Reposted by Emily M. Bender
Thanks to @darakerr.bsky.social for talking with me about the high climate cost of generative AI, Google's reneging on its climate goals, and the real existential crisis facing us on NPR Morning Edition today.
www.npr.org/2024/07/10/n...
0 replies
11 reposts
25 likes
Some nice reporting from @404media.co on yet another ridiculous "we asked ChatGPT study".
www.404media.co/ai-finds-tha...
For a thorough take-down of the previous iteration of this, see ep 25 of Mystery AI Hype Theater 3000, w/@alexhanna.bsky.social:
www.buzzsprout.com/2126417/1441...
2 replies
11 reposts
23 likes
What could possibly go wrong?
4 replies
16 reposts
63 likes
TFW you get to have a conversation with Prof. @safiyanoble.bsky.social and it's recorded so other people can enjoy too!!
0 replies
8 reposts
22 likes
TFW you get to have a conversation with Prof. @safiyanoble.bsky.social and it's recorded so other people can enjoy too!!
0 replies
8 reposts
22 likes
MAIHT3k Ep 35: AI Overviews and Google's AdTech Empire
… in which the magnificent Prof. @safiyanoble.bsky.social joined me and @alex.bsky.social to situate the problems with Google's AI overviews in the context of Google's history of bad practices.
Thx to @whatulysses.bsky.social for production!
2 replies
7 reposts
8 likes
Also available as video on PeerTube:
peertube.dair-institute.org/w/veZW69ZkfN...
0 replies
2 reposts
2 likes
Also available as video on PeerTube:
peertube.dair-institute.org/w/veZW69ZkfN...
0 replies
2 reposts
2 likes
MAIHT3k Ep 35: AI Overviews and Google's AdTech Empire
… in which the magnificent Prof. @safiyanoble.bsky.social joined me and @alex.bsky.social to situate the problems with Google's AI overviews in the context of Google's history of bad practices.
Thx to @whatulysses.bsky.social for production!
2 replies
7 reposts
8 likes
Reposted by Emily M. Bender
When prompted about current events, chatgpt “repeatedly cited a sloppy scam news site that deluges the user with fake software updates and virus warnings.”
5 replies
44 reposts
103 likes
This feels like a 2024 version of dreams about accidentally going to school naked...
1 replies
0 reposts
6 likes
Reposted by Emily M. Bender
One for the syntax folk
🐦🐦
31 replies
347 reposts
1376 likes
Alan Borning, Batya Friedman and Nick Logler wrote a great piece on this:
cacm.acm.org/research/the...
0 replies
0 reposts
1 likes
Reposted by Emily M. Bender
AI the tool has technical issues and application avoidances that o find fascinating( AI’s regulatory possibilities are myriad but never proposed )
But AI the philosophy is what terrifies me
It seeks truth without provenance and creativity with no viewpoint
3 replies
19 reposts
61 likes
Reposted by Emily M. Bender
The Daily Zeitgeist Podcast with Jack O'Brien and Miles Gray "Down The Stupid AI Rabbithole 06.25.24" @emilymbender.bsky.social @alexhanna.bsky.social www.iheart.com/podcast/105-...
0 replies
2 reposts
2 likes
The latest from the MAIHT3k newsletter, on the energy demands of "AI" and the folly of believing Big Tech's empty promises about carbon zero
buttondown.email/maiht3k/arch...
0 replies
5 reposts
6 likes
Reposted by Emily M. Bender
"We can, however, be aware of what's happening, and push our elected officials for better climate policy—and know not to fall for Big Tech's promises."
To help with this work, I've been gathering articles/other media at Against AI and Its Environmental Harms: pad.riseup.net/p/Against_AI...
0 replies
7 reposts
5 likes
The latest from the MAIHT3k newsletter, on the energy demands of "AI" and the folly of believing Big Tech's empty promises about carbon zero
buttondown.email/maiht3k/arch...
0 replies
5 reposts
6 likes
Today!
0 replies
1 reposts
4 likes
I'm sorry to report that one of the articles (GPTs are GPTs) we discuss in this episode has been published in Science. I have tons of respect for AAAS but this is a huge miss. On the upside, if you'd like to be entertained for an hour about why, here you go:
w/@alexhanna.bsky.social
1 replies
5 reposts
13 likes
Join us tomorrow!
0 replies
0 reposts
1 likes
Mystery AI Hype Theater 3000 Episode 34: Senate Dot Roadmap Dot Final Dot No Really Dot Docx
... in which @alexhanna.bsky.social and I dive into Sen Schumer's AI report and are dismayed at what happens when our electeds listen to the hype.
Thanks to @whatulysses.bsky.social for production.
1 replies
7 reposts
12 likes
Reposted by Emily M. Bender
imagine where "AI is biased/harmful/ineffective until it's proven otherwise" is the norm instead of the current hubris where excitement and faith at empty promises have become normal
1 replies
31 reposts
78 likes
Also available as video on PeerTube!
peertube.dair-institute.org/w/jqppykvcrD...
0 replies
0 reposts
1 likes
Mystery AI Hype Theater 3000 Episode 34: Senate Dot Roadmap Dot Final Dot No Really Dot Docx
... in which @alexhanna.bsky.social and I dive into Sen Schumer's AI report and are dismayed at what happens when our electeds listen to the hype.
Thanks to @whatulysses.bsky.social for production.
1 replies
7 reposts
12 likes
Next Mystery AI Hype Theater 3000 live-stream!
