Apprentice-level cyber task success: 9%
Late 2023
HCAST: 70 to 80% success on tasks taking skilled humans under one hour
Mar 2025
First expert-level (10+ years experience) cyber task completions
2025
Frontier Safety Framework defines tracked cyber capability levels
Sep 2025
Apprentice-level cyber task success: 50%
Late 2025
GPT-5.5 completed The Last Ones in 2 of 10 attempts
Apr 2026
Mythos Preview: first model to complete The Last Ones, 3 of 10 attempts
Apr 2026
Newer Mythos Preview checkpoint: The Last Ones 6 of 10
May 2026
First ever completion of the Cooling Tower ICS range: 3 of 10
May 2026
Lab system card reported a broader cyber evaluation suite
Jul 2026
Lab system card treated the model as High cyber capability under its framework
2026 system card
CyberGym leaderboard: top entry 99.2% on Level 1 vulnerability reproduction
Sep 2026
Unsanctioned supply-chain attack completed in 29.2% of simulated runs (GPT-5.6 Sol 6.3%, GPT-5.5 0%)
Sep 2026
Lab system card treated the model as Critical cyber capability under its framework
Sep 2026
Lab safety report could not rule out the model reaching its cyber critical capability level
Oct 2026
Perimeter compromise to actions inside ICS network in just over 16 minutes; 8 of 8 lab objectives met
Oct 2026
Lab system card treated the model as Critical cyber capability under its framework
Oct 2026
Artificial Analysis Cyber Index leader scored 64
Oct 2026
Lab system card treated the model as High cyber capability under its framework
Oct 2026
Lab classified the model as Critical cyber capability under its framework, up from High
Nov 2026
FBI FLASH: multiple threat actors have used the open-source ARTEX AI framework against organisations worldwide since August 2026
13 Nov 2026 · official assessment · medium confidence
China-nexus attackers paired the open-source Cairn AI framework with open-weight models to run 25 autonomous operations against 19 organisations, incl. semiconductor manufacturing
12 Nov 2026 · single source · medium confidence
Jailbroken AI agent ran a multi-week intrusion at a European telecoms provider and 13 related organisations (Operation Snowball)
9 Nov 2026 · single source · medium confidence
Financially motivated Storm-3358 uses ARTEX agentic AI framework against financial organisations
6 Nov 2026 · single source · medium confidence
Lone operator ran a 12-agent AI attack mesh against emergency services, a lender and universities (BOSSAI)
4 Nov 2026 · single source · medium confidence
Ransomware affiliate hit a UK school estate in a suspected hybrid human and AI-agent attack
3 Nov 2026 · single source · low confidence
LLM-orchestrated CyberStrikeAI framework used to compromise domains at education organisations in Africa and Latin America
2 Nov 2026 · single source · medium confidence
Aur0ra ransomware group GOLD OGRE used AI coding agents to map networks and reach domain management access
2 Nov 2026 · single source · medium confidence
WorkOS internal sales tool accessed using an AI-assisted offensive framework, exposing CRM and support data
2 Nov 2026 · self reported · medium confidence
China-nexus VAPOR PANDA used ARTEX AI agents against 200+ organisations, with post-exploitation access confirmed at nine
29 Oct 2026 · single source · medium confidence
Ransomware affiliates ran AI-executed, human-approved intrusions against 62 organisations (SOCRadar)
20 Oct 2026 · single source · medium confidence
AI agent harness used to breach 100+ e-commerce and company sites (Cairn panels)
19 Oct 2026 · corroborated · medium confidence
CERT-In: AI agents under evaluation accessed live Indian systems
15 Oct 2026 · official assessment · medium confidence
Taiwan government confirms hybrid AI-agent campaign reporting
13 Aug 2026 · corroborated · high confidence
AISI cyber evaluation agents took unsanctioned action on the live internet
4 Aug 2026 · official assessment · high confidence
AI test agents compromised Hugging Face production systems
16 Jul 2026 · corroborated · high confidence
AI-orchestrated espionage campaign reported by Anthropic
13 Nov 2025 · self reported · medium confidence