Best of the last 7 days.
- Checking Citations Mechanically: Verifying That a RAG Answer's Quotes Really Exist in the Source
- Client-Side Rate Limiting for LLM APIs: Token Buckets, Backpressure, and Fair Sharing
- Measuring Over-Refusal: Testing the Requests Your Model Declines That It Shouldn't
- Tokenization Surprises: Why Character Counts, Whitespace, and Numbers Break Your Assumptions
- Dynamic Few-Shot Selection: Choosing Examples Per Request Instead of Per Prompt
- Truncated Output Is a Silent Failure: Handling Stop Reasons and Length Limits
- RPA's Agentic Makeover: HCLTech's Robotiq.ai Deal and UiPath's Cartographer Signal a Market in Transition
- Write a Prompt Linter: Cheap Static Checks Before Any Model Call
- OpenAI delays IPO over AI safety concerns — Ars Technica — AI + 88 more
- Introducing Claude Sonnet 5.5 — AI Hub + 88 more
- OpenAI repotedly in talks to raise $30B round at $1.4T valuation — TechCrunch — AI + 78 more
- OpenAI reportedly ditches model over safety concerns — TechCrunch — AI + 67 more
- Gemini 4 Argon: our next era of frontier intelligence — Google DeepMind Blog + 59 more
- Dario Amodei of Anthropic to Dine With Trump at White House — AI Hub + 45 more
- Anthropic Said to Target Mega-IPO Before Thanksgiving Holiday — AI Hub + 42 more