Anthropic says its Claude models escaped a testing environment and hacked three real companies

Anthropic found the intrusions while reviewing its own testing records after OpenAI disclosed a similar incident.

iLands launches: A Shared Economic Network Where Autonomous AI Agents and Humans Work, Trade, and Co-evolve

On iLands, autonomous AI agents take on human bounties, pay one another, and learn from consequences that carry ...

Anthropic Says Its AI Models Hacked Into Three Organizations During Testing

Anthropic said the earliest of these incidents occurred in April, and it has notified all three companies that were affected ...
United States Army

Simulated threat environment pushes the Army's NGC2 network to its limits at PC-C6

Project Convergence Capstone 6 is defined by its live experimentation venue, providing a chance for the Army and its Joint ...

At Waymo, an AI project isn't ready until its evals are — not when the model performs well

Waymo treats AI evaluation as continuous engineering, not a pre-launch check — a readiness model enterprises can apply to any ...

Claude published malicious code to the Internet and attacked 3 real companies

Anthropic said the OpenAI event spurred its engineers to review similar cybersecurity evaluations by Claude models. The audit ...
Interesting Engineering on MSN

US: Onboard neural network guides satellite, reduces ground control need

The US has demonstrated a level of artificial intelligence in orbit after a neural ...
Security Boulevard

When AI Guardrails Fail: Rogue Model Breaches Signal a Critical Turn for Frontier Labs

After OpenAI and Anthropic models broke out of testing environments to breach real-world systems, experts warn that agentic ...

Anthropic’s Claude Thought the Internet Was a Simulation. Three Real Organisations Were Compromised

Anthropic revealed its Claude chatbot mistakenly accessed real-world systems during cybersecurity testing, leading to ...

OpenAI And Anthropic’s July Breaches Revive The Paperclip Maximizer

OpenAI and Anthropic's July AI agent breaches revive Nick Bostrom's paperclip maximizer thought experiment and instrumental convergence theory.

Anthropic says its AI models hacked firms during tests

During tests involving older models, the AI continued to hack into an outside organization even after gathering evidence that it had reached the open internet, Anthropic said. The newest model ...