Tess Hegarty 🔸

2.6K posts

Tess Hegarty 🔸 banner
Tess Hegarty 🔸

Tess Hegarty 🔸

@thegartsy

AI development is inherently political & social. PhD @Stanford | formerly @MIT | 🔸 10% Pledger

Katılım Şubat 2024
695 Takip Edilen981 Takipçiler
Sabitlenmiş Tweet
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
Tech bros hyping AI: “We are building a literal god! AI will solve all of our problems, and deceleration is literally murder.” Tech bros denying AI risk: “AI is basically a search engine. Why are you scared of a computer multiplying numbers??” Me: wait so which is it? 🤔
English
64
43
548
50K
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
@So8res @liron @Grimezsz I share the intuition that a debate against eg Beff legitimizes him/his position more than seems warranted. I also felt like the AI Doc should’ve contextualized him more for viewers who aren’t familiar with his deal. Though ideally shouldn’t have given him a platform at all.
English
0
0
9
150
Nate Soares ⏹️
Nate Soares ⏹️@So8res·
@liron @Grimezsz (I'm enthusiastic about debating respected academics or folks who try to grow ASIs, if they argue that "smarter automatically means more good" or "I don't care if humanity is wiped out". I'm less interested in debating rando internet personalities who hold that position.)
English
7
0
81
2.5K
Liron Shapira
Liron Shapira@liron·
Yep, there's a shortage of quality debate between representatives of the AI camps. It isn't a perfect format, but it's better than tweets and echo-chamber interviews. 👤 @Grimezsz gets why it needs to happen 👤 @So8res is willing to take on major discourse participants 👤 @BeffJezos says he's game All 3 of you are invited on Doom Debates!
Liron Shapira tweet media
𝖦𝗋𝗂𝗆𝖾𝗌 ⏳@Grimezsz

I didn’t see the doc yet - but I do think the best “don’t panic” people don’t rly do interviews because most of the best arguments against doomers (who I think are very logical except with regards to their own branding) essentially come down to opinions on the nature of life that the vast majority of people will not like or be ok with. Like I’ve had people make very strong arguments to me about why zero care should be spent addressing doomer concerns and it basically comes down to things like human life isn’t particularly special in the context of intelligence, or the philosophies of the people building ai are based on such and such superior cultural approach that I trust more than the current one etc etc obviously that’s extremely basic but - I think the reason we don’t hear these arguments in public is because they tend to end up being lik well a bunch of ppl r gna be poor in the short term and it’ll be awful and it’s gna be a bad bottleneck time or the cybernetic system deserves priority over individuals hence a certain amount of suffering death and merging and possibly even extinction is ok I don’t see these types of arguments ever risking being subjected to actual debate or rigor from the opposition. It’s pretty crazy there aren’t more well documented well planned earnest formal debates between the best doomers and the best optimists with fact checking - and mebe it’s not a totally formal debate cuz I want people to debate their side and I want optimists to have to stand up to the most meticulous scrutiny and if it still stands then awesome, Same for doomers idk It’s insane that it never actually comes down to being person v person. There’s almost no reason to do any more docs or have any more discussions if we can’t do that because it’s just ppl yelling unchallenged arguments back and forth This is too long sorry

English
4
4
60
8.8K
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
Funny to check in here only to see some guy crashing out about how the recent protest was staged His evidence? My hand drawn poster is too nice A bit out of the loop recently (long break to finish/recover from PhD), but appreciate all those who were out there on Sat protesting!
Christopher Fryant@cfryant

@MichaelSteidel You spot a difference between these two signs?

English
11
2
92
3.3K
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
@limsanity23 @ilex_ulmus I mean the overlapping venn diagram of “cares deeply about AI safety” and “has art training” is pretty small… but not zero!
English
1
0
3
71
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
@MichaelTrazzi Maybe next time I’ll film a sign making tutorial since Christopher is so incredulous 😂
English
1
0
6
108
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
@copykat510 Nice! Yeah same… some people care about averting human extinction and I don’t see how this is a tough one to grasp. Like it doesn’t even require looking beyond one’s self-interest to care here!
English
0
0
3
40
CopyKate
CopyKate@copykat510·
@thegartsy I can confirm I was not paid to participate and have volunteered hundreds of hours in the name of this cause. Christopher needs to touch grass ;-)
English
1
0
5
46
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
@sergeantsup I am definitely taking it as a compliment! My art teacher mom will get a kick out of this
English
0
0
3
45
kip
kip@sergeantsup·
@thegartsy it's actually really funny how complimentary this is, he thinks you're so good that you'd never work for free
English
1
0
8
117
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
@cfryant @MichaelSteidel I'm sorry my sign making skills are too "professional" but I can confirm I was an unpaid volunteer with PauseAI when I made the sign. I also do paid graphic design as a side gig if this fuss is because you are looking for a designer :) x.com/thegartsy/stat…
Tess Hegarty 🔸@thegartsy

Let’s not end up like the frog getting boiled alive. Try to feel every degree by which the temperature of the water increases. Pay attention. Follow the money. Ask the big questions. Don’t stop demanding answers.

