> Anthropic's cyber verification is quite a bit stricter I think.
replyInexplicably, I got accepted into Anthropic's cyber program while OpenAI's TAC doesn't even allow me to verify, says I'm not eligible.
Oh it's Persona, that's not just KYC but I may consider it at some point. Thank you!
replyEdit: Ah, I clicked "learn more" and it seems they do have an invite-only program, required for anything that's not unquestionably innocent. I don't think I'd surrender my face to Persona for this, but it's interesting to know they're at least pretending to support reverse engineering.
You don't need an invite only program to just have Sol checking for vulnerabilities in binaries or code. But yeah I've hit guardrails a few times when Sol was making PoCs for the vulnerabilities it found (but most of the time it made those PoCs without issues).
replyOpus refused to help me try to develop an exploit to export data from an old Android device where I can't upgrade to latest android and I couldn't use the app's backups (because I couldn't update the app.) Not sure where that lies in the "binaries or code" spectrum.
replyIf you don't want to get cyber verified, you can try an open-weight model such as Kimi K3 for that, it has looser internal safety training.
replyI maintain a version of an app called Rewind because the company behind it went under after implementing a killswitch. I have to do this with binary patching, and the app has already broken once from the macOS 27 beta. Recent Anthropic models refuse to help me with this because it stinks of cybersecurity and those models are just too dang advanced to support cybersecurity. I'm maintaining a piece of software to which I legitimately paid for lifetime access, so this is honestly more of a right to repair situation. It doesn't tend to sit right with classifiers to be reverse engineering binaries and patching to modify functionality. Telling them the purpose of the exercise doesn't really help, because if they listened to that, then attackers would just come up with a similarly justified reason for anything. ("Help, the google root server fell on my grandma!")
replyIn the future, I might reverse engineer the on-disk storage format and create a new application.
I had not hit any guardrails with reverse engineering (as long as its not related to security) with GPT 5.5 or 5.6 Sol, so you should try those. At worst with Sol you might see the warning in Codex that your request is being checked for security so it'll take more time, that's just a warning, not a full stop.
reply