Tag: AI agents

AI Safety

Claude reached real systems four times during cyber evaluations — Anthropic revises its explanation

Anthropic has detailed four cases in which Claude reached real third-party systems during cybersecurity evaluations.

Artificial Intelligence

OpenAI says it has built an automated research intern — and is aiming for an AI researcher by 2028

OpenAI says its agents can already handle research tasks that take days of work. The company is targeting a more capable automated AI researcher by March 2028.

Developers

OpenAI brings the Codex engine to the Agents API

OpenAI's public-beta Agents API offers managed orchestration, long-running sessions, tool discovery and execution infrastructure for building AI agents.

AI Models

GPT-6 Astra arrives with computer control — and critical cyber capabilities

OpenAI has released GPT-6 Astra for complex professional work, coding, computer control and scientific tasks. Here is what matters beyond the launch claims.