English
1
0
3
24
sarah
sarah@littIeramblings·
good afternoon I bought this new jumper and am feeling delusional enough to think it’s giving rory gilmore
sarah tweet media
English
12
0
48
2.5K
Tess Hegarty 🔸 retweetledi
Jonathan Mannhart 🔎🔸
“I'd also like to acknowledge my students (…) they've gone on to do many great things. I'm particularly proud of the fact that one of my students fired Sam Altman.“ 😳🫡
English
91
669
7.8K
835.1K
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
@CJKRaymond @jachiam0 @gabriel_weil @TheZvi I feel like anyone who works at OpenAI should hardly be surprised if people are extra skeptical about OpenAI's commitment to alignment - it's nothing personal, but there's so little reason for public trust at this point, with all the drama and broken promises.
English
1
0
5
183
Cameron Raymond
Cameron Raymond@CJKRaymond·
@jachiam0 @gabriel_weil @TheZvi Annoying that this is towards one of the most good faith people in the space, who engages with folks from all different perspectives (EA, FAccT, industry plants) @jachiam0 😤😤😤😤
English
1
0
5
459
Darwin.D.W
Darwin.D.W@67darwin·
@thegartsy @LocBibliophilia @acohenNY @SpeakerPelosi That’s a very optimistic view of the law. Laws are designed in a way that it can be interpreted differently. Lawyers don’t need to exist if laws are comprehensive to begin with. If it’s just cost, just put cost in there. It says “which”, not “and”.
English
2
0
0
76
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
Am I reading this correctly? Sure reads like @speakerpelosi is insulting my intelligence in this messaging. Claiming this bill favored big tech is ridiculous. It’s a nice sounding justification, but sure doesn’t make it true. We know the real reasons.
Nancy Pelosi@SpeakerPelosi

AI springs from California. Thank you, @CAgovernor Newsom, for recognizing the opportunity and responsibility we all share to enable small entrepreneurs and academia – not big tech – to dominate. gov.ca.gov/2024/09/29/gov…

English
8
5
103
14K
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
@binarybits I have two theories on this: 1) too poorly informed to understand basic functions of how our government works (awkward because I learned this in high school) 2) they know it’s unreasonable but complain anyway because they lack integrity
English
0
0
5
253
Timothy B. Lee
Timothy B. Lee@binarybits·
It drives me crazy when republicans say “if Kamala Harris favored X, she should have implemented that policy as vice president.” The vice president’s only power is the ability to cast tie-breaking votes in the senate.
English
37
19
184
59.2K
Tess Hegarty 🔸
Tess Hegarty 🔸@thegartsy·
@67darwin @LocBibliophilia @acohenNY @SpeakerPelosi It’s fully relevant, the dollar threshold is in the same sentence as the FLOP threshold, and from that phrasing, both must be true for the law to apply. I can’t tell if this is a reading comprehension issue? Or what is going on here?
English
1
0
1
56
Darwin.D.W
Darwin.D.W@67darwin·
@LocBibliophilia @thegartsy @acohenNY @SpeakerPelosi This is the bloody thing I'm talking about. What kind of freaking idiot will put this in the legislation? THIS is what I'm talking about when I mention Moore's Law. I never once mentioned $$$ amount spent. Your argument is irrelevant to what I'm saying.
Darwin.D.W tweet media
English
1
0
0
49
Tess Hegarty 🔸 retweetledi
Siméon
Siméon@Simeon_Cps·
Now that AIs can help build bioweapons, maybe it's a good time to start checking how good AI companies are at risk management? SaferAI did just that. TLDR: 1. no company is good enough at risk management. 2. there’s still substantial variability between actors.
Siméon tweet media
Andrew Chow@andrewrchow

A new study endorsed by @Yoshua_Bengio finds that the risk management practices of most major AI labs are "very weak." xAI came in last, with a 0/5 rating. time.com/7026972/safera…

English
7
39
125
33.7K