Company
Offensive-security research company (hacktron.ai) that publishes vulnerability research and exploit-chain write-ups, including AI-model exploitation benchmarks.
Hacktron chained a heap overflow in libheif to an OpenAI SSO flaw and opened a pull request in OpenAI's internal monorepo in under 72 hours. The bug had been fixed upstream a year earlier, but the fix never reached Debian's packages.
Reality
The libheif bug was fixed upstream a year earlier without a security label or a CVE, so a Discourse image check that ignored HEIC left it reachable, and forum sign-in tokens carried full API access to the accounts behind them.
Perspective Coverage
6 publishersReality
Hacktron reports that GPT-5.6 Sol Ultra built a complete Chrome/V8 exploit chain in a controlled test for $1,596.89 of model compute. The benchmark handed the model the source tree and the public security-fix commits, so the figure covers only the compute for one lab task.
Reality
CVE-2026-5430 lets a token signed with an algorithm WSO2 does not support pass authentication as an administrator. watchTowr says tokens with administrator privileges baked in reached its honeypots on September 13.
Reality
WSO2 published the fix in May. watchTowr saw forged tokens arrive at its honeypot in September. Its own replay against a correctly targeted deployment came back with the credentials the gateway holds.
Reality
Cisco says CVE-2026-76460 is under active exploitation. It scores 10.0, needs no credentials, and puts an unauthenticated attacker past the web management interface of an Identity Services Engine appliance.
Reality
Hacktron's report puts the vulnerable code in libheif, two dependency steps below an ImageMagick call on OpenAI's Discourse forum, and says the upstream fix never went through the usual security advisory process.
Reality
Hacktron's path into OpenAI's internal repos ran through a libheif bug Debian had not backported and a single sign-on flow that trusted community.openai.com. The dev.to breakdown says the model changed what the attack cost. The category of attack was the same either way.
Reality