On AI & water usage, it looks like all US data center usage (not just AI) ranges from 628M gallons a day (counting evaporation from dam reservoirs used for hydro-power) to 200-275M with power but not dam evaporation, to 50M for cooling alone. So not nothing, but also a lot less than golf courses.

A 20/80 rule for AI in education. And what that looked like yesterday in my classroom. [An interesting example of how a teacher is using AI in the classroom.] aiwaypoints.substack.com/p/a-2080-rul...

Meta Ray-Ban Display Sells Out As Early Adopters Struggle To Buy https://www.uploadvr.com/meta-ray-ban-display-sold-out-early-adopters-struggle/ #AI #glasses

they literally gave sonnet 4.5 context anxiety cognition.ai/blog/devin-s...

PayPal's Honey will now with with ChatGPT to find shopping deals.

OpenAI is launching a social app where you can generate deepfakes of you and your friends and obviously nothing will go wrong here at all ever techcrunch.com/2025/09/30/o...

The entire structure of how OpenAI presents ChatGPT is deceptive. There’s nothing wrong with having safety precautions but this is basically a magic trick where they are attempting to both foist an experimental language model on consumers and have them humanize it the-decoder.com/chatgpt-quie...

OpenAI’s New Sora Video Generator to Require Copyright Holders to Opt Out [This is not how copyright works.] www.wsj.com/tech/ai/open...

Buy it in ChatGPT: Instant Checkout and the Agentic Commerce Protocol openai.com/index/buy-it...

Beyond AI Deepfakes & the Uncanny Valley https://nettricegaskins.medium.com/beyond-ai-deepfakes-the-uncanny-valley-4af4c0036561 #AI #deepfakes #history

New from 404 Media: we've closely been following the wave of lawyers getting caught using AI. But now we know why: here 18 lawyers explain why they used AI. They blame IT, family emergencies, their own poor judgment, their assistants, illness, and more. www.404media.co/18-lawyers-c...

Paper: Towards an AI-Augmented Textbook ( arxiv.org/abs/2509.13348 )

French filmmaker groups have penned an open letter opposing OpenAI’s plan to debut an AI-generated animated film at Cannes, urging the protection of human-led artistry. www.cartoonbrew.com/feature-film...

Are deepfakes reshaping our news reality? Explore how AI transforms satire and journalism in today’s media landscape. #Deepfakes https://www.theguardian.com/technology/2025/sep/26/deepfakes-ai-slop-now-part-of-news-cycle-south-park-v-tru#Deepfakesp

Spotify removed over 75 million tracks deemed 'spammy' in an AI music crackdown. How much do you think AI will change music? #MusicIndustry https://www.musicbusinessworldwide.com/spotify-has-deleted-75m-spammy-tracks-as-it-unveils-new-ai-music-policies/

A new test from OpenAI aims to understand how close AI is to outperforming humans at economically valuable work.

Replacing humans with AI is harder than it seems. "The algorithm will see you now" worksinprogress.co/issue/the-al...

SOURCES: OpenAI is planning to bring ads to ChatGPT. Also: In a message to employees, Sam Altman says he wants 250 gigawatts of compute by 2033. He calls OpenAI's team behind Stargate a "core bet" like research / robotics. "Doing this right will cost trillions." sources.news/p/openai-ads...

It was inevitable. ChatGPT-selected stocks beating professional money managers. @reuters.com www.reuters.com/business/fin...

how likely a bond will trade at a given price. HSBC and the US technology giant applied quantum processing to an anonymized set of European bond trading data and found it could significantly enhance the efficiency of the market." www.bloomberg.com/news/article...

UK's new AI tool has recovered nearly £500m lost to fraud in just a year. How do you think AI can further tackle financial crimes? #FraudPrevention https://www.bbc.com/news/articles/cpd92gpld0g#FraudPrevention

💥New: ChatGPT is blind to bad science ✍️ @yelusikuata.bsky.social & @mikethelwall.bsky.social #AcWri #SciComm #ChatGPT #Retractions

Excited about Facebook's new AI dating assistant? How do you think it’ll change the dating game for users? #AIDating https://techcrunch.com/2025/09/22/facebook-is-getting-an-ai-dating-assistan#AIDating/

AI can walk through all the doors at once - Mike Caulfield https://mikecaulfield.substack.com/p/ai-can-walk-through-all-the-doors (& a fun Letterman clip) #AI #usefulness

Are teen innovators shaping the future of tech in San Francisco? Discover their role in the AI gold rush! What do you think? #AIFuture https://nymag.com/intelligencer/article/san-francisco-ai-boom-artificial-intelligence-tech-industry-kids.html

Curious about how The New York Times and The Washington Post are leveraging AI in journalism? Discover their innovative strategies! #AIinJournalism https://www.niemanlab.org/2025/09/a-look-inside-the-ai-strategies-at-the-new-york-times-and-the-washington-post/

Not only is solar more than capable of supplying all the world’s energy, in the long term it’s the only power source that won’t fry the planet.

Wow, I'm just seeing that the judge denied the class-action settlement in the Anthropic copyright case! "Without prejudice" means the parties can "file another motion to fill in the details Judge Alsop found lacking." chatgptiseatingtheworld.com/2025/09/08/s...

I'm grateful to the folks at IBM for platforming our work assessing impacts of genAI on the fed workforce. This work was designed before the draconian, ham-fisted cuts that were (partly) based on an absurd faith in AI as a replacement to labor. This work refutes that premise in full relief.

The previously sensible advice to never use ChatGPT for search needs to be rethought - GPT-5 in thinking mode is shockingly good at running searches now simonwillison.net/2025/Sep/6/r...

Are rising electricity bills a price we pay for AI’s growth? Explore how massive data centers are impacting our wallets! #AIBusiness https://www.pbs.org/newshour/show/how-ai-infrastructure-is-driving-a-sharp-rise-in-electricity-bil#AIBusinesss

California Attorney General Rob Bonta and Delaware Attorney General Kathy Jennings met with and sent an open letter to OpenAI to express their concerns over the safety of ChatGPT, particularly for children and teens. 

If a teacher is going to try some form of AI assistance in their teaching process and/or invite students to engage with AI in some way, which options should they consider first? I shared some possibilities with faculty at West Valley College today. Here are my slides: link.annarmills.com/WestValley

Is AI truly a threat or are we overreacting? Eliezer Yudkowsky shares his fears and solutions. What’s your take? #ArtificialIntelligence https://www.wired.com/story/the-doomers-who-insist-ai-will-kill-us-al#ArtificialIntelligence/

Why language models hallucinate - OpenAI https://openai.com/index/why-language-models-hallucinate/ (new paper) #AI #hallucination

A skeptic’s take on AI electricity load growth https://www.latitudemedia.com/news/catalyst-a-skeptics-take-on-ai-electricity-load-growth/ (good podcast episode) #AI #energy

READ HN 📖 news about AI, LLM and NLP: AI web crawlers are destroying websites in their never-ending content hunger https://www.theregister.com/2025/08/29/ai_web_crawlers_are_destroying/ #technews #NLP #genAI

Green and intelligent: the role of AI in the climate transition https://www.nature.com/articles/s44168-025-00252-3.pdf (excellent) #AI #ClimateTransition

I'm up to 124 little mostly-vibe-coded apps on my tools.simonwillison.net site and most of them are trivial but some of them aren't, so I wrote up this guide to the ones I think are most worth checking out simonwillison.net/2025/Sep/4/h...

AI operating behind the scenes in grocery stores is significantly reducing food waste www.linkedin.com/pulse/waste-...

Harvard’s Generative AI Policy Is Inequitable | Opinion | The Harvard Crimson https://www.thecrimson.com/article/2025/9/3/siddiqui-harvard-ai-inequitable/ (thoughts from a student) #AI #education #equity

New on the GAIN blog today from @clarespencer.bsky.social . Local journalists are finding value in using GenAI tools to monitor public meetings and find important stories. More here: generative-ai-newsroom.com/how-genai-to...

LLM traffic converts about the same as organic search: Research LLM referrals convert no better than search and make up less than 1% of traffic, challenging claims of higher-quality clicks, new data shows. #seo #smm

A beautiful articulation of why LLMs are not good at writing. Link: www.theinterference.io/p/writing-is...

ChatGPT's Projects feature for organizing AI chats will now be available to free users as part of a recent update.

The rumour is OpenAI and co. are seeking revenue splits from retailers as a way of making their products profitable. This raises big concerns about dark patterns and online manipulation. A whole new frontier for consumer protection. Some of my thoughts below. theconversation.com/openai-looks...

Could AI truly elevate living standards as they claim? What are your thoughts on the future of AI and the economy? #ArtificialIntelligence https://www.ft.com/content/887531bf-646a-46b5-a821-649d3948a574