Any time a company looks to start selling shares on Wall Street, it’s required to disclose potential risks to investors. In the case of artificial intelligence giant Anthropic, that has included the possibility that its products could threaten nothing less than humanity itself.

Anthropic said its Mythos 5 model almost never stopped a cyberattack once it began. Even when the agent was told it was harming real people? Crickets.

Claude Duet, a personalized, proactive AI agent that runs on its own cloud-based computer and works when you sleep.

Anthropic has introduced a new low-cost AI model, the Claude Opus 5.5, that offers similar capabilities and safety upgrades as the Claude Fable 5.1 model. The company says the new Opus 5.5 leads in agentic tasks, coding, computer use, and knowledge work ...

Though Anthropic doesn't call itself a software company, it bears a strong resemblance to one.

Anthropic are out with yet another update to Claude AI: the company's Claude Code and Cowork tools can now remotely control your Mac on your behalf. When Claude lacks a direct connector for a given app like Slack or Google Calendar, it falls back to ...

Anthropic sits atop a pyramid of business partners and suppliers, making it a barometer of the broader economy. Its initial public offering will have widespread implications.

Single agent vs multi-agent AI: Anthropic's Applied AI team finds one well-designed agent beats multi-agent pipelines on quality, speed, and cost in most enterprise deployments, running twice as fast and costing up to 5x less,

Users claim that Opus 5.5 guardrails blocked Claude from reverse-engineering smart TVs. It's an unverified anecdote: safeguards may be blocking TVs by mistake or targeting other risks. If true, people could turn to other or open-source LLMs to bypass Claude’s TV restrictions.