canucker201624 minutes ago
Perhaps this is the exploit chain (they mention libheif at the start of their exploit chain)
```
  https://www.hacktron.ai/blog/hacking-openai
```
and HN discussion
```
  https://news.ycombinator.com/item?id=49749656
```
K0nserv2 hours ago
The one thing you'd expect on a website like this: how the exploit works, is missing.
The entire thing feels like marketing.
- neuronexmachina2 hours ago
  I mean, they kind of do? I assume they're hesitant to write a how-to on how to tailor the exploit image:
  >These are not out-of-the-box exploits. Exploitation requires fingerprinting the target version and tailoring the payload image(s). Some of our RCE attempts landed only after thousands of image uploads. That said, an AI agentic approach with a frontier model like GPT-5.6 Sol cut exploit development time down to roughly 1 to 3 days from initial probe to remote RCE. A motivated attacker can convert a vulnerable upload endpoint into RCE or an info leak.
  - K0nserv2 hours ago
    I was under the impression the underlying issue has been patched. If it hasn't, this page and their initial blog post seem irresponsible. If it has indeed been patched why not provide a detailed write up?
owebmasteran hour ago
Is this the new way to go viral ? Registering a domain and vibecode the content