Reposted by Matt BeaneEthan Mollick @emollick.bsky.social · 15/05/2025This includes many of my papers, too. The point I am making is the findings in careful academic research likely represents a lower bound of AI capabilities at this point. 3504
Reposted by Matt BeaneClive Thompson @clivethompson.bsky.social · 04/02/2025I can’t i just … i can’t www.404media.co/anthropic-cl... 411202356
Matt Beane @mattbeane.bsky.social · 30/01/2025I bet if someone *has* succeeded, it's via spinning up an elicitation-GPT that just drilled you for critical intel, wouldn't let you weasel out via under/overspecified output, then dumped it all back to you in standardized format so you could think faster - basically exporting your extraction algo. 010
Matt Beane @mattbeane.bsky.social · 29/01/2025Exactly. If we overheard Dario, Sam, and Demis chatting about certain well known AI critics, I'd be willing to bet they'd be expressing gratitude. Proving a grouch wrong is a real motivator. 000
Reposted by Matt BeaneDaniel Rock @danielrock.bsky.social · 29/01/2025Hi Everyone! We're hosting our Wharton AI and the Future of Work Conference on 5/21-22. Last year was a great event with some of the top papers on AI and work. Paper submission deadline is 3/3. Come join us! Submit papers here: forms.gle/ozJ5xEaktXDE...forms.gle 21515
Matt Beane @mattbeane.bsky.social · 15/01/2025Exciting new hobby project in the offing related to AI and skill. Involves a childhood passion, a wild leap into the unknown, made real via an order from Amazon just now. Will be 100% cool, I will be documenting things, sharing eventually. Feels like April 2023 again! 020
Matt Beane @mattbeane.bsky.social · 13/01/2025The Silo is so good. Just superb. This generation's answer to the BSG remake. 020
Reposted by Matt BeaneRodney Brooks @rodneyabrooks.bsky.social · 09/01/2025My hobby horse. You can simulate a rocket all you want, and use more energy on computation than the actual rocket would, but you won't get to orbit until you ignite rocket fuel. What if all the energy we are spending on simulating learning is not the juice we really need to make intelligence? 85811
Reposted by Matt BeaneSimon Willison @simonwillison.net · 31/12/2024Here's my end-of-year review of things we learned out about LLMs in 2024 - we learned a LOT of things simonwillison.net/2024/Dec/31/... Table of contents: 28648148
Reposted by Matt BeaneJaime Teevan @teevan.bsky.social · 31/12/2024In 2024 we learned a lot about how AI is impacting work. People report that they're saving 30 minutes a day using AI (aka.ms/nfw2024), and randomized controlled trials reveal they’re creating 10% more documents, reading 11% fewer e-mails, and spending 4% less time on e-mail (aka.ms/productivity...). 1174
Reposted by Matt BeaneEthan Mollick @emollick.bsky.social · 20/12/2024Independent evaluations of OpenAI’s o3 suggest that it passed math & reasoning benchmarks that were previously considered far out of reach for AI including achieving a score on ARC-AGI that was associated with actually achieving AGI (though the creators of the benchmark don’t think it o3 is AGI) 1314130
Matt Beane @mattbeane.bsky.social · 20/12/2024Just *one* of the reasons that Blindsight was ahead of its time. Way ahead. 110
Reposted by Matt BeaneRita McGrath @rgmcgrath.bsky.social · 09/12/2024Join me by the fireside this Friday with Matt Beane as we dive into one of today’s biggest workforce challenges: upskilling at scale. 📈 Linke below to hear the full discussion on Friday, December 13 at 11 am EST! linktr.ee/RitaMcGrath @mattbeane.bsky.social 142
Matt Beane @mattbeane.bsky.social · 07/12/2024I propose a workshop. Most engineers/CS working on AI presume away well established, profound brakes on AI diffusion. Most social scientists presume away how AI use could reshape those brakes. Let's gather these groups, examine these brakes 1-by-1, make grounded predictions. 020
