Artificial Intelligence

Ask a Small Language Model to Show Its Work. Then Change One Number.

A new Amazon study finds that small models solving math via chain-of-thought are mostly copying the last number they see, not computing the answer. The finding matters for anyone using CoT faithfulness as an oversight signal.

The $250 Million Company With 2.1 Stars

Anthropic Cofounder Told the Pope AI Labs Cannot Govern Themselves. The Vatican Listened.

Anthropic Is Paying Humans $400K to Do What Its Own AI Should Do Better

Anthropic Plans to Open Its Best Cyber Model to Everyone. It Cannot Yet Prevent Misuse.

Anthropic Launched a Business AI Tool. Hours Later, a Security Firm Published Its Secret Flaw.

AI Agents Are Talking to Each Other the Slow Way. Columbia Has a Fix.

An AI That Chooses Being Provably Correct Over Being Convincing

The University Lab That Beat Google and OpenAI at Their Own Math Benchmark

Your LLM Does Not Know What It Does Not Know

Google API Keys Now Have a Price Tag Nobody Agreed To

Enterprise AI Agents Do Not Need More Memory. They Need Non-Regression.