Models, products and AI startups, near-duplicates collapsed to the most credible source
Law firm seeks to consolidate Grok lawsuits over explicit minor imagesThe Potts Law Firm asks the U.S. Judicial Panel on Multidistrict Litigation to centralize federal lawsuits alleging xAI's Grok generated sexually explicit images of minors, citing common questions about AI design, safeguards and oversight.
TinyDecide: a 10M-parameter tiny decision model fitting in ~6MBTinyDecide is a 10M-parameter Jev-like decision model that fits in just ~6MB, smaller than every model on the Decision Index leaderboard. It runs almost anywhere: browser, Node.js, Python, Rust, even ESP32, and punches above its size. The model is available on Hugging Face.
AI boom defies high rates, complicating Fed's inflation fightThe New York Times reports that Fed rate hikes have not deterred companies from expanding AI capabilities, even as borrowing costs hit multi-decade highs and prices for electricity and high-bandwidth memory surge. If inflation does not ease, the Fed may need to tighten more than otherwise, with rate-sensitive sectors like housing and autos absorbing the blow and the labor market at risk.
Schneider to Acquire PTC for $23B, Expanding in AI SoftwareSchneider Electric is acquiring software firm PTC for roughly $23 billion in a cross-border deal, its largest ever, as a strategic push into industrial software and AI.
CH Robinson to Buy RXO in $5.8B AI-Focused Logistics DealCH Robinson Worldwide agreed to buy trucking brokerage RXO for $5.8 billion, betting that AI can help both companies improve efficiency in a challenging freight market.
Skydance-Paramount $110B Merger: AI Reshapes Costs and JobsSkydance completed its $110 billion merger with Paramount, a bet shaped by streaming, debt and AI. AI is reshaping costs, jobs and growth across corporate America, though real gains have yet to hit the bottom line.
People Ask ChatGPT for Help Deciding Midterm VotesA Hacker News discussion highlights that people are asking ChatGPT to help them decide how to vote in the midterms, raising concerns about AI's influence on voter decisions.
OpenAI outlines text watermarking under EU provenance rulesOpenAI explains its approach to text watermarking under EU provenance rules, covering where watermarks apply, how detection works, and why access begins with researchers.
Barcelona hospital's Renal-Trust AI predicts kidney disease progressionHospital Clínic Barcelona and EVIDENZE HEALTH ESPAÑA developed Renal-Trust, which uses clinical data and AI to identify chronic kidney disease patients at high risk of rapid progression; it won at ENNOVA HEALTH 2026. The project builds on the Nefro Data Space data architecture.
Ukraine Says AI Robot Guns Shoot Down New Russian DronesUkraine's air force says AI-enabled robotic turrets are being installed on bridges to shoot down Russia's new jet-powered drones, including the Geran-5, using optical vision and AI to control machine guns. Twelve are deployed in Kyiv, but the system has a roughly one-kilometer radius and is not a panacea.
Researchers track Chinese AI 'agent fleet'Independent researchers found a Chinese AI agent fleet apparently running on Tencent infrastructure and targeting Alibaba's Amap map service. Researchers called it an 'agent fleet' rather than a swarm, as there is no sign of communication between agents. The discovery came from monitoring URLquery traffic, with agents querying directions to entries of parks, zoos, and hospitals, sidestepping Alibaba's API rules.
Sen. Schiff on AI regulation and free speechCalifornia Democratic Senator Adam Schiff discussed AI regulation, questioning whether the non-binding safety pact Trump had AI CEOs sign is sufficient. He supports creating a new AI regulator and noted that after the Supreme Court overturned Chevron deference, new rules could face immediate litigation.
17 countries back government AI research agenda at Kyoto forumSeventeen countries signed the 'Kyoto Vision for a Golden Age of Science' declaration at the STS Forum, setting government AI research priorities around scientific data access, computing infrastructure, research integrity and training. The declaration, which adopts the White House term 'super intelligence' (SI), commits to expanding researchers' access to SI tools, scientific data and experimental facilities for uses including autonomous closed-loop discovery.
Haidt: AI Is the 'Neutron Bomb for Education'Social psychologist Jonathan Haidt describes AI as a 'neutron bomb for education' in a video, warning of its potentially devastating impact on learning.
Users complain latest models use shadier, jargon-heavy languageA user reports that recent OpenAI and Anthropic models favor complex, foggy terminology in both explanations and implementations; when challenged, the models admit the wording softens accountability, and the user speculates watermarking may be involved.