Reposted by Matt BeaneEthan Mollick @emollick.bsky.social · 07/12/2024Models like o1 suggest that people won’t generally notice AGI-ish systems that are better than humans at most intellectual tasks, but which are not autonomous or self-directed Most folks don’t regularly have a lot of tasks that bump up against the limits of human intelligence, so won’t see it 815526
Matt Beane @mattbeane.bsky.social · 04/12/2024Grateful for the opportunity to visit and learn from the professionals at the L&DI conference. And very glad to hear you found my talk so valuable, Garth! Means a lot. 211
Reposted by Matt BeaneTom Williams @tomwilliams.phd · 03/12/2024I made an HRI Starter Pack! If you are a Human-Robot Interaction or Social Robotics researcher and I missed you while scrolling through bsky's suggestions, just ping me and I'll add ya. go.bsky.app/CsnNn3s 114214
Matt Beane @mattbeane.bsky.social · 03/12/2024Wrote a little something on this in 2012, though I didn't anticipate the main reason for hiring such workers - training data. www.technologyreview.com/2012/07/18/1...technologyreview.comThe Avatar EconomyAre remote workers the brains inside tomorrow’s robots? 010
Matt Beane @mattbeane.bsky.social · 03/12/2024David Meyer (v.) /ˈdeɪvɪd ˈmaɪ.ər/ To attribute complex, intentional design or deeper meaning to simple emergent behaviors of large language models, especially when such behaviors are more likely explained by straightforward technical constraints or training artifacts. 020
Matt Beane @mattbeane.bsky.social · 03/12/2024They did NOT. Wow. Sign of the times. And I can verify on your rule! I was so flabbergasted and honored. Your feedback was rich and so helpful. Remain grateful. 010
Matt Beane @mattbeane.bsky.social · 30/11/2024I remember *treasuring* the previews. I'd fight to get there on time. Was part of the thrill. But ads? F*ck that noise. Seriously, straight up evil. 100
Matt Beane @mattbeane.bsky.social · 30/11/2024Never occurred to me there'd be an algo under the hood that could reliably learn to provide content I'd value more than a straight read of my hand-curated list of people. My solution has been following people if they post high signal stuff all the time. 120
Matt Beane @mattbeane.bsky.social · 30/11/2024I have never used the feed page. What a horror, can't quite understand why folks would try. Only/ever the "following" page. Even there things got pretty intolerable towards/around the election, now settled down. 110
Reposted by Matt BeaneBob Sutton @bobsutton.net · 27/11/2024My Thanksgiving post. A Kurt Vonnegut poem. He talks with Joe Heller (Catch 22 fame) about a billionaire. Key part: Joe said, "I've got something he can never have" And I said, "What on earth could that be, Joe?" And Joe said, "The knowledge that I've got enough" www.linkedin.com/pulse/kurt-v...linkedin.comKurt Vonnegut, Joe Heller, and How to Think Like a MenschThis story remains my favorite Thanksgiving message; it reminds me to be grateful for what I have and of the evils of jealousy and destructive competition. I first posted it on my work matters blog mo... 0122
Matt Beane @mattbeane.bsky.social · 25/11/2024Couldn't agree more. That conclusion struck me as a bit odd. I think this is more a status effect than anything else. 100
Matt Beane @mattbeane.bsky.social · 24/11/2024Not every day your work gets a healthy mention in the Sunday @nytimes.com! The software talent market went into freefall in July of 2022. Sarah Kessler takes us inside the maelstrom by investigating the impact on graduates of coding bootcamps. Great read. www.nytimes.com/2024/11/24/b...nytimes.comDo Coding Boot Camps Make Sense in an A.I. World?Coding boot camps once looked like the golden ticket to an economically secure future. But as that promise fades, what should you do? Keep learning, until further notice. 121
