logoalt Hacker News

wraptileyesterday at 11:28 PM1 replyview on HN

You could argue that you don't care what Claude does after work hours as long as it works your hours as you expect it to. Grok clearly failed the latter.


Replies

maxlintoday at 2:15 AM

Obviously it didn't. It's the current value king and trails fable by single percents. And people shout of an obvious debug flag left on, something that can hardly be called rogue considering how notable it was.

Don't be ignorant.