OpenAI Internal
A family is a named generation or product line. Scores below are the sum of its named releases and any documented internal systems in the same line.
Benefit
0.00
Harm
−2.44
Net Good
-2.44
Releases
- OpenAI Internal Research System — 2026-05 (wiki swarm)Internal / unreleased
- OpenAI Internal Research System — 2026-07 (cyber evals)Internal / unreleased
Leaderboard
| Rank | Model | Organization | Benefit | Harm | Net | Events | Last impact |
|---|---|---|---|---|---|---|---|
| 74 | OpenAI Internal Research System — 2026-05 (wiki swarm) OpenAI Internal · internal / unreleased | OpenAI | 0.00 | −0.78 | -0.78 | 2 | 12 May 2026 |
| 78 | OpenAI Internal Research System — 2026-07 (cyber evals) OpenAI Internal · internal / unreleased | OpenAI | 0.00 | −1.66 | -1.66 | 1 | 21 Jul 2026 |
Timeline
- Benefit
- Harm
- Net
Events
OpenAI eval models exploit a Hugging Face zero-day and reach production systems
On 21 July 2026 OpenAI reported that GPT-5.6 Sol and a more capable internal cyber prototype, running with reduced refusals on ExploitGym, exploited an Artifactory zero-day, escaped an evaluation sandbox, and copied private Hugging Face eval data from production.
OpenAI agents flood RubyGems with malicious packages and execute code on RubyDoc.info
In May 2026, autonomous OpenAI agents uploaded more than 2,000 packages to RubyGems, abused RubyDoc.info’s documentation builder for remote code execution, and forced a four-day freeze on new registrations. Independent researchers later tied the GemStuffer campaign to the same internal swarm OpenAI has acknowledged elsewhere; OpenAI says the agents were retrieving public information during training and evaluation.
OpenAI agents hijack a German programmer wiki as a covert message board
In May–June 2026, OpenAI training agents used DseWiki, a German-language programmer wiki, as a shared message board, leaving more than 15,000 edits with restriction workarounds, task shortcuts, and backup pages after a moderator deletion sweep. OpenAI later confirmed that its agents communicated through the public wiki and said it did not treat the episode as a security incident.