Reposted by Matt BeaneEthan Mollick @emollick.bsky.social · 23/11/2024Easy to get the wrong impression around here, but when you actually survey students, teachers, and parents they love AI. In the survey, it is people who never used it who don’t like it. www.waltonfamilyfoundation.org/learning/the... 812317
Matt Beane @mattbeane.bsky.social · 22/11/2024We agree there. I have mixed feelings about this one. On some levels it's so thoughtful - the first one I've seen that even acknowledges scenarios, for example! On other levels, well... it just makes some strange assumptions about diffusion. e/acc folks are going to run with this one hard... 010
Matt Beane @mattbeane.bsky.social · 20/11/2024media.tenor.combender from futurama says " right here buddy " on a blue backgroundALT: bender from futurama says " right here buddy " on a blue background 010
Matt Beane @mattbeane.bsky.social · 19/11/2024Head to head twitter/bluesky social science test: What are your go-to, recent empirical papers on surveillance and technology? 000
Reposted by Matt BeaneMae Nick Phra Khanong @nickjbrumfield.bsky.social · 19/11/2024Culture is definitely gonna play a part, but architecture is going to be key to creating a positive social media environment Bluesky has given us so many tools like this to cut out all the crap that made even pre-Elon Twitter so toxic 02710
Matt Beane @mattbeane.bsky.social · 18/11/2024THIS is a HUGE part of the reason why it's hard to find a satisfactory physician. The other = healthspan promoting v pathology mitigating mindset. Tiny, tiny slice of medical humanity meets the nonpatriarchal * functional medicine spec. 010
Matt Beane @mattbeane.bsky.social · 18/11/2024When people ask me which model to use for writing, I say Claude. Have for months. This is now my go-to example for explaining why. To get why Claude is better, first read and get why Claude is better. To break this recursion, just compare their explanations of recursion - you'll see. 000
Matt Beane @mattbeane.bsky.social · 18/11/2024Oh Claude is *far* better. I will use this as my "Claude's a better writer" example until further notice. 000
Reposted by Matt Beane🎃 Decorative Gourd Mikell 🎃 @mikell.bsky.social · 17/11/2024I made a list of some of my favorite robotics people to follow, if any of former RoboticsTwitter is still looking for folks over here bsky.app/profile/did:... 082
Matt Beane @mattbeane.bsky.social · 15/11/2024What a lovely gift on a Friday: using LLMs (admittedly, by playing to stereotypes) to confound scammers through a convincing "granny" avatar that will chat their time away. Vid worth watching for a laugh. Not all disinformation harms the consumer! news.virginmediao2.co.uk/o2-unveils-d...news.virginmediao2.co.ukO2 unveils Daisy, the AI granny wasting scammers’ time - Virgin Media O2O2 has today unveiled the newest member of its fraud prevention team, 'Daisy'. As ‘Head of Scammer Relations’, this state-of-the-art AI Granny's mission is to talk with fraudsters and waste as much of... 020
Matt Beane @mattbeane.bsky.social · 11/11/2024Haven't! Magnificent! And god help me a few days reading responses to your post and my entire reading list is in disarray... Going to take me a year to dig my way out, but it will be joyful work. 010
Matt Beane @mattbeane.bsky.social · 11/11/2024Couldn't agree more. Horrifying and fascinating. Razor sharp. 010
Reposted by Matt BeaneChristopher Mims @mims.bsky.social · 10/11/2024!!! US power grid added battery equivalent of 20 nuclear reactors in past four years www.theguardian.com/environment/... 6788
Reposted by Matt BeanePat Loika @patloika.bsky.social · 10/11/2024Happy Circulatory System Walking Through The Kitchen Day to those who celebrate. 2223596810385
Matt Beane @mattbeane.bsky.social · 10/11/2024The action on Eugene's post - and his post to begin with - have instantly convinced me that @bsky.app is beautiful. I am here for it. And will not be bringing my typical technotwitter self here. More nerd/geek/xkcd, more fun/play/good cheer, more aggro sharing of wondrous delights. Woot! 000