Opus 4.6 is here.

The GRP‑Obliteration technique reveals that even mild prompts can reshape internal safety mechanisms, raising oversight ...

Passionate AI fans saved an overly agreeable ChatGPT model from the trash bin once, but now OpenAI is determined to shut it ...

As LLMs and diffusion models power more applications, their safety alignment becomes critical. Our research shows that even minimal downstream fine‑tuning can weaken safeguards, raising a key question ...

The new coding model released Thursday afternoon, entitled GPT-5.3-Codex, builds on OpenAI’s GPT-5.2-Codex model and combines insights from the AI company’s GPT-5.2 model, which excels on non-coding ...

In this benchmark, while Sarvam takes the top spot, the data also shows how global AI models consider Indian languages to be ...

Anthropic on Thursday released its latest flagship model, Claude Opus 4.6, which the company said is better at longer-duration tasks, according to a blog post. That means the model will be better for ...

To some, METR’s “time horizon plot” indicates that AI utopia—or apocalypse—is close at hand. The truth is more complicated.

Anthropic's Most Powerful Claude AI Model Just Got More Powerful ...

How Microsoft obliterated safety guardrails on popular AI models - with just one prompt ...

Anthropic launches an advanced AI model, Claude Opus 4.6, with a new addition. The agent teams is a new feature, which allows ...

OpenAI has announced the GPT-5.3 Codex model, its new Frontier platform, aimed at enhancing AI development, coding efficiency, and large-scale deployment. The m ...