AI researchers lament models refusing vulnerability analysis of old PHP code despite 'use AI to defend' messaging
voooooogel · x · 2026-10-26
zetalyrae forwards a complaint about absurd AI guardrails: the official line is "AI cyberattacks are here, use our AI to defend," yet models refuse to even analyze vulnerabilities in long-dead PHP junkware. jdpressman argues there should be a clear difference between probing an obsolete, unused PHP app and auditing freshly written code where flaws impact real users — a one-size-fits-all refusal makes no sense.
More from Models
- AI cyber classifier reportedly refuses to help audit Apache code — jd_pressman · 2026-10-26
- Claude Opus 4.5's essay on language as a medium draws 'beautiful mind' praise — repligate · 2026-10-26
- Ternary 1.58-bit Nemotron 30B-A3B GGUF Runs 368K Context on 12GB GPU — pmttyji · 2026-10-26
- Dev shares $5,107 monthly AI bill: 10B tokens, DeepSeek cost $0.70 — cneuralnetwork · 2026-10-26
- Skipping eval checks means you're likely fitting benchmark biases and noise — antoine_chaffin · 2026-10-26
- DeepSeek V4.1-Flash hits 49 tok/s decode on dual Radeon R9700 in local test — Longjumping-Elk-7756 · 2026-10-26