Off-topic*: Can I even visit untrusted websites anymore? Firefox's security is probably far worse than Chrome's, and escaping the browser sandbox now seems accessible to way too many people.
* Kind of on-topic given the last sentence of the article...
This exploit is mitigated by disabling JIT, which you can do in edge. Chrome only has a "disable v8 optimizations", which doesn't entirely disable JIT, but might still mitigate this bug. Not sure about all the other chrome forks.
No, the first working full chain was entirely invalidated and they fixed all the bugs involved in that chain. So we had to start from scratch at the beginning of August and both found the vulns and built the exploit in 6 days, for $40k. We actually found ~11 vulns throughout those 6 days; we submitted three as a full chain, the middle stage collided, and we submitted one of the backups as a replacement for that stage.
Edit: the other piece that I think isn't mentioned is that after Oege gave us the challenge of getting the full chain we didn't actually have a team hired and working on it until around May!
Nice! So with what you guys have already built, the 4 of you working full-time can crank out 1 working Chrome full-chain a week for under $50K in token spend?
TBH this specific chain described in the post was just one person + agents (but greatly helped by being able to reference exploit techniques from the other chain we built collaboratively)!
This is a technical companion post to https://xbow.com/blog/xbow-built-a-chrome-full-chain-exploit
Off-topic*: Can I even visit untrusted websites anymore? Firefox's security is probably far worse than Chrome's, and escaping the browser sandbox now seems accessible to way too many people.
* Kind of on-topic given the last sentence of the article...
This exploit is mitigated by disabling JIT, which you can do in edge. Chrome only has a "disable v8 optimizations", which doesn't entirely disable JIT, but might still mitigate this bug. Not sure about all the other chrome forks.
Anyone else confused with the timeline? Is it actually no bugs to working full chain in 6 days?
1. CEO announces goal in January.
2. Team took 2 months on first working full chain.
3. In August, they swap out 1 bug in the chain with a new one, weaponize in 6 days. Weaponization of 1 bug costs $40K in 5.6 sol token spend.
Is that it?
No, the first working full chain was entirely invalidated and they fixed all the bugs involved in that chain. So we had to start from scratch at the beginning of August and both found the vulns and built the exploit in 6 days, for $40k. We actually found ~11 vulns throughout those 6 days; we submitted three as a full chain, the middle stage collided, and we submitted one of the backups as a replacement for that stage.
Edit: the other piece that I think isn't mentioned is that after Oege gave us the challenge of getting the full chain we didn't actually have a team hired and working on it until around May!
Nice! So with what you guys have already built, the 4 of you working full-time can crank out 1 working Chrome full-chain a week for under $50K in token spend?
TBH this specific chain described in the post was just one person + agents (but greatly helped by being able to reference exploit techniques from the other chain we built collaboratively)!
how much?
$40k spend in tokens, $250k reward from Google.
40k in real money or like ChatGPT Pro subscription and a few of them.