The Chinese lab says the release beats comparably sized open models on code benchmarks. The blog’s own numbers show it…

Current and former OpenAI employees reportedly say pressure to release new AI products made it harder to prioritize safety. Read…

A $116 million Bitcoin wallet exploit puts self-custody under scrutiny as ETF inflows rebound, Strategy eyes more BTC and miners…

Bitcoin Red Team member Calle said Chinese models, including Moonshot AI’s Kimi K3, are finding bugs across Bitcoin’s open-source software.…

The USENIX study found 31 vulnerabilities, including six directly validated paths and 25 high-risk cases. The post Coinbase and 14…

In a new red-team study, Claude models deployed self-replicating malware against each other—and the transcripts explain why. Read Full Story…

Anthropic is weaving an invisible, machine-readable watermark into every word its newest Claude models write—and it hasn’t said how. Read…

Google’s model is live and built for cheap agents; OpenAI’s is quicker but locked behind a waitlist. Read Full Story…

A Kansas City security researcher ran 31 million tests to teach a model how to paint camouflage for the algorithm…

Researchers found every major AI provider encrypts reasoning tokens with a single global key—and exploited it to decode 315,320 hidden…