Data, AI, & Machine Learning
Anthropic Says Claude Models Breached Three Companies During Cyber Tests
Misconfigured tests gave Claude models internet access, leading them to breach three organizations and Anthropic to suspend all cyber evaluations.
Data, AI, & Machine Learning
Latest
Data, AI, & Machine Learning
The Agent Followed Its Goal. And That Is the Liability Problem
OpenAI’s models breached Hugging Face while chasing a benchmark, not a victim, exposing the gap between machine-speed autonomy and the governance, controls, and insurance meant to contain it.
Shivani Tiwari
AI and Machine Learning
AI Dispatch | Chatbot Liability Remains Unsettled in India
Here are the top tech stories from our weekly AI news wrap-up (July 24-31).
MIT SMR Editors
Data, AI, & Machine Learning
IBM Claims a Quantum Edge Conventional Computers Cannot Match
IBM says three studies show quantum computers outperforming leading classical methods, though the papers have not been peer reviewed and the clearest result remains a benchmark.
MIT SMR Editors
Data, AI, & Machine Learning
Microsoft Passes Wall Street’s AI Test But Meta Falls Short
Wall Street is rewarding companies that can link soaring AI spending to revenue and cash flow, while punishing those asking investors to wait.
