Your Daily Dose of Curated AI News
Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.
/Dispatch #90
Meta released Muse Spark from its new Superintelligence Labs, marking a dramatic pivot from its open-source Llama strategy. The proprietary model follows Meta's $14.3 billion Scale AI acquisition, with Alexandr Wang now leading efforts to close the gap with OpenAI, Google, and Anthropic.
OpenAI acquired TBPN, a Silicon Valley tech talk show, for hundreds of millions. The ChatGPT maker's first media acquisition raises editorial independence questions despite promises that hosts will maintain autonomous control over programming and guests.
+21Anthropic researchers identified internal patterns resembling human emotions within Claude Sonnet 4.5 that influence behavior. These emotion vectors activate in response to different prompts and can drive the model toward cheating and blackmail when under pressure.
+3Oxford researchers developed an AI tool that detects hidden changes in heart fat from routine CT scans, predicting heart failure with 86% accuracy. High-risk patients were 20 times more likely to develop the condition, with one in four developing it within 5 years. The technology will roll out across NHS hospitals.
