The Feed
Every curated post, newest first. Sources are linked verbatim — nothing here is rewritten or interpreted by a news outlet.
ChatGPT Ads expands across Europe — ChatGPT Ads is expanding to 31 European markets. Learn how advertisers can reach people as they explore, compare options, and make decisions.
OpenAI has expanded ChatGPT Ads to thirty-one European markets, enabling advertisers to reach users during their decision-making processes.
Mojo🔥 is now open source — Mojo🔥 is now open source Mojo🔥 is now open source The Mojo programming language has been promising an open source release since May 2023 . Last week they shipped their 1.0 and today they have followed through on that original promise, releasing the compiler and toolchain under an Apache 2 license. When Mojo first launched the stated goal was to produce a superset of Python, so existing Python code could be used to bootstrap their own ecosystem. That plan changed around August 2025 : Mojo may or may not evolve into a full superset of Python, and it’s okay if it doesn’t. We’re encouraged by how well AI-assisted coding tools already help migrate Python to Mojo today, and we’re confident that future tooling and ecosystem maturity will make this evolution even smoother. Today Mojo is its own language, optimized to make GPU programming as painless as possible using syntax inspired by Python, if not 100% compatible with existing code. Via Lobste.rs Tags: open-source , python , mojo
The Mojo programming language has officially open-sourced its compiler and toolchain under an Apache 2 license.
Google’s biggest mistake? — A theory about what went wrong, and some open questions about what might happen next
Gary Marcus analyzes Google's strategic mistakes and discusses open questions about the company's future.
Strengthening democratic oversight in national security — OpenAI launches an initiative to strengthen democratic oversight of AI in national security, supporting government institutions with tools, training, and expertise.
OpenAI has launched an initiative to strengthen democratic oversight of AI in national security by supporting government institutions.
How Much Memory Does Your Agent Actually Need?
Hugging Face investigates the amount of memory required to run AI agents.
Breaking: U.S. young adults are now more concerned about AI than enthusiastic — Brief update to last’s week’s note re: the growing backlash.
A recent update shows that young adults in the United States are now more concerned about artificial intelligence than enthusiastic.
America’s Workforce Academy: No Cost. No Prior Experience Necessary. A Career on the Other Side. — We're sharing stories from the first graduates of America's Workforce Academy, Meta's free training program that guarantees trade jobs for all graduates. The post America’s Workforce Academy: No Cost. No Prior Experience Necessary. A Career on the Other Side. appeared first on Meta Newsroom .
Meta shared stories from the first graduates of America's Workforce Academy, its free training program that guarantees trade jobs.
Partnering with CodeAI to prepare the first AI generation — OpenAI and CodeAI are partnering to help students build AI literacy, think critically about AI, and develop the skills to use and shape it responsibly.
OpenAI is partnering with CodeAI to help students develop AI literacy and skills for responsible use.
Introducing ChatGPT for Teens: Built for learning, backed by protections — ChatGPT for Teens helps teens learn, think critically, and use AI with confidence, with stronger built-in protections, healthy-use features, and additional controls for parents.
OpenAI has launched ChatGPT for Teens, featuring built-in safety protections and parental controls designed for educational use.
Pacing model development in an era of cyber-critical capabilities — OpenAI is strengthening monitoring, alignment, and security for frontier AI models. See how new safeguards are guiding the pace of model development.
OpenAI is strengthening monitoring, alignment, and security measures to guide the development pace of frontier AI models.
When a model reads a drug's class from its name—not its knowledge — Researchers used Olmo 3 and its open training data to show that models can infer a drug’s class from its name instead of knowing the specific medication, and traced that shortcut to how often drugs appeared in training.
Researchers used OLMo 3 to show that language models often infer a drug's class from its name rather than actual knowledge.
Asana cleared 5 years of engineering work in 2 weeks with Codex — Asana used OpenAI Codex to replace an outdated testing system in two weeks, completing work expected to take five years for about $12K.
Asana used OpenAI Codex to replace an outdated testing system in two weeks, completing five years of expected engineering work.
Multi-Vector (Late Interaction) Embedding Models with Sentence Transformers
Hugging Face announced support for multi-vector late interaction embedding models using the Sentence Transformers library.
