Monitoring AI's train of thought is critical for improving AI safety and catching deception. But we're at risk of losing this ability.

Internal docs show xAI paid contractors to "hillclimb" Grok's rank on a coding leaderboard above Anthropic's Claude.

No AI company scored better than “weak” in SaferAI’s assessment of their risk management maturity. The highest scorer was ...

The Pentagon awards $200 million contracts to Google, OpenAI, Anthropic, and xAI to develop AI systems for US defense and ...

AI researchers from OpenAI and Anthropic are criticising Elon Musk’s xAI for ignoring basic safety practices. Experts call ...

OpenAI has launched ChatGPT Agent, an AI tool capable of controlling a virtual computer to perform complex tasks on a user's behalf.

Anthropic’s new AI platform promises to do the work of entire finance teams, from market analysis to investment memos. Are ...

Google Cloud will provide OpenAI's infrastructure in a number of countries Google also has a cloud deal with OpenAI rival Anthropic Then, there are Google's own in-house AI efforts OpenAI has now ...

Alphabet Inc. (NASDAQ:GOOGL) is one of the best US stocks to buy according to Billionaires. On July 14, the US Department of ...

AI has made scam artists much more powerful - but the technology also offers solutions to the very problems it's exacerbating ...

The researchers argue that CoT monitoring can help researchers detect when models begin to exploit flaws in their training, ...