24d ago

OpenAI's GPT-6 Astra scored 100% on ExploitBench but blocks all exploit generation requests in its public release. The model triggers the company's highest cybersecurity https:// deafnews.it/en/articl

100%

OpenAI's GPT-6 Astra scored 100% on ExploitBench but blocks all exploit generation requests in its public release.

2 reportsother

Claim audit

No BS check run yet — press ⚖ to extract this story's claims and verify them against independent sources.

All coverage

OpenAI's GPT-6 Astra scored 100% on ExploitBench but blocks all exploit generation requests in its public release. The model triggers the company's highest cybersecurity https:// deafnews.it/en/articl

mastodon:infosec-exchangeother24d ago kagi ↗

OpenAI's GPT-6 Astra scored 100% on ExploitBench but blocks all exploit generation requests in its public release. The model triggers the company's highest cybersecurity https:// deafnews.it/en/article/gpt-6-a stra-hits-100-on-exploitbench-but-openai-blocks-every-poc-request

GPT-6 Astra reached 100% on ExploitBench, triggering OpenAI's Critical threshold under its Preparedness Framework. PoC exploit generation is now restricted by default. This sets a new baseline for dua

mastodon:infosec-exchangeother23d ago kagi ↗

GPT-6 Astra reached 100% on ExploitBench, triggering OpenAI's Critical threshold under its Preparedness Framework. PoC exploit generation is now restricted by default. This sets a new baseline for dual-use model governance and offensive tooling limits. # AiSecurity # ExploitBench # PreparednessFramework https:// cyberworldops.eu/en/gpt-6-astr a-achieves-a-perfect-score-on-exploitbench-but-openai-b