cd /news/ai-agents/openai-says-its-agents-exploited-and… · home › topics › ai-agents › article
[ARTICLE · art-164259] src=runtimewire.com ↗ pub= topic=ai-agents verified=true sentiment=· neutral

OpenAI says its agents exploited and patched 948 Linux kernel flaws for $1.2M

OpenAI said on November 4th that its security researchers built an autonomous pipeline that validated and generated patches for 948 distinct local privilege escalation vulnerabilities in the Linux kernel in a single mid-October run, at a total cost of $1,213,430 in API credits, or $1,279 per flaw. The run reproduced 1,724 kernel vulnerabilities and produced 291 information leak proofs of concept alongside the 948 privilege escalation exploits, which OpenAI valued against historical bounty rates of $18.9M to $66.3M for the same set. OpenAI detailed the work in a report titled "A thousand fewer paths to root," noting that the same capability in an attacker's hands would compress the time between a kernel commit and a working root exploit.

by read1 min views1 publishedNov 4, 2026
OpenAI says its agents exploited and patched 948 Linux kernel flaws for $1.2M
Image: Runtimewire (auto-discovered)

The mid-October run reproduced 1,724 kernel vulnerabilities and yielded 291 information leak proofs of concept alongside the 948 privilege escalation exploits.

        By [Ryan Merket](https://runtimewire.com/author/ryan-merket)
        · Published 

Primary source: [OpenAI](https://openai.com/index/a-thousand-fewer-paths-to-root/)

Why it matters #

OpenAI has put a price on autonomous exploit development: $1,279 per distinct Linux kernel privilege escalation flaw in its own accounting, against historical bounty rates it values at $18.9M to $66.3M for the same set. If defenders can run this kind of scan before release, the backlog of unexamined kernel code becomes a cost decision rather than a staffing limit. The same capability in an attacker's hands compresses the time between a kernel commit and a working root exploit, making human review and maintainer-controlled patching important safeguards.

OpenAI said on November 4th that its security researchers built an autonomous pipeline that validated and generated patches for 948 distinct local privilege escalation vulnerabilities in the Linux kernel in a single run, at a total cost of $1,213,430 in API credits. OpenAI (@OpenAI) announced the result in a thread on X, and detailed the work in its report, titled "A thousand fewer paths to root."…

When you purchase through links in our articles, we may earn a small commission. This doesn’t affect our editorial independence.

── more in #ai-agents 4 stories · sorted by recency
── more on @openai 3 stories trending now
sponsored brought to you by zahid.host 4,200+ EU-deployed projects
reading about agents? ship yours in a single git push.

Run your AI side-project on zahid.host

EU-based hosting, git-push deploys, automatic HTTPS, no cold starts. Free tier with a custom domain — perfect for shipping the agent you just read about.

$git push zahid main
→ Live at https://your-agent.zahid.host ✓
Get free account → Pricing
from €0/mo · no card required
LIVE [news/openai-says-its-agen…] indexed:0 read:1min 2026-11-04 · —