Anthropic and NSA: An Agreement Despite Pentagon Warning
Le brief IA que les pros lisent chaque soir
Les 7 actus IA du jour, décryptées en 5 min. Gratuit.
Inclus dès l'inscription : notre sélection des meilleurs guides & comparatifs IA.
Choisis ton rythme
Gratuit · Pas de spam · Désabonnement en 1 clic
A Controversial Partnership Between Anthropic and the NSA
Anthropic, a technology company recently classified as a potential supply chain risk by the Pentagon, may nonetheless continue to provide its artificial intelligence models to the National Security Agency (NSA). This decision, reported by the New York Times, was personally approved by Susie Wiles, the White House Chief of Staff.
The situation arises from a critical shortage of the latest Grace Blackwell chips from Nvidia, essential for operating models from OpenAI and other companies. In contrast, Anthropic's "Mythos" model is compatible with older chips, making it indispensable for the NSA's classified networks in the short term.
Security Stakes and Contractual Clauses
The Pentagon had previously expressed concerns about Anthropic due to its refusal to release its technology for "any lawful purpose." However, a new contract, currently being finalized, would include a clause prohibiting the model from processing data of American citizens. The controversial language that had stalled previous discussions would not be included in this agreement.
The White House is considering using this contract as a template for future agreements with other technology companies, despite the potential risks identified by the Pentagon.
A Massive Investment to Address Chip Shortages
To address the long-term chip shortage, the White House has approved a $9 billion investment for the development of new artificial intelligence chips. However, this investment still requires Congressional approval to be implemented.
Brief IA — L'actualité IA en français
L'essentiel de l'actualité de l'intelligence artificielle, décrypté et expliqué chaque jour.