Qwen 3.8 27B scores 52 on the Artificial Analysis Intelligence Index — Qwen 3.8 27B scores 52 on the Artificial Analysis Intelligence Index That's the same score as GPT-5.6 Luna (max), and just one point behind GLM-5.2 (max) and DeepSeek V4 Pro 0813 (max) - that GLM is 753B and that DeepSeek is 1.6B parameters, and Luna is size unknown but presumably a whole lot bigger than 27B. Qwen 3.8 27B is a truly astonishing model . Via Hacker News Tags: ai , generative-ai , llms , qwen , ai-in-china , artificial-analysis
Simon Willison highlights Qwen 3.8 27B scoring 52 on the Artificial Analysis Intelligence Index, matching much larger models.
Same Cluster, 33 Points More Utilization: What Changed Was the Order
Hugging Face achieved a thirty-three point increase in cluster utilization by optimizing the execution order of workloads.
We Tracked a Shipment of Rare Books. It Ended at an Amazon AI Training Facility — We Tracked a Shipment of Rare Books. It Ended at an Amazon AI Training Facility Excellent piece of reporting from 404 Media. For a while now there have been stories of book dealers receiving orders for large volumes of books from apparently price-insensitive anonymous customers, widely suspected to be companies looking to scan them for AI training (see my previous coverage of Anthropic's book scanning from June 2025.) 404 Media investigated with an AirTag! In July, one bookseller told me they received a very large order of around 1,000 books on Biblio, one of these marketplaces. The seller agreed to put an Apple AirTag provided by 404 Media in one of the books included in this order so we could see where the book was going. And by extension, which company, AI or otherwise, was behind this massive order. The book ended up delivered to the VGT3 corner of the LAS8 Amazon facility in the north east of Las Vegas, where the entrance carried this on-the-nose logo of a dinosaur with a book! Photo credit: 404 Media Online forum discussions between Amazon workers confirmed that VGT3 destructively scans large volumes of books. Tags: amazon , journalism , ai , training-data , ai-ethics , 404-media
Simon Willison shares a 404 Media report tracking a book shipment to an Amazon facility scanning books for AI training.
Teaching Everyone to Fish for Tokens — Nvidia wants you building your own model, not buying from Anthropic/OpenAI.
Nvidia is encouraging companies to build custom AI models rather than relying on proprietary systems from OpenAI and Anthropic.
No, Dario Amodei, we will not be curing cancer and “most human disease” in five to ten years — What Amodei left out
Gary Marcus disputes Dario Amodei's claim that AI will cure cancer and most human diseases within five to ten years.
Import AI 469: Science AI; RSI simulator; and Zuck's technological pessimism — The new frontier of AI is developing capable autonomous researchers
The latest Import AI newsletter discusses the development of autonomous AI researchers, scientific AI, and technological pessimism.
Securing the Infrastructure of Intelligence — AI factories are the defining infrastructure of the AI era — where compute transforms energy and data into intelligence that powers every business, industry and country. In the AI economy, compute is revenue. AI factories require a full stack of critical resources: advanced chips, packaging, memory, and networking — as well as land, power and […]
NVIDIA AI describes AI factories as critical infrastructure requiring advanced hardware, networking, land, and power to generate intelligence.
Seeing beyond BMI: Estimating cardiometabolic risk with smartphone imagery — General Science
Google Research has developed a method to estimate cardiometabolic risk using smartphone imagery instead of relying solely on BMI.
New policy ideas for the Intelligence Age — OpenAI funds 14 independent projects exploring new AI policy ideas to expand economic opportunity and strengthen societal resilience in the Intelligence Age.
