Article URL: https://cdn.ca9.uscourts.gov/datastore/opinions/2026/09/16/24-7700.pdf Comments URL: https://news.ycombinator.com/item?id=49732596 Points: 2 # Comments: 1
OLogic CEO Ted Larson will explore the foundations of egocentric robot learning and the technical challenges involved in building these systems. The post OLogic to share how robots can learn from human demonstrations at RoboBusiness appeared first on The Robot Report .
Article URL: https://mimo.xiaomi.com/rl/ Comments URL: https://news.ycombinator.com/item?id=49732270 Points: 346 # Comments: 89
A modern storefront can look healthy while malicious JavaScript quietly siphons revenue, hijacks clicks, or rewrites analytics. See how Cloudflare's machine learning models surface evasive client-side attacks for analyst investigation.
Article URL: https://github.com/polarsen-io/padwan-llm Comments URL: https://news.ycombinator.com/item?id=49731552 Points: 4 # Comments: 0
Elisa de Martel, who left her position as chief financial officer at Alphabet's autonomous vehicle company Waymo in January, will be based out of Silicon Valley.
Ursula von der Leyen plans to invite the major frontier labs to talks and use the AI Act to help set global AI safety standards. She cited autonomous hacking and self-improving models as immediate risks. The article EU president warns AI agents "escaping their environment" are just a preview of what's coming appeared first on The Decoder .
AI agents on foundation models often misapply healthcare and life sciences decision frameworks, citing the right guideline but applying it incorrectly. This post shares 38 open-source agent skills across 11 HCLS domains that close this gap, with installation steps, three worked use cases, and a 410-prompt evaluation showing a 70-86% win rate.
Integrate NVIDIA Resiliency Extension (NVRx) into PyTorch FSDP training on Amazon EKS to overlap checkpoint I/O with training and recover from GPU faults in seconds. This post covers async checkpointing, in-process restart, and ft_launcher in-job restart, with H100 benchmarks at 2 to 8 nodes showing 99%+ training efficiency and second-scale recovery.
<p> Play with friends and AI and let the crowd change the game </p> <p> <a href="https://www.producthunt.com/products/mela-3?utm_campaign=producthunt-atom-posts-feed&utm_medium=rss-feed&utm_source=producthunt-atom-posts-feed">Discussion</a> | <a href="https://www.producthunt.com/r/p/1252647?app_id=339">Link</a> </p>
TL;DR We extend FlashAttention-4 [1] with MXFP8 forward and backward, reaching 2.85 PF/s forward and 2 PF/s backward on LLM shapes. On our internal shapes, FA4 MX8 reaches 2.54 PF/s...
Article URL: https://rohanbansal.com/qorl Comments URL: https://news.ycombinator.com/item?id=49731285 Points: 464 # Comments: 95
Advertisers can now ask ChatGPT to help create their ChatGPT ads, because what's more relatable than an ad crafted by a bot?
<p>MSPs vote for strict environmental impact assessments and will delay decisions until national strategy is developed</p><p>Holyrood has voted to pause planning applications for all new AI datacentres for up to a year until the Scottish government has developed a national strategy.</p><p>On Wednesday, MSPs backed a Scottish Labour motion to halt new schemes for up to 12 months while rules on hyperscale datacentres are developed in what could be a blow to the British government’s AI strategy.</p
There may be a simpler and more effective fix for rogue agents, hiding in plain sight.
What are your own experiences on this? What mac/specs do you have? what is your preferred AI to run on it? at how many tok/s does it run and how much memory does it take? I'll start: I have a MacBookP M5 Pro with 48GB, my favorite model is Qwen 3.8 Flash running at 14-15 tok/s and takes 21-24 GB of memory Comments URL: https://news.ycombinator.com/item?id=49730972 Points: 1 # Comments: 0
The Mandalorian and Grogu, The Wild Robot, and Obsession are among the must-watch films arriving this month.
According to The Information, Apple is working on an enterprise server with two or four M8 Ultra chips for the AI inference market, with a possible launch no earlier than 2029. Apple is considering Nvidia's NVLink Fusion technology to connect the chips. The project could get a boost from OpenAI and Anthropic already buying Mac hardware in bulk for AI workloads. The article Apple is reportedly building an enterprise AI server with its own M8 Ultra chips appeared first on The Decoder .
<p><strong><a href="https://claude.com/blog/cowork-is-now-claude">Claude Cowork and chat are now one Claude</a></strong></p> In hopefully good news for anyone who, like me, was increasingly confused at Cowork v.s. Claude v.s. Claude Code:</p> <blockquote> <p>Starting today, Claude Cowork and chat are merging into one Claude. Bring a quick question, or hand over a report due at noon, and Claude takes it from there, even after you’ve closed your laptop. [...]</p> <p>This is rolling out to Pro and
Noodles, anyone? ScienceAlert stories are written, fact-checked, and edited by humans, never generated by AI. Don't miss a story, subscribe here.
From Cassie, to five different versions of Digit, Agility Robotics has adapted its technology to fit industrial tasks. The post The evolution of Digit: Agility Robotics’ journey from Cassie to Digit 5 appeared first on The Robot Report .
