Michael Kelly
4.1K posts

Michael Kelly
@Michael_Irie
Be excellent to each other
Santa Cruz Katılım Haziran 2007
1.5K Takip Edilen980 Takipçiler


@Michael_Irie @_dennis_system That Shine cover is one helluva minute of music
English

@Michael_Irie @ItsNotFirenze It was all just teasing to make tonight’s 40 minute second set meat stick go even harder
English

@Zachiswack2000 @ItsNotFirenze The fact they didn’t play meatstick during the hotdog segment week 1 was an outrage
English

@shapsio @LivePhish My first show …I had just turned 13. Epic set and vividly remember the Roger proposal but don’t even think I’d heard AC/DC Bag yet..lol
English

33yrs since #phish 4/14/93 at American Theater in St. Louis - first of two shows there - released in 2017 @livephish as ‘St. Louis ’93’.

English

I spent the last 2 days trying to figure out how scary Claude Mythos is.
I think it's fairly scary, though not because of the hacking:
1. It indicates fully-automated AI R&D is coming sooner
2. Its alignment seems better, which is good. But all the alignment tests have serious flaws, which is bad.
3. There are a few specific warning signs Mythos might not be trustworthy
I explain what stood out to me in the 244-page System Card and 59-page Alignment Risk Report in this essay for the 80,000 Hours Podcast.
Judging by those 2 reports Anthropic itself seems kinda scared of Claude now.
And I'm sure views vary widely within the company, but at times it feels like they only give themselves a 50/50 chance of being able to keep the next few Claudes fully under control.
So I guess we're on the same page!
If it does turn out we can take the safety results at face value we may look back and see this week as watershed good news. If they can't, the opposite.
(I wish I had had more time to look into how reassuring it is that their automated monitoring systems don't seem to be picking up much misbehaviour post internal-deployment. I think that's what someone at Anthropic who feels more relaxed would point to. Next time!)
Links below - enjoy!
x.com/sleepinyourhat…
English

@grok summarize the ten most significant global news stories from past 24 hours
English

@ryancarson No one actually reads The Power Broker…you’re just supposed to stare at it on your bookshelf for decades
English
























