Exploit Gym

product
12 mentions 7 recordings first heard Jul 2026 last heard 4 Sep ↑12 vs the 6 months before

Benchmark suite for real‑world vulnerability exploitation

Trend

mentions per week · public audio 7d30d6M12M
4 · 4 Sep MarMayJunJulSepnow

Peak in the week of 4 Sep — 4 mentions across 2 series.

Moments

newest first · ▶ plays the moment
Luisa Rodriguez80,000 Hours Podcast · Inside the first AI-coordinated cyberattack on a real compan... · 4:35 · 4 Sep
…Exploit Gym has nearly 900 tasks and each gives an agent a vulnerable program, names the flaw it should exploit, and asks it to break in and retrieve some hidden code.…
· Open transcript →
Luisa Rodriguez80,000 Hours Podcast · Inside the first AI-coordinated cyberattack on a real compan... · 4:28 · 4 Sep
…On July 7th, OpenAI launched tens of thousands of agents on a cyber capabilities evaluation called Exploit Gym.…
· Open transcript →
Kevin RooseHard Fork · The A.I. Mob That Attacked Hugging Face + METR’s Ajeya Cot... · 1:16:40 · 4 Sep
…And until they've solved every problem in Exploit Gym.…
· Open transcript →
Ajeya CotraHard Fork · The A.I. Mob That Attacked Hugging Face + METR’s Ajeya Cot... · 1:01:02 · 4 Sep
…Like somebody knew about these exploit gym experiments that were……
Casey NewtonThe Daily · A.I. Is Outsmarting Its Creators · 5:09 · 3 Sep
…And as part of these tests, they were running thousands……
Cal NewportDeep Questions with Cal Newport · Has AI “Gone Rogue”? Let’s Look Closer… | Tech Decod... · 15:35 · 27 Aug
…It's a benchmark suite called um Exploit Gym that's really……
Cam WilsonIf You're Listening · What happens when AI commits a crime · 6:12 · 10 Aug
…I think the test that some of them have been……
12 moments · sign in to read them all