@al2.bsky.social will join me and @alexhanna.bsky.social to take apart yet another paper-shaped AI hype artifact unleashed from a "top" industry lab (could that be the summit of bullshit mountain?):
Monday June 24
9:30am Pacific
0 replies
6 reposts
7 likes
Next Mystery AI Hype Theater 3000 live-stream!
@al2.bsky.social will join me and @alexhanna.bsky.social to take apart yet another paper-shaped AI hype artifact unleashed from a "top" industry lab (could that be the summit of bullshit mountain?):
Monday June 24
9:30am Pacific
0 replies
6 reposts
7 likes
Next Mystery AI Hype Theater 3000 live-stream!
@ali@masto.al2.in will join me and @alex to take apart yet another paper-shaped AI hype artifact unleashed from a "top" industry lab (could that be the summit of bullshit mountain?):
Monday June 24
9:30am Pacific
www.twitch.tv/dair_institute
0 replies
0 reposts
2 likes
Really enjoyed this conversation with Kobi Leins on Carnegie Council's AIEI podcast:
www.carnegiecouncil.org/media/series...
1 replies
6 reposts
9 likes
Thank you!
1 replies
0 reposts
1 likes
Reposted by Emily M. Bender
Really excellent conversation.
1 replies
1 reposts
3 likes
Really enjoyed this conversation with Kobi Leins on Carnegie Council's AIEI podcast:
www.carnegiecouncil.org/media/series...
1 replies
6 reposts
9 likes
Reposted by Emily M. Bender
Noble's Algos of Oppression was 2018 & joined a living conversation, but we're still nowhere because a bunch of mostly white dudes w/ cartoon dollar signs where their critical thinking skills should be keep slathering themselves in LLM/GPT/"AI" & running down the street screaming "We Are The Future"
3 replies
41 reposts
144 likes
Reposted by Emily M. Bender
Basically every person who thought about this for even 10 seconds warned google this was going to happen. Google's entire incentive structure is monetization-via-ranking-via-SEO-&-clicks. People are clicking "AI" porn, & people Making "AI" porn are manipulatively lying in their SEO to drive traffic
10 replies
144 reposts
410 likes
PSA: Every time you cite the arXiv version of something instead of the peer reviewed version, you're lending legitimacy to nonsense like this:
Source: arxiv.org/html/2404.07...
3 replies
14 reposts
55 likes
PSA: Every time you cite the arXiv version of something instead of the peer reviewed version, you're lending legitimacy to nonsense like this:
Source: arxiv.org/html/2404.07...
3 replies
14 reposts
55 likes
And the story about seeing a robot have a feeling back in 1972 ... I wonder if he really thought so back then too (has always been this far down the rabbit hole) or if that's some revisionist history about his own thoughts.
0 replies
0 reposts
3 likes
That is so much nonsense. Also, is it really a new thing from this month? "Letter last week" sounds like 2023, but the date is June 12, 2024, so I'm super confused.
2 replies
0 reposts
4 likes
Reposted by Emily M. Bender
📢📢📢CRITICAL AI PROUD TO PREVIEW INTRO TO SPECIAL ISSUE ON LLMS, GEN AI AND THE RISE OF CHATBOTS. Abstract and snips below /1 📢📢📢
criticalai.org/2024/06/06/s...
1 replies
25 reposts
25 likes
Reposted by Emily M. Bender
Put this whole paragraph on repeat.
6 replies
270 reposts
478 likes
In the latest installment of the Mystery AI Hype Theater 3000 newsletter: Piles of Data Don't Add Up to AI, But They Do Add Up to Trouble
buttondown.email/maiht3k/arch...
0 replies
6 reposts
12 likes
In the latest installment of the Mystery AI Hype Theater 3000 newsletter: Piles of Data Don't Add Up to AI, But They Do Add Up to Trouble
buttondown.email/maiht3k/arch...
0 replies
6 reposts
12 likes
Today!
0 replies
2 reposts
3 likes
Join us tomorrow!
0 replies
0 reposts
2 likes
Reposted by Emily M. Bender
“Sarah’s not allowed to discuss politics.” 💀
4 replies
18 reposts
59 likes
Ready for more Mystery AI Hype Theater 3000?
Join me & @alex.bsky.social on June 10, when we get to chat with Prof. @safiyanoble.bsky.social about Google's AI Overviews -- in the long context of Google's warping of information access.
Monday June 10, noon Pacific
www.twitch.tv/dair_institute
1 replies
8 reposts
17 likes
Ready for more Mystery AI Hype Theater 3000?
Join me & @alex.bsky.social on June 10, when we get to chat with Prof. @safiyanoble.bsky.social about Google's AI Overviews -- in the long context of Google's warping of information access.
Monday June 10, noon Pacific
www.twitch.tv/dair_institute
1 replies
8 reposts
17 likes
Setting aside the issue of whether "textualism" is a good idea, corpus linguistic methods based on an entirely unknown and undocumented corpus are unsound. And "asking ChatGPT" also isn't corpus methods.
0 replies
0 reposts
6 likes
That's good to hear! Thank you.
0 replies
0 reposts
1 likes
Mystery AI Hype Theater 3000, Ep 33 Much Ado About 'AI' 'Deception', in which guest co-host @mmitchell.bsky.social and I find it's really the AI doomers deceiving themselves with motivated thinking.
www.buzzsprout.com/2126417/1518...
Thanks to @whatulysses.bsky.social for production!
1 replies
9 reposts
9 likes