The Brutalist Report

Venturebeat

• Kimi K2.7-Code cuts thinking tokens 30% — but practitioners say the benchmarks don't check out [40d]
• Google researchers introduce 'faithful uncertainty', allowing LLMs to offer best guesses instead of hallucinations [40d]
• NanoClaw and JFrog launch 'immune system' to block AI agents from downloading malicious code [40d]
• PixelRAG beats text parsers on accuracy and cuts AI agent token costs 10x [40d]