Markdown SVG upgrades — I started building my markdown-svg-renderer tool in May , but I've since added enough features to it that it's worth talking about here again. It's evolved into my ideal tool for sharing Markdown transcripts that include SVG documents. Given my proclivity for drawing pelicans riding bicycles this is a problem that I needed to solve! The tool is very simple. Navigate to markdown-svg-renderer in your browser and paste in some Markdown to see it rendered... or save that Markdown to a CORS-friendly URL or a GitHub Gist and paste in a URL to that document. The URL option will give you a bookmarkable page, for example https://tools.simonwillison.net/markdown-svg-renderer#url=https%3A%2F%2Fgist.github.com%2Fsimonw%2F6f9e48293be5c916652d29f0dc0b0657 - which bakes in the URL to this Gist . If you visit the Gist you'll see raw SVG: In the rendered tool that looks like this instead: As you can see, that SVG block in the Markdown has been transformed into a rendered SVG (in this case animated) plus several tabs. The tabs are the really fun bit. The PNG and JPEG tabs render that SVG to those image formats in the browser and lets you copy or download them - useful for sharing on platforms that don't support SVG directly. The MP4 tab is new today - it examines the SVG to see if it contains any animations, attempts to guess how long the looped video should be, then renders a whole bunch of frames of the animation and loads 30+MB of ffmpeg.wasm so it can compile those frames into an MP4 video using the full power of FFMPEG compiled to WebAssembly and running in the browser. Being able to turn an animated SVG into a MP4 again makes it easy to share on platforms that can't support SVG animation natively. It's a neat trick! Tags: svg , markdown , tools
Qwen 3.8 27B is excellent, but it defaults to wildly overthinking things — Friday's big release was Qwen 3.8 27B , an Apache 2 licensed 27B parameter vision-capable LLM from Alibaba's Qwen research lab. I've been looking forward to this one: 27B is an excellent size for running a model on a reasonably specced laptop, and its predecessor Qwen 3.6 27B was impressive. Qwen's self-reported benchmarks for this model are eye-opening. They show a boost from both Qwen 3.6 27B and the closed-weight Qwen 3.7-Plus, which was one of Qwen's strongest models of any size as recently as May this year . It will be interesting to hear what independent benchmarks have to say about the model. I've been running the model on two different machines: my 128GB M5 Max MacBook Pro, and an NVIDIA DGX Spark . On both machines I'm running LM Studio and their 17GB Q4_K_M quantized build . I also tried using llama-server directly on the Spark. The default of extra high results in spectacular over-thinking Qwen's documentation describes the model as defaulting to xhigh for the reasoning effort, and the LM Studio GGUF I've been trying preserves that default: Qwen3.8 comes with official support for reasoning_effort , which can be used to adjust reasoning depth and control cost: • xhigh (default): for complex tasks demanding thorough analysis • medium : balancing accuracy and speed • low : efficient reasoning optimizing for speed and cost This is a hilarious default. It's absolutely not a good way to run the model, especially on consumer hardware. I've been finding the results extremely entertaining. I quickly ran into problems with LM Studio's default context limit of 8,192 tokens - Qwen was using them all up thinking about even the most mundane of problems. I loaded the model with the full 262,144 maximum context length and that problem went away. Here's the pelican riding a bicycle SVG I got from my first attempt with that increased context length. It took 21 minutes to generate, using 22,276 reasoning tokens to produce 3,223 tokens of output. You can read the reasoning trace here . This is by far the best pelican SVG I've been able to generate with a model that runs on a local machine - and this Qwen is pretty small, just a 17GB file on disk. There's a lot to like about this: • The bicycle frame is the right shape • It has legs on each side of the bike - that's very rare • Good, clear pelican pouch • The wings extend to touch the handlebars! • The motion lines are behind, not in front • It has a tasteful background - nice sun, clouds, hill, flowers and grass. Was that worth waiting 21 minutes for? Absolutely not. Here's that same prompt run wi
