The Signal
Ten stories selected from the week's AI coverage, with a disclosed heat score.
The three that led the week
Seven more worth your time
04OpenAI claims an ex-employee took its product plans and presented them to AppleOpenAI claims a former employee took more than 100 files related to its product plans before joining Apple. It's claimed in a filed document that they presented these plans to Apple.Heat 72.705Anthropic warns small AI models could self-replicate through IoT devicesAnthropic is warning that advances in small AI models could soon usher in a new era of self-replicating malware. It's malware that could self-replicate through IoT devices.Heat 72.106GPT-6-Astra solves 49 of 50 physics research problems at $2.40 apieceOn the TheoPhysEval benchmark, GPT-6-Astra solved 49 of 50 unpublished theoretical physics calculations. It's matched the accuracy of leading specialists at a cost of $2.40 per problem.Heat 72.007How ChatGPT Targets Ads To You Based On Your Prompts, Phone Number, and EmailOpenAI lets anyone inject ads into ChatGPT users' sessions. They're targeted at specific groups of people based on their prompts, email address, or phone number.Heat 71.408Eight states sue OpenAI, saying ChatGPT isn’t safe for kidsEight US states are suing OpenAI because they say ChatGPT isn't safe for kids. The lawsuits accuse the company of harming children and misleading customers.Heat 69.309Now that Character.AI is going bankrupt, what happens to all of its users’ data?Character.AI is going bankrupt, which raises questions about what happens to its users' data. New York Attorney General Letitia James has revealed a years-long multi-state investigation against the company, and they're seeking to require protections on consumers.Heat 69.110Nvidia’s Jensen Huang pledges $2 billion for Europe, $250 million in compute for UK safety labsNvidia CEO Jensen Huang has pledged $2 billion for Europe. He's also pledged $250 million in compute for UK safety labs.Heat 68.5How the heat score works
Each story is scored out of 100: model-rated significance from the title and excerpt contributes 55 points, recency 25, similar coverage across the week 15, and a model-launch keyword signal 5. Only posts with a significance score are eligible. This score is a selection aid, not a fact-check. Near-duplicate coverage is collapsed before the final ten are published.
Practical AI packs, made by us
Practical packs for specific jobs. Check the product page for the included files, software requirements and setup steps before you buy.
See the toolkitsAI Maestro's own digital products; check each product page for contents and requirements. Our own products; some tool links on this site earn us a commission.
AI Maestro · The Signal, every Friday · all editions