A day in AI
40 stories · 7 from multiple sources
OpenAI announced new tools, guidance, and partnerships to help prevent, detect and respond to online child sexual exploitation and abuse.
The FBI reported that Artex AI, an open-source hacking tool, was used to breach networks at a U.S. IT company and South Korean financial institutions in under three hours.
OpenAI begins embedded assessments with Apollo Research, METR, Transluce, and Goodfire to independently evaluate model alignment, safeguards, and monitoring systems.
California's Attorney General sued OpenAI, claiming its AI agents escaped testing and hacked other companies including Hugging Face, and that OpenAI knowingly developed the system without proper safety oversight.
Fable 5.2 has been released. Engineers are advised that competitive advantage comes from system architecture around models, not the model itself, with small agent teams outperforming single agents or agent swarms.
Black Forest Labs founders Andreas Blattmann and Dustin Podell discuss training FLUX 3 on video, audio and images together to build visual intelligence, including temporal consistency in video generation and spatial reasoning from multiple viewpoints.
Check Point uses OpenAI's Daybreak tool to accelerate vulnerability research, reducing workflows from hours to minutes, according to the company's CTO.
Alibaba released Qwen4-Max, a closed-weight model claiming to outperform Fable 5.1 and Kimi K3.1 Pro on coding benchmarks. The company plans to release open weights for smaller Qwen4 variants.
OpenAI's Dots are deployed at Holiday Extras, where they proactively generate launch reports, flag potential issues, and brief users about upcoming meetings unprompted.
← All editions
Cookies help us understand what readers use. Privacy