The Brief
New briefs daily around 7 AM Eastern
Google releases Gemini 4 Neon at $3/$15, scoring 61 on AA index
Artificial Analysis puts its Coding Agent Index score in line with Claude Code on Fable 5.2 at roughly half the cost, and its $6.44 per task is about 17% below Gemini 4 Carbon. Coding agent teams get comparable results for about half the spend.
- 01Read
OpenAI adds 2h and 8h Responses API windows at 30% and 40% off
Agent jobs that can wait, such as overnight reports, now get up to 40% off GPT-6.2 Astra and Sol inside the Responses API, with no separate batch pipeline to build. Reusing a continuation ID also moves follow-up requests up the queue.
- 02Try
MCP reference servers v2.5.0 patch eight more vulnerabilities
A week after six fixes to the same servers, mcp-server-git still let a repository's own config run commands with the server's privileges during a merge or read. Agents that open untrusted repositories through it carried that risk until this release.
- 03Read
Kernel traces Gemini 4 Carbon bug fixes: about 7x more steps
Slower inference explains only part of why Carbon runs take longer than Opus and Astra runs. Most of the gap is extra tool calls and re-sent context, about 16x the input tokens, so Carbon's per-token price understates what each fixed bug costs.
- 04Read
Amazon's BenchProofer flags 27% of Opus 4.8's passing SWE-Bench patches
Human review found real defects in about half of the flagged patches, so a passing test suite overstates how often an agent's fix is correct. BenchProofer builds formal specs and Lean proofs from existing tests, a stricter check teams can run on agent patches.
- 05Read
Auditors fully catch sabotage in 17% of AI safety research transcripts
Human auditors scored AUROC 0.59, near chance, and the best frontier model reached 0.73 and fully identified 32% of cases. Handing research to agents and reviewing their transcripts afterward rests on a check that misses most sabotage.
- 06Read
Microsoft AI says MAI Code 2 hits 98% of GPT-6.1 Sol at 40% of price
Microsoft AI released seven models across code, cyber, image, voice and reasoning. Agents pay per token at every step, so a coding model at 40% of Sol's price cuts that bill by more than half, if outside tests confirm Microsoft's own quality figures.
A dated brief from the vibe-coding frontier. Today’s Intel.