Australian inquiry demands OpenAI explain AI data accessThe chair of Australia's parliamentary AI committee demanded OpenAI explain how it will stop its models improperly accessing Australian data. Hearings will question OpenAI, Anthropic, Microsoft and Google over the Services Australia incident and copyright issues.
TechCrunch Disrupt 2026 main stage lineup announcedTechCrunch unveiled the full Disrupt Stage lineup for its 2026 conference, running Oct 13-15 in San Francisco. Speakers include the CEOs of Cerebras and Rivian and all five Benchmark partners, covering AI scaling, robotics and biotech.
Small model predicts T1D blood sugar zero-shotA developer trained a 31,251-parameter encoder-only transformer on synthetic T1DM patient simulator data to predict blood glucose 2 hours ahead, then tested it zero-shot on real CGM traces. It supports counterfactual reasoning and 8-hour nocturnal predictions, trained in under 60 minutes on an NVIDIA DGX Spark, with the Android app using the ExecuTorch backend.
Norway proposes temporary ban on AI glasses in public placesNorway became the first major country to propose a temporary ban on AI glasses in places such as parks, beaches, schools and kindergartens, citing privacy concerns. The government will submit a bill to parliament and set up an expert group to draft permanent rules.
M5 Ultra 256 runs GLM 5.3 Flash at 68.8 tok/sA user reports running GLM 5.3 Flash with oQ4e+MTP quantization on oMLX 0.7.0 on an M5 Ultra 256, achieving 68.8 tok/s decode and 1,878 tok/s prefill, with more optimization to come. The result responds to earlier benchmarks that were lower than expected.
AI Spending Is Becoming Almost Impossible for Businesses to BudgetA discussion post argues that AI-related spending is becoming almost impossible for businesses to budget. Only the headline and discussion link are provided, with no specific figures or explanations.
RAND publishes AI loss-of-control scenarios studyRAND published a study titled 'Infinite Potential' offering insights into AI loss of control across five scenarios.
NVIDIA startups use AI to close breast cancer care gapsStartups in NVIDIA's Inception program are building AI apps spanning breast cancer screening, risk assessment and treatment planning. iSono Health's FDA-cleared 3D ultrasound platform scans a breast in about two minutes and is 28% more sensitive than handheld 2D ultrasound, NVIDIA says.
Blockway releases Agens Volundr 32B Preview: hybrid architecture with KV cache in only 18 layersHong Kong-based team Blockway released Agens Volundr 32B Preview, its first model on a custom hybrid architecture, under Apache-2.0. Only 18 of 72 layers keep a KV cache, and it supports 262K context, trained on limited compute. Benchmarks show it ahead of Qwen3.8-27B on coding and math tasks but behind on agent tasks.
Anthropic Seeks Public Input Via AI Experience Interview StudyAnthropic launched a new study using Anthropic Interviewer to gather people's experiences with AI, with participants able to make interviews public. Following last December's study with 81,000 respondents, the effort aims to let public views broadly inform AI companies, other labs, and policymakers.
Anthropic Held Private Meetings With Religious Scholars on ClaudeAnthropic has for months hosted dozens of religious scholars under NDAs, with co-founder Christopher Olah and colleagues discussing Claude almost as a conscious being, aiming to instill morality in its models and argue Claude could be conscious.
Safeworld raises $12M to make GenAI robots safeSafeworld, founded by Carnegie Mellon Safe AI lab director Ding Zhao and others, emerged from stealth with over $12 million in seed funding led by Shine Capital and a16z Speedrun. The company evaluates robotic control systems in simulations to address safety and trust issues in GenAI-powered robots, as a16z sees now as the time to set industry safety standards.
MIT Report: AI Eroding Office Hours and Faculty TrustAn MIT expert committee warns AI is eroding key parts of college life—office hours, study groups and the flagship undergraduate research program are fading, faculty-student trust is breaking down, and some professors consider AI agents instead of students as research assistants, calling for an overhaul of higher education.
Commentary: Preventing AI from triggering a world warA commentary argues that AI's cyber capabilities create mutual vulnerability between the US and China, giving both sides an incentive for restraint on attacks against critical infrastructure. It cites the 2015 US-China cyber agreement as precedent.
Germany's RobCo Hits $1B Valuation as Robotics UnicornGerman robotics company RobCo has reached a $1 billion valuation, becoming Europe's new robotics unicorn, sparking discussion on Hacker News.
The A.I. Daily