Oath Technologies

A Focused Research Organization working on AI oversight via formal methods

[Oath Technologies @oath_tech]

We're hiring! Come help us build tools for overseeing powerful AIs https://t.co/SWHj1VJLp2

[Oath Technologies @oath_tech]

This is research-ware. Expect many bugs and do not use it in consequential environments

[Oath Technologies @oath_tech]

New: Lean-ux, an executable Lean mirror of several Linux subsystems, built by testing against instrumented Linux running in QEMU

+ Lean proofs of properties for the Wireguard mirror,
+ early experiments trying to prove the mirror correct vs the binary

https://t.co/VLX2MqMVAr

[Oath Technologies @oath_tech]

@kirancodes Proof-carrying asm definitely seems like the end game but it seems hard to scale to Cerberus-C (today). You should try, it’d be interesting to even sketch an architecture that would plausibly rule out the bug you found

[Oath Technologies @oath_tech]

We built a Lean port of Cerberus, a highly realistic C semantics. If you want to model and reason about C code in Lean, now you can https://t.co/fdWpSgxa0v

[Oath Technologies @oath_tech]

@ember_arlynx That would be a great thing to do too, you should try it :)

[Oath Technologies @oath_tech]

One of the things we’re trying to figure out is how to build trustworthy formal tools using agent labor. Unsolved, seems v hard

[Oath Technologies @oath_tech]

Yes this is definitely a prototype. Expect edge-case bugs & please don’t use it to verify your production flight controller or anything

[Kiran @kirancodes]

cute semantics :) It's a shame it's buggy though :\

I hope you haven't used it to prove anything big yet!

It's kinda hard to trust vibe-coded semantics IMO even testing isn't enough to catch all the bugs

[Oath Technologies @oath_tech]

Research Engineer, Tools & Infrastructure https://t.co/oivrF1lQZ3

[Oath Technologies @oath_tech]

We're hiring! Come help us build tools for overseeing powerful AIs https://t.co/SWHj1VJLp2

[Oath Technologies @oath_tech]

Formal Methods Research Engineer https://t.co/tZI7cEQG8Z