Good write up. I wonder what approach an LLM would have taken for this optimization, say using autosearch.

If you mean auto research it is not bad idea to run it tbh. I have over 1M products now in the database not a bad set to train.

Yes, I meant auto research. Sorry about the typo.

Looks like you have a pretty good closed-loop and data set to run it on.

Even in the world of AI good to see good old fashioned prop making for a video.

I've seen the markets and they sure did move wildly, but what confirms that this was manipulated? (The alternative I was thinking of is "this twitter poster saw those two market resolutions and made up a narrative connecting them").

Somehow a ton of people are caught in the variant.

Long-term relative to lifespan of the 401K holder. Outcome changes a lot for those who are ready to retire.

Interesting that they actually acknowledge there was a change on March 6th. Kudos to the prompt analysis work that uncovered it!

To give credit where it is due: Boris actually submitted a few PRs this week to OpenClaw to increase prompt cache hits. You can see them here: https://github.com/openclaw/openclaw/pullsq=is%3Apr+author%3...

I think the usage patterns of a lot of harnesses are pushing against their planned capacity. I would say they can certainly explain themselves a lot better.

Link is broken.

missing a ? character. https://github.com/openclaw/openclaw/pulls?q=is%3Apr+author%...

I can't see any results any more.

Thank you!

AH! Thanks, that's useful context!

Thanks for sharing the prompts: https://github.com/califio/publications/blob/main/MADBugs/CV...

the prompts show how this was a back-and-forth with a lot of nudging, interruptions and steering: it's not Claude writing a full exploit just from a vulnerability description.

Chuck Norris hasn’t died, he summoned the death. RIP.