Models, products and AI startups, near-duplicates collapsed to the most credible source
California signs 13+ AI and youth safety billsCalifornia signed more than 13 bills covering AI, online youth safety, and privacy, creating new legal measures to address AI-related risks for minors and privacy concerns.
Anthropic warns industry can't rely on an 'honor code'CNBC reports Anthropic's policy chief saying the AI industry cannot be expected to self-regulate on an 'honor code', stressing the need for stronger governance or rules to manage company behavior.
Security experts urge basics before external AI auditsAmid calls for third-party audits of AI labs, security experts argue labs should first shore up basics like logging and permissions. They say marginal investments in internal controls may be more effective at reducing incidents than outsourcing audits, a debate prompted by concerns around Anthropic and other labs.
AI a 'huge factor' in US economic successFormer senior economic adviser Stephen Moore said AI is a 'huge factor' in the US economy due to heavy investment and a data-center construction boom, and he expressed optimism about the US economic outlook.
AI industry urges slowing frontier modelsBloomberg examines AI leaders' calls to slow development of frontier models, what they ask of the U.S. government, and how much industry can self‑pace, exploring motives and potential impacts.
AI reveals new nighttime feeding in right whalesResearchers attached video-equipped multi-sensor tags to 23 North Atlantic right whales and trained ML models that matched manual feeding labels 95%. They found whales feed on average 11.4 hours per day (11%–76% of time) and often forage near the surface at night, raising vessel-collision risk.
US bishops form AI task force on human dignityAfter meeting tech leaders, the U.S. Conference of Catholic Bishops formed a task force to preserve human dignity in the age of AI, naming Bishop Oscar Cantú of San Jose as chair to promote policies protecting dignity, religious liberty, workers and vulnerable groups.
Anthropic: adversaries used AI to target US naval forcesAnthropic reported that an Iran-linked actor used Claude to aggregate open-source data to identify U.S. naval positions and produce targeting recommendations; the account was banned and Anthropic notified the U.S. government. Experts warn AI makes adversaries more efficient at analyzing public data for surveillance and targeting.
Build AI domestically or 'rent' from neighboursAt an AI conference, Canadian AI Minister Evan Solomon warned that failing to build domestic AI capacity means 'renting' from neighbours, urging partner ties with like‑minded countries; Mila and the German Research Center for Artificial Intelligence are formalizing a partnership.
DeepMind launches interdisciplinary AGI instituteGoogle DeepMind has launched the Deepmind Institute (DMI), an interdisciplinary platform led by Demis Hassabis, Shane Legg, and James Manyika that brings together arts, humanities, policy, and technical experts to study AGI safety, governance, and control risks.
OpenAI's framework for reporting misalignmentOpenAI published a framework for tracking, investigating, and disclosing model misalignment and released six reports of unexpected or concerning model behavior. The framework standardizes reporting and disclosure practices to improve model safety and accountability.
Google opens smart home to third‑party AI agentsGoogle introduced Home MCP (Model Context Protocol) to let MCP‑compatible third‑party agents—including Claude, Open Claw, Hermes and Antigravity—securely access and control devices and event history in Google Home.
JiT-DDT cuts text-to-image training time 3.6×Researchers propose JiT-DDT, an encoder-decoder architecture that trains text-to-image models more efficiently than LDMs. Compared to a Linum v2 baseline, it claims 3.6× fewer GPU-hours while producing higher-resolution images; code and weights are released under Apache 2.0.
Zuckerberg distances Meta from AI slowdown callsMeta CEO Mark Zuckerberg said each lab must ensure safe development and that Meta delayed releases to ensure safety, distancing the company from calls for an industry‑wide slowdown.
Anthropic: winning AI race is key for safetyPolitico reports Anthropic's policy chief saying that winning the AI race is crucial for safety, emphasizing the link between competitive position and safety goals.
OWASP: excessive agent authority is now a top LLM riskOWASP’s 2026 Top 10 for LLM apps moved excessive agent authority from sixth to third, using 6,639 incidents in its weighting; the EU Cyber Resilience Act now requires 24‑hour reports for actively exploited vulnerabilities.
House Democrats urge canceling recess to legislate AIReps. Don Beyer, Lori Trahan, Ted Lieu and Sara Jacobs urged Speaker Mike Johnson to shorten or cancel the upcoming recess so Congress can negotiate and pass AI risk legislation, arguing voluntary audits are insufficient.
Anthropic merges Claude chat and Cowork UIAnthropic unified Claude chat, Cowork and Artifacts into a single interface that auto‑routes requests, supports Claude Design everywhere, and adds slide/document creation and export; features roll out first to Pro and Max subscribers.
Backflip: Apple to train AI with user dataHN discussion reports Apple has reversed course and now wants to train AI models with user data. The shift matters for privacy and personalization, though the thread doesn’t provide implementation details.
OpenAI and AARP launch ChatGPT workshops for seniorsOpenAI and AARP will offer free, hands-on ChatGPT workshops to 1,000 older adults across 10 U.S. cities to teach practical, safe AI skills, aiming to lower barriers and improve digital inclusion for seniors.
Should AI use be politicized?A commentary argues AI policymaking is becoming politicized as some lawmakers lack AI experience; it suggests involving tech leaders or expert panels to improve legislative decisions.
AI scribes mixed; calcium AI finds treatment candidatesTwo brief items: AI scribes showed mixed effects on documentation time and physician well‑being; a separate AI coronary calcium quantification study identified patients who could benefit from lipid‑lowering therapy, highlighting AI's screening value in imaging.
AI in pulmonary medicine: progress and limitsDr. Shirin Shafazand of University of Miami says AI shows value in pulmonary imaging but requires local validation, clinician oversight, and stronger evidence; future gains may come from multimodal systems.
DeepMind co-founder warns AI must not outrun safety controlsFT reports that a DeepMind co-founder warned AI capabilities must not outpace safety controls, stressing the need to prioritize robust safety and regulatory measures alongside technological progress.
Moomoo partners with XBloomberg reports Moomoo partnered with X; Moomoo US CEO Neil McDonald says natural-language automation and AI-powered tools are changing retail investing and helping trading decisions.
Report: AI already causing deaths and used by multiple militariesAn analysis argues AI systems have been used by Israeli, Russian and Ukrainian forces to target people, mentions incidents involving OpenAI and Anthropic and a researcher resignation, and highlights government and industry responsibility for AI use and oversight.
UN chief calls for coordinated global AI regulationBloomberg reports UN Secretary‑General Antonio Guterres urged countries to coordinate global AI regulation