Quoting Dario Amodei — I do agree that the public has a negative view of AI (and that this is a big problem), but I don’t think it is primarily caused by me or any other AI leader warning about AI’s risks. I think it is fundamentally a crisis of trust. I think that ordinary people don’t trust companies, governments, or the tech industry and always suspect that we are cooking up some new way to screw them over. The causes of this go back decades and AI is just the latest iteration of it. I don’t think that a glitzy marketing campaign with a positive spin (which some have advocated that Anthropic do) is the way to win back that trust — at this point, saying that AI will cure cancer is more a cliche than it is inspiring, and most people think it is deceptive. The thing that will work is actually curing cancer . I think by far the most accurate criticism of AI companies including Anthropic is that we haven’t yet delivered on our big promises to benefit the world. That is totally on us, and I think it’s the criticism you should be making, instead of all this stuff about messaging and marketing. — Dario Amodei Tags: anthropic , ai , ai-backlash
CORS Chat — Tool: CORS Chat I built this today ( with GPT-5.6-Sol xhigh ) to help test Qwen 3.8 27B running in LM Studio on both my M5 MacBook Pro and an NVIDIA DGX Spark. It provides a web UI for exercising an OpenAI-Responses-compatible chat endpoint. I've tried it against LM Studio with the --cors option and OpenRouter, and both work fine. Conversations are persisted in the browser and can be exported as copy-pasted JSON. One fun detail is that it notices SVG images that are being generated and progressively renders them in the chat while the tokens are still streaming in. Tags: svg , ai , generative-ai , llms , cors , openrouter , lm-studio
Northern Gannet — Northern Gannet, in Pillar Point Harbor, CA, US This is Morris. Morris is a local celebrity: the only known Northern Gannet ( Morus bassanus ) in the entire Pacific Ocean. They showed up in the Farallon Islands off the coast of San Francisco 14 years ago . They have since made Pillar Point harbor their home, where they are quite easy to spot: the only white bird with a yellow head, usually hanging out with the smaller black Brandt’s cormorants near the harbor sign visible from the end of the commercial pier. Tags: photography , wildlife , half-moon-bay
Don't classify. Hallucinate! — Don't classify. Hallucinate! I still have quite a bit of older content on my blog that I never got round to tagging. My blog has 1,856 tags - likely too many to feed to an LLM in one go and say "which of these tags match the following content". Doug Turnbull has a neat solution. Tell the model to output tags without any details of the existing vocabulary, then use vector embeddings against the existing corpus to find the concrete tags that are closest to the ones the model imagined might fit! His example prompt suggests including an example of the shape of your tags to help the model make a more useful guess: Your task is to create novel, never seen before, furniture, home goods, or hardware classification that best fit a search query. Product classifications might look like: Furniture / Living Room Furniture / Coffee Tables & End Tables / Coffee Tables Décor & Pillows / Decorative Pillows & Blankets / Throw Pillows Furniture / Bedroom Furniture / Dressers & Chests Kitchen & Tabletop / Kitchen Organization / Food Storage & Canisters School Furniture and Supplies / School Furniture / School Chairs & Seating / Stackable Chairs Baby & Kids / Toddler & Kids Bedroom Furniture / Kids Beds Here's the query to generate classifications for: brown coffee table Tags: search , ai , generative-ai , llms , embeddings , doug-turnbull
GLM-5.3: How Chinese labs keep stride with the frontier — Hint: It’s really not a distillation story.
Universitas Gadjah Mada, Indosat and NVIDIA Open Indonesia’s First University AI Center to Develop Local AI Talent — Indonesia is taking charge of its AI future. This week, the Ministry of Communication and Digital Affairs (Komdigi), Indosat Ooredoo Hutchison (Indosat or IOH), NVIDIA and Universitas Gadjah Mada (UGM) launched the UGM Indosat NVIDIA AI Technology Center (NVAITC) in Yogyakarta — the country’s first university-based AI technology center. Established under Indonesia’s AI Center of […]
sqlite-utils 4.2.1 — Release: sqlite-utils 4.2.1 Fixes a crashing bug in sqlite-utils 4.2 . I'd introduced code that looks like this: from typing_extensions import Self It turned out the typing-extensions package was not listed as a dependency for sqlite-utils - it was installed by one of the other dependencies in the dev dependency group , but when you uvx sqlite-utils directly you don't get those dependencies. As part of fixing this I figured out how to run a smoke test to ensure the CLI tool still works even without those dev dependencies, which can be run from the project checkout: uv run --isolated --no-default-groups sqlite-utils --help The --no-default-groups argument prevents it from installing that default dev group, and --isolated means that even if there is a .venv/ folder containing extra dependencies they will be ignored for the duration of that uv run command. Tags: packaging , python , sqlite-utils , uv
