AI drops attack costs to ~$100 of inference, making legacy code worth exploiting
karinanguyen · x · 2026-10-26
Security observer karinanguyen points out a counterintuitive reality: much of the world's software is implicitly secured by the opportunity cost of attacking it. Your legacy code hasn't changed — but the attacker's calculus has.
When AI reduces vulnerability discovery and exploitation to roughly a $100 inference bill, systems that were never worth targeting suddenly are. This implies the risk exposure of existing software is being systematically amplified in the AI era, and security teams should re-evaluate their assumptions.
More from AGI Musings
- WEF Future of Jobs 2025: 170M new roles vs 92M displaced by 2030 — tjb-studio · 2026-10-26
- 1,000+ mathematicians pledge no-AI research; critic says fire them all — tak3sh8 · 2026-10-26
- Scott Alexander challenges AI skeptics to name a Metaculus score AIs will never hit — slatestarcodex · 2026-10-26
- Dev's take: each new model makes your project 2-5x less exciting, so aim 100x higher — zack_overflow · 2026-10-26
- One week: 80-year-old cells reset to under 20, AI designing working proteins — rand_longevity · 2026-10-26
- "I'm Dario pilled": Anthropic points nearly all compute at models making models — iruletheworldmo · 2026-10-26