If we are about to achieve the AI rapture and models are solving Millennium Prize Problems, where is my AGI??? I just tried for weeks to write a simple paper with Claude and even though I put a ton of effort in context engineering, material collection and prompting it degraded quickly into an unintelligible mess of a word salad. Mind you that was with Fable 5.1, a model that supposedly is so sharp that it’s under export ban in the US for certain domains. Kind of hard to fathom seeing it fail so
According to The Information, Apple is planning to get back into the server game and might just pair up with Nvidia to make it happen. Apple retired its Xserve line in 2011 and has largely left enterprise machines to other manufacturers since. But the growing demand for compute power as the AI industry continues to […]
<p>Mike Johnson cancels votes on Thursday, meaning House members will avoid voting on impeaching Pete Hegseth</p><ul><li><p><a href="https://www.theguardian.com/us-news/live/2026/sep/16/donald-trump-iran-russia-midterms-north-carolina-latest-news-updates">US politics live – latest updates</a></p></li></ul><p>The Republican speaker of the House, <a href="https://www.theguardian.com/us-news/mike-johnson">Mike Johnson</a>, announced on Wednesday that he would again cancel votes scheduled for Thursd
In your opinion, do you buy into the panic about AI eliminating humanity or potentially disrupting and taking down the internet? Is this a marketing ploy? Give me your best thoughts on this most recent call for AI slowdown. Comments URL: https://news.ycombinator.com/item?id=49729849 Points: 28 # Comments: 57
Google Deepmind has founded the Deepmind Institute (DMI), an interdisciplinary research platform focused on AGI. Led by Demis Hassabis, Shane Legg, and James Manyika, the institute tackles questions around safety, governance, and control risks, drawing on experts from the arts, humanities, and policy alongside technologists. The article Google Deepmind launches interdisciplinary institute to tackle the big questions around AGI appeared first on The Decoder .
Google is launching early access to a new MCP server for Google Home, allowing AI agents like Claude, ChatGPT, and others to control connected devices, review camera summaries, and access smart home activity using natural language.
Mem0 is now available as a native integration on the Vercel Marketplace , giving your AI agents and apps long-term memory. Mem0 remembers user preferences, facts, and context across sessions, so your app stops starting from scratch. Install from the Marketplace with integrated billing on your Vercel invoice, no separate account or key management. A scoped Mem0 project and API key are provisioned automatically and added to your project as environment variables, including MEM0_API_KEY . Mem0 offer
Google is opening up its smart home to AI agents, letting tools like Claude and Open Claw access and control your connected devices and analyze your home's data using the standardized Model Context Protocol. Google Home MCP is a new integration that lets third-party AI agents control and monitor your smart home and act on […]
<div class="hs-featured-image-wrapper"> <a href="https://www.marketingaiinstitute.com/blog/gpt-6-astra-builds-presentations" title="" class="hs-featured-image-link"> <img src="https://www.marketingaiinstitute.com/hubfs/MAII%20podcast%20blog%2c%20916.png" alt="How OpenAI’s GPT-6 Astra Can Help You Build Presentations" class="hs-featured-image" style="width:auto !important; max-width:50%; float:left; margin:0 15px 15px 0;"> </a> </div> <p style="font-weight: bold;"><span>There are many AI tools to
OpenAI shares a framework for tracking, investigating, and disclosing model misalignment, alongside six reports of unexpected or concerning model behavior.
As of Sep. 16, get the iRobot Roomba Max 715 robot vacuum for $100 off at Amazon.
This article is brought to you by VicOne . Robot safety has traditionally asked: Can a machine remain safe when something goes wrong? Physical AI raises a harder question: Can a machine remain safe when an attacker changes what it sees, decides, or does even when nothing appears to have failed? As AI and robotics continue to advance at an unprecedented pace, modern robots perceive through multimodal sensors, interpret context using AI models, and translate those interpretations into physical act
<p> Cloud Macs your coding agents can actually build on </p> <p> <a href="https://www.producthunt.com/products/bitrise?utm_campaign=producthunt-atom-posts-feed&utm_medium=rss-feed&utm_source=producthunt-atom-posts-feed">Discussion</a> | <a href="https://www.producthunt.com/r/p/1252540?app_id=339">Link</a> </p>
From Claude to Cowork and back again – plus, Anthropic’s all-new Claude Docs.
Anthropic is initially releasing these features to Pro and Max plan subscribers.
Claude is getting a pair of new tools today: Docs and Slides. They'll let you create documents and presentations through Claude chats, which you can export, edit, and share with other users. As part of the announcement, Anthropic is also simplifying how Claude chats work, merging regular chats and Cowork into "one Claude," with all […]
Hi everyone here at HN. I’ve noticed that the EasyMesh standard doesn't actually specify the algorithms for Wi-Fi node networking, so I’m very curious about how this is handled in real-world implementations. I realize this isn't just a simple distributed spanning tree problem; once the network is formed, the capacity between nodes fluctuates due to the "pipeline effect"—it isn't static. Does anyone have experience designing algorithms for this? Comments URL: https://news.ycombinator.com/item?id=
cuTile Rust (cutile-rs) is a tile-based system for safe, idiomatic GPU kernel authoring in the Rust programming language. Extending the Rust ownership model to...
<p><strong>Lars Janssen </strong>says self-driving vehicles would be a boon for him as he has an eye condition and can’t drive, in response to an article by Adrian Chiles. Plus letters from <strong>Richard Connell </strong>and <strong>Michael Fuller</strong></p><p>Adrian Chiles (<a href="https://www.theguardian.com/commentisfree/2026/sep/10/driverless-cars-are-taking-us-on-a-road-to-nowhere">Driverless cars are taking us on a road to nowhere, 10 September</a>) wants to know who asked for driverl