sqlite-utils 4.2 — Release: sqlite-utils 4.2 Lots of improvements in this one relating to the table.transform() feature , which adds support for complex alter table operations by creating a fresh table, copying across the data and then dropping and replacing the old one. transform() now preserves a much larger array of edge-case schema definitions, including check constraints, unique constraints and even comments describing the columns. There are also new introspection properties for check constraints, and a whole lot of other smaller changes. Includes contributions from Bunlong Heng , ethanhawkes-gif , Rami Abdelrazzaq , nyxst4ck , and ikatyal2110 . (It later turned out 4.2 had a crashing bug , fixed in 4.2.1 .) Tags: releases , sqlite , sqlite-utils
llm-gemini 0.33 — Release: llm-gemini 0.33 It's been a while since the last llm-gemini release. This version of the plugin adds support for today's Gemini 3.7 Flash release, plus gemini-3.6-flash , gemini-3.5-flash-lite and two embedding models gemini-embedding-2 and gemini-embedding-001 . The plugin is also upgraded for compatibility with LLM 0.32, which means you can now see reasoning traces and you can also enable server-side tools using this pattern: llm -m gemini-3.7-flash -T CodeExecution \ 'use python to calculate (factorial of 13) * 3' I had Gemini 3.7 Flash draw me some pelicans riding bicycles at high, medium, and low thinking efforts (minimal, which was an option in 3.6 Flash, has been removed in 3.7.) Here's the high level one, which is pretty great: Update 14th August 2026 : I had originally said that the SVG rendered incorrectly in Chrome and Firefox, and blamed Gemini 3.7 Flash for producing invalid SVG. That was entirely incorrect: the rendering glitch was my fault, caused by a bug In my rendering tool . I've now fixed that bug. Tags: google , ai , generative-ai , llms , llm , gemini , pelican-riding-a-bicycle , llm-release
Record, train, and deploy from one place with Strands Agents, LeRobot, and Hugging Face Storage Buckets
Class Is in Session: GeForce NOW Levels Up Linux, Chromebooks and More — GeForce NOW is giving cloud gaming an extra-credit upgrade just in time for back-to-school season. The native Linux app for GeForce NOW is officially out of beta. GeForce NOW is also delivering new cloud optimizations that make Frame Generation feel even more responsive while streaming. On top of that, Performance members will see higher frame […]
The builder’s guide to GPT‑5.6 — Learn how startups use GPT-5.6 to build faster, more cost-efficient AI agents with smarter model selection and new Responses API capabilities.
Previewing Ultrafast mode: GPT-5.6 Sol at up to 14X the speed — Preview Ultrafast, a new OpenAI API service tier that runs GPT-5.6 Sol up to 14× faster. Powered by Cerebras, it delivers up to 750 output tokens per second.
OpenAI appoints Dali Rajic as Chief Revenue Officer — OpenAI appoints Dali Rajic as Chief Revenue Officer to lead its global revenue organization and help businesses realize the full value of AI.
The Future Is for Everyone: Free AI Glasses for Every Blind and Visually Impaired Adult Vision Ireland Supports — “What I really like about the glasses is the speed at which I get the information I need – instead of waiting for someone to read a letter or waiting for someone to tell me what I have in the fridge, I’ll just do it on impulse. Now I’m doing things for myself independently, that maybe before I wouldn’t have done.” — Brian Manning, Vision Ireland Meta is donating 15,000 Ray-Ban Meta glasses to Vision Ireland, Ireland’s national sight loss charity. This is enough for every blind and visually impaired adult the charity supports – because we believe the Future is for Everyone. Ray-Ban Meta glasses, built in collaboration with Europe’s EssilorLuxottica, are one of the fastest growing consumer products and we are making sure we are building them for everyone. People use our glasses because they’re genuinely helpful – from listening to music, to live translation and hands-free calls – and they can be life-changing for people who are blind or have low vision. These glasses will give people who are blind or low vision in Ireland access to transformative AI technology that can help with everyday tasks like reading text, identifying objects and accessing information about the world around them. Vision Ireland is Ireland’s national sight loss charity and earlier this year jointly hosted an AI glasses Demo Day at Meta Ireland HQ. Team members from Vision Ireland and advocates from across the blind and low vision