d'accorder un accès à certaines sociétés et agences gouvernementales américaines.

Plus de 100 organisations sont désormais éligibles pour recevoir le modèle connu sous le nom de Claude Mythos 5. Les responsables ont décrit cette mesure comme le résultat de longs pourparlers avec l'administration Trump.

Ce changement de politique ne s'applique qu'aux entités nationales. Aucun détail n'a été divulgué concernant les organisations concernées ni le calendrier de déploiement.

Articles connexes

Illustration of US official revoking Anthropic AI access due to China concerns, showing models going offline.
Image générée par IA

Trump administration restricts Anthropic AI models over China ties

Rapporté par l'IA Image générée par IA

The Trump administration directed Anthropic to revoke access to its advanced AI models following concerns about ties to China. The move led the company to take its most powerful offerings offline last week.

Anthropic has released a new cyber-focused AI model called Mythos, capable of detecting software flaws faster than humans and generating exploits. The model has raised alarms among governments and companies for potentially turbocharging hacking by exposing vulnerabilities quicker than they can be patched. Officials worldwide are scrambling to assess the risks.

Rapporté par l'IA

Anthropic leaders met with Trump administration officials in Washington on Monday but failed to resolve a dispute over export controls on its advanced AI models. The controls, imposed last week due to jailbreaking concerns, remain in place after the high-level talks.

In the latest development of the Anthropic supply chain risk controversy, a U.S. federal appeals court on April 9 denied Anthropic's emergency motion to block the Trump administration's blacklisting of its AI technology. The court expedited oral arguments for May 19 but ruled the balance of equities favors the government, marking a setback following a prior district court injunction.

Rapporté par l'IA

Anthropic has launched Claude Opus 4.7, a new AI model designed to assist developers with complex coding tasks. The company emphasized its improved instruction-following and memory capabilities. This release follows the earlier announcement of the more advanced Claude Mythos Preview.

The UK government’s AI Security Institute has released an evaluation of Anthropic's Mythos Preview AI model, confirming its strong performance in multistep cyber infiltration challenges. Mythos became the first model to fully complete a demanding 32-step network attack simulation known as 'The Last Ones.' The institute cautions that real-world defenses may limit such automated threats.

Rapporté par l'IA

President Donald Trump signed an executive order on June 2 calling for voluntary government review of advanced AI models 30 days before their release. The order focuses on cybersecurity risks but imposes no mandatory requirements on companies.

 

 

 

Ce site utilise des cookies

Nous utilisons des cookies pour l'analyse afin d'améliorer notre site. Lisez notre politique de confidentialité pour plus d'informations.
Refuser