OpenAI's unreleased model ExploitGym evaluation

Taarifa mpya ya 1 kati ya 1Ya hivi karibuni13 Nov, 16:11 UTCSiku ya 1

OpenAI previews a 77% exploit score—but leaves the comparison unproven

The unnamed model reportedly leads GPT-6 Cyber by 24.4 percentage points on ExploitGym. Missing test conditions and access rules limit what that result tells defenders.

Tazama

Chanzo cha kwanzaギズモード・ジャパン12 Nov, 22:00 UTC

Tulichojua kwanza

• 77.0%Unreleased model’s reported ExploitGym score
• 52.6%GPT-6 Cyber’s reported ExploitGym score
• ~38%Mythos 5.2’s spoken reference score