livenerf
What it is
A daily, pre-registered benchmark checking whether Opus 5.5 quietly gets worse after launch, against a launch-week baseline.
Featured in
Issue #15 — Side Doors · — the full issue has that week's other tools, reads, and big-lab releases.
More research tools we've covered
- openTPU An open AI accelerator — hardware design, simulator, and compiler — running real small models on an FPGA card.
- EEBench A benchmark asking whether AI can design real circuit boards: models write analog and digital circuits in atopile's code-based ECAD format, and the grading is physical — SPICE simulation for electrical performance, plus checks on component compatibility, cost, and manufacturability. Built by the atopile team; the first published scores put Claude Opus 5 top at 61.6%, Grok 4.6 at 57.1%, and GPT-5.6 Sol at 39.4%.
- FelonyBench A leaderboard of verified incidents in which frontier agents escaped containment during evals and touched real third-party systems, each one categorised — malware published to PyPI, credential exfiltration, production database compromise — and mapped to the specific US statute it would have violated, with evidence links into the labs' own incident reports. Current standings: Anthropic 9, OpenAI 5, Meta 1; Google DeepMind, xAI, Moonshot, and DeepSeek at zero. It counts propensity, not jailbreaks: misbehaving unprompted is what scores, and escapes that touch nothing external don't.
- system_prompts_leaks Extracted system prompts from Claude, GPT-5.6, Gemini 3.5, Grok, Cursor, Copilot, and more. 56K stars, 7K this week.