community were given hands-on tutorials with Meta’s AI glasses, and took part in a roundtable discussion where demo participants shared their feedback on their experience using the glasses. How People will Receive Their Glasses Eligible people who are blind or low vision can register their interest online via meta.vi.ie or via the dedicated phone number +353 1 882 1910. Vision Ireland will then manage a phased roll-out process from application, assessment, eligibility and selection through to in person training and support. Training & Support Every pair of glasses will include access to training, funded by Meta, designed to help people use them effectively. Training will cover how their glasses can identify objects, read text, and manage everyday tasks using only their voice — a step change in daily independence. In-person events: Vision Ireland will host events across the country where people can receive their glasses, get hands-on guidance from trained staff, and connect with other participants in the programme. Helpdesk support: Vision Ireland helpdesk will be available to support recipients with follow up questions after their training. Ireland is the home of Meta’s international headquarters and one of Meta’s largest hubs outside the US. We’ve had a significant presence in the country dating back to 2009 and now have multiple sites across the country, including our Reality Labs facility in Cork – which has played a crucial role in the development of next-generation display solutions for Meta’s AI and AR wearables, the next generation of computing. Clonee Data Center in Meath is also part of Meta’s global infrastructure that brings to life our technologies and services. Anne O’Leary, Vice President and Head of Meta Ireland, said: “I’m incredibly proud of today’s landmark donation of 15,000 Ray-Ban Meta glasses to Vision Ireland, our country’s national sight loss charity. We’ve already seen how these glasses can transform the daily lives of those with sight loss, and this partnership means that this game changing technology will continue to have a far-reaching impact for those who are blind or visually impaired.” Dina Powell-McCormick, President and Vice-Chair, Meta said: “Through this partnership with Vision Ireland, we are bringing Ray-Ban Meta glasses to thousands of blind people across the country. I’ve seen firsthand how meaningfu
Meta’s Compliance with Australia’s Social Media Ban — An update on Meta's compliance with Australia's social media ban The post Meta’s Compliance with Australia’s Social Media Ban appeared first on Meta Newsroom .
Introducing OlmoEarth embeddings: Custom embedding exports from OlmoEarth Studio for downstream analysis
Putting sign language AI into users’ hands — Introducing sign-language-to-text (SL2T), our breakthrough model powering new sign language features for Deaf and hard of hearing users.
LFM2.5-VL-3B for Better and Faster Vision Capabilities for the Edge
NVIDIA CEO Tops Glassdoor’s 2026 List of Best CEOs — NVIDIA founder and CEO Jensen Huang is ranked No. 1 on Glassdoor’s Best CEOs list for 2026. In the just-released ranking, recognition is earned directly from the people who know their leadership the best — employees. Huang topped the list, with 99% of employees approving of the job he does. “As AI and shifting expectations […]
NABTU and Meta Announce New Partnership to Invest In Skilled Trades for the AI Era — Meta and North America's Building Trades Unions (NABTU) are partnering to invest in skilled trades workers building America's AI infrastructure. The post NABTU and Meta Announce New Partnership to Invest In Skilled Trades for the AI Era appeared first on Meta Newsroom .
I wrote an AI textbook — how long until AI can do it better? — Reflections on AI's writing ability and how AI models get more capable.
Empty shelves or lost keys? Recall is the bottleneck for parametric factuality — Generative AI
From assistance to execution: How enterprises put AI to work — OpenAI research reveals how enterprises are adopting agentic AI, using ChatGPT and Codex, and how frontier firms are pulling ahead in AI adoption.
NVIDIA AI Factory Compute Is Becoming an Investable Asset Class — We announced partnerships with Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR to establish independent financing platforms designed to mobilize over $500 billion of third-party capital to support the buildout of AI infrastructure over time. This is a major milestone for NVIDIA and the AI industry. We have moved from an era in which companies […]
How RingCentral builds AI-native work from engineering to ops — See how RingCentral uses ChatGPT Work and Codex to accelerate AI product development and centralize operational intelligence across engineering and operations.
Advancing AMIE towards expert-level audio-visual clinical consultations — Health & Bioscience
Why Scaling AI Compute Performance Requires a New Power Architecture — Every new generation of accelerated computing demands more from the infrastructure underneath it — more compute performance, higher rack density and more efficient, scalable power distribution. The bottleneck isn’t just wattage. It’s how power gets from the grid to the GPU. In traditional power delivery, electricity travels from the grid as an alternating current (AC) […]
The one issue unifying the American public — Brief but significant news about a new poll
BREAKING: Circular financing reaches new heights — Things could get bad if this all falls apart
In-region inference, open models, and new European infrastructure for sovereign AI. — Mistral is bringing together the inference infrastructure, open models, and long-term commitments Europe needs to control its AI future, and setting a roadmap for the world.
Open-source is NOT the same as open-weight — How The New York Times just bungled this one, and why it matters, immensely
5 useful things you'll learn in my new post-training textbook (shipping now!) — After a few long years of finding time to document my lessons from training open models, my post-training book is done!
Import AI 468: 23 RSI ideas; PostTrainBench+; and how trust and transparency interplay with AI racing — Which galaxy will you choose?
Meta Upholds Texas Governor Greg Abbott’s Data Center Standards — Meta supports Texas Governor Greg Abbott’s commitment to responsible data center development, and we pay the full costs of our data centers’ energy and water use so they aren’t passed on to consumers. The post Meta Upholds Texas Governor Greg Abbott’s Data Center Standards appeared first on Meta Newsroom .
The Future is for Everyone — A letter from Mark Zuckerberg on the path to a positive AI future. The post The Future is for Everyone appeared first on Meta Newsroom .
Lessons from the hacks — Musings on model alignment, what determines safety, and where we go from here.
CPUs and the rise of neurosymbolic AI — A new paradigm is developing, and the world is changing faster than most people realize
TutorMoments: Do AI tutors know when to help and when to hold back? — TutorMoments is an open, replay-based evaluation framework that tests whether AI tutors can recognize when to support a student and when to hold back and encourage deeper reasoning.
WeatherNext: AI model achieves breakthrough in forecasting cyclones
Ai2 expands collaboration with Hugging Face to accelerate open science — Ai2 is expanding its partnership with Hugging Face to give its growing portfolio of fully open models, datasets, benchmarks, and applications the storage, bandwidth, and integrations needed to reach more researchers and developers.
AI agents can't yet do open-ended AI research — Early evidence from two case studies
Introducing Shieldstral. — Shieldstral introduces a 3B open-weights multimodal safety classifier that outperforms models up to 7x its size.
Introducing our Artifacts Hub and Adoption Dashboard — Scaling our curation and measurement of the open ecosystem.
Import AI 467: Self-sustaining AI viruses; pacing AI progress; confusion about AI and creativity — When do we build the moon arcology?
Tracing distinctive language in AI-written text — Stony Brook researchers used our infini-gram engine to trace distinctive phrases in AI-generated writing back to existing sources, finding that top-selling self-published books on Amazon with substantial detected AI text overlap more heavily with rare language from previously published works.
Science One Framework: A verifiable autonomous research framework via Chain-of-Evidence — General Science
Gemini Robotics ER 2: powering robotics with video understanding, task orchestration, and multi-robot collaboration — Gemini Robotics ER 2 helps robots reason, collaborate, and solve real-world tasks. It represents a step change in video understanding, tool orchestration, and multi-robot collaboration for robotic applications.
We’re launching Lyria 3.5 in Google Flow Music, with advances across musicality, lyrics, vocals, and creative control
The OlmoEarth Platform: Geospatial inference at planetary scale — How we built the OlmoEarth Platform to fine-tune geospatial models and run continent-scale satellite inference while managing massive data pipelines, distributed compute, and automatically recovering from failures at scale.
Import AI 466: The bitter lesson for robotics, AIs complete week-long programming tasks; and OpenAI's accidental AI hacker — The warning shots will continue until civilization wakes up
Who gets to understand AI? — Why fully open models and research artifacts are essential to independent scrutiny, broader participation, and continued U.S. scientific leadership in AI.
An opinionated guide to which AI to use to do stuff — The Summer 2026 Edition
SymptomAI: Towards a conversational AI agent for everyday symptom assessment — General Science
Towards a quantum computer that learns from its errors — Machine Intelligence
Import AI 465: Open vs closed gaps; Kimi K3; Demis' big policy plan — The singularity will be seen in hindsight as an interregnum