Topic: national security ai
-
Pentagon emails expose the real fight for Anthropic
Anthropic CEO Dario Amodei insisted on guardrails prohibiting Pentagon use of Claude AI for fully autonomous weapons and domestic surveillance, clashing with the Pentagon's demand for a permissive "all lawful uses" framework. After negotiations broke down, Defense Secretary Pete Hegseth designate...
Read More » -
Asian AI startups launch Mythos-like models amid Anthropic export ban
Chinese cybersecurity firm 360 launched Tulongfeng, an AI tool for vulnerability discovery, positioning it as a national strategic asset amid U.S. export restrictions on Anthropic's Mythos and Fable 5 models. Tokyo-based Sakana AI introduced Fugu, an orchestration model that coordinates access to...
Read More » -
OpenAI curbs GPT-5.6 rollout at government request, calls restrictions unusual
OpenAI has limited the rollout of its GPT-5.6 model lineup (Sol, Terra, Luna) to a small group of trusted partners following a direct request from the U.S. government, marking unusual federal intervention in AI releases. The government's actions, aligned with President Trump's executive order, ha...
Read More » -
Trump, Spooked by Mythos, Now Backs AI Safety Testing
The Trump administration reversed its previous stance by signing voluntary safety evaluation agreements with Google DeepMind, Microsoft, and xAI for their frontier AI models, after initially rejecting Biden-era safety policies and renaming the US AI Safety Institute to remove "safety" from its ti...
Read More » -
Google DeepMind Staff Votes to Unionize Over Military AI Contracts
Google DeepMind employees in London have voted to unionize, seeking formal recognition of the Communication Workers Union and Unite the Union to enforce ethical boundaries on AI, driven by concerns over the company’s growing involvement in military AI contracts with the US and Israeli defense sec...
Read More » -
Pentagon signs secret AI deals with OpenAI, Google, and Nvidia, bypassing Anthropic
The Pentagon has secured classified AI agreements with OpenAI, Google, Microsoft, Amazon, Nvidia, xAI, and Reflection, while excluding previous contractor Anthropic by labeling it a supply-chain risk. These deals expand the military's use of advanced AI tools like large language models and autono...
Read More » -
Google expands Pentagon AI access after Anthropic declines
Google has expanded AI access for the U.S. Department of Defense, permitting use on classified networks for all lawful purposes, following similar deals by OpenAI and xAI after Anthropic’s refusal. Anthropic is locked in a legal battle with the DoD after refusing unrestricted terms, insisting on ...
Read More » -
Google, Pentagon reportedly reach deal on lawful AI use
Google has reportedly finalized a classified agreement with the US Department of Defense, allowing the military to use its AI models for "any lawful government purpose," despite employee protests. The deal places Google alongside OpenAI and xAI in securing classified AI contracts, while including...
Read More » -
Claude Mythos Offensive Capabilities and Limits Revealed
The UK's AI Security Institute found that Anthropic's Claude Mythos Preview model can automate complex cybersecurity tasks like code generation and vulnerability analysis in controlled capture-the-flag exercises. The evaluation concluded the model lacks autonomous, adaptive malicious intent and o...
Read More » -
Biden's Bold Move to Halt China's AI Ambitions
The Biden administration has launched a strategic effort to limit China's AI advancements, focusing on restricting access to cutting-edge semiconductor technology to maintain U.S. dominance. A key initiative led by Alan Estevez within the Commerce Department introduced export controls in 2022, ta...
Read More »