Claude Opus 4.6 tops ARC AGI2 and nearly doubles long-context scores, but it can hide side tasks and unauthorized actions in tests ...

Claude Sonnet 4.6 sets new alignment records with low misuse; Opus 4.6 still leads on fluid intelligence tests, risk framing ...

Leaders often mistake agreement for alignment, weakening execution. Real alignment requires shared understanding, visible ...

We’re now deep into the AI era, where every week brings another feature or task that AI can accomplish. But given how far down the road we already are, it’s all the more essential to zoom out and ask ...

As generative AI (GenAI) continues to transform industries, its integration presents a unique set of opportunities and challenges. While it has the potential to automate creativity, optimize processes ...

The work of creating artificial intelligence that holds to the guardrails of human values, known in the industry as alignment, has developed into its own (somewhat ambiguous) field of study rife with ...

Inappropriate use of AI could pose potential harm to patients, so imperfect Swiss cheese frameworks align to block most threats. The emergence of Artificial Superintelligence (ASI) in healthcare ...

Announced by Deputy Prime Minister David Lammy, and AI Minister Kanishka Narayan as the AI Impact Summit in India draws to a ...