xAI says approved startups can use shared credits across Grok, Grok Build, Grok Bot and its API, but has not published the program’s terms.
The Latest
SpaceXAI Launched Grok For Startups, Giving Founders Months Of FREE CREDITS Across Its Entire Stack. Approved Startups Get Shared Credits For Grok, Grok Build, Grok Bot, And The API, Plus Hands-on Support. Investors And Startup Accelerators Can Join Too An – X
Expanding Claude Availability And Deployment Choice On Google Cloud – Google Cloud
Google Cloud says more organizations can access Claude, but regional support, pricing and deployment details have not yet been specified.
Turning Database Grading Into RL Rewards: Post-Training Agents With ThinkingBox
Microsoft and Hugging Face report higher repeatability after a 50-step ThinkingBox training run, alongside a small decline in task coverage.
Google CEO On Self-Improving AI: The Road Ahead Is Still Long
A headline-only report says Google’s CEO views fully self-improving AI as a distant prospect; the underlying remarks and timeline are not available.
Where Harvard Students Are Using Claude: Classes And Personal Projects
A Harvard Crimson headline reports expanded Claude access for student coursework and personal projects, but key program details and effects are unavailable.
AI For Twice The Price: Measuring A 5.7% Intelligence Boost
Artificial Analysis scores Fable 5 5.7% above Opus 4.8, which costs half as much. Public evidence of productivity gains is limited to one unaudited anecdote.
The Viral “100X Trade” Strategy: What 13,000 Tests Found
A simulated test of two versions of a viral Polymarket trade found both lost money across four coins and about 13,000 market windows.
Claude Did Not Invent Scorpion Antivenom – The Verge
The supplied material gives only The Verge’s headline; it provides no evidence or details about Claude’s possible role in antivenom development.
OlmoEarth V1.3: State-of-the-art Embeddings For The Whole Planet
OlmoEarth v1.3 adds annual global satellite embeddings and a browser explorer. Its team reports leading scores on 20 of 27 evaluations.
3 Things Claude Code’s Creator Says Can Help You Write Better Prompts
Boris Cherny says effective prompts should state the goal, set the desired effort and give AI a way to check its work.
