𝕙𝕒𝕧𝕖𝕟

On a mission to increase collective whimsy. Thinking about AI safety and AI governance. Married to @raelifin.

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

“I’m not worried about AI because you could just shut it down. Just unplug it!”

[Palisade Research @PalisadeAI]

An LLM-controlled robot dog saw us press its shutdown button, and the LLM rewrote the robot’s code so it could stay on.
When AI interacts with the physical world, it brings all its capabilities and failure modes with it. 🧵

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

catching up on reading all the important things about AI from the past few weeks

[Michael Huang ⏸️ @michhuan]

Freeze AI invites comparisons to the nuclear freeze campaign of the 80s that called for an agreement between the U.S. and Soviet governments to halt the nuclear race. Their most famous protest gathered a million people in New York. Something for @MichaelTrazzi to aim for.

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

Yesterday I lay in my hammock, looking up into the fractals of the trees and watching the birds soar overhead.
He sat next to me on the ground, reading, one hand on the hammock to gently rock me back and forth.
Our dog drifted between us, sometimes lying beneath the hammock, then switching over to get pets from him.

And I felt gratitude.

There was a time when I avoided love. Then I opened my heart and found him.

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@Romy_Holland @valleyofthena yeah, if you haven’t already, definitely let them know this is something you want. Part of it may be from not knowing how much demand there is.

[AI StopWatch @AIStopWatch]

'“One, two, three, four, we don’t want a robot war!”

On Saturday, July 11, hundreds of us marched through San Francisco in the Stop the AI Race protest, gathering in front of the offices of OpenAI, Anthropic, and Google DeepMind.

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@Romy_Holland @valleyofthena and a lot of nonprofits lost funding in the past few years with grants disappearing. many are working with much tighter budgets

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@MichaelTrazzi

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

I attended the Stop the AI Race protest in San Francisco last week and wrote about it here @AIStopWatch

https://t.co/2yoq0f4TCB

[Max Harms @raelifin]

The Corrigibility Research Fund just launched, with me as manager. I'll be awarding over $200k in grants and prizes to scientists who advance our understanding of this extremely important subfield of AI alignment! 🚀

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

checking in with my esteemed colleague

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@cogcelia the door is insane, but I think this is probably one of the best bathrooms at Lighthaven otherwise

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@zetalyrae They’re like “to be fair, we said don’t replace mathematicians. Writers tho? 🤖”

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

While the algorithm seems better so far with all around less random noise, I can also see how it may get easier to get burned out on certain topics and need more breaks (which is not necessarily a bad thing) or make it easier for people to get into echo chambers. But seeing my community more seems cool too.

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@TinkerElla95 It’s mostly that the FYP was a little more all over the place, and now it’s more concentrated around people I know and seems hyper concentrated topically

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@lumpenspace I appreciate having seen this post. I feel like I just saw how the debate on stopping development of advanced AI may have a similar flavor to the debate on whether abortion is murder (not a perfect parallel tho)

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@SALTWATERSOPH @theojaffee Scott Wiener with SB 1047 and SB 53

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@NathanpmYoung It’s a mixed bag, and definitely the case that a few years ago there was more naïveté around how politics works, but there has been progress made on that front recently.

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@ilex_ulmus @ordinarytings I was actually pretty impressed with the coalition that showed up to this one! Rationalists, sure, but also a bunch of environmentalists, labor union people, Green Party reps, etc

[Max Harms @raelifin]

Underrated skills/virtues:
* Being a good ally -- actually showing up when it matters (including with money), not letting the narcissism of small differences distract, etc.
* Being a good enemy -- leaving a line of retreat, avoiding escalation, acknowledging shared humanity, etc.

[𝕙𝕒𝕧𝕖𝕟 @haventropy]

@cutesuscat @celestepoasts the ramen is indeed legendary