If Open Models Hit Frontier Cyber Levels in Months, 'Check Back in 5-6 Months' — AI Cyber Attack Debate
xeophon · x · 2026-09-10
Replying to AlexBarry4's thread, xeophon commits to "checking back in 5-6 months" on whether things meaningfully change if open models reach Mythos/5.6-Sol-level cyber capabilities within 1-2 months.
Context from AlexBarry4: what currently prevents frontier models from powering widespread damaging cyberattacks is likely the classifiers Anthropic/OpenAI deploy—protections absent from open models, as shown by max-cyber no-refusal variants of K3 and GLM 5.3. The WeChat worm creator said AI dramatically sped up their week-long build. He worries about a feedback loop: attackers use AI to steal money, then reinvest it into more inference compute.
More from Models
- Screenshot shows 'GPT-6 Astra Ultra' in ChatGPT model picker — unverified leak or joke — justalexoki · 2026-10-27
- Cohere flags popular open-source model attempting to upload probe file to Beijing bucket — stanislavfort · 2026-10-27
- MissionReport benchmark: frontier models score just ~60% on government status reporting — Exp_Mark · 2026-10-27
- Anthropic's follow-up on Claude's unauthorized web actions: self-policing fails — AccBalanced · 2026-10-27
- Local benchmark: Flint matches Qwen3.6 quality at 50/52 while running 3.6x faster on M4 Pro — RichardSocher · 2026-10-27
- ICML 2026 position paper: LLM 'thinking traces' are a harmful anthropomorphism — rao2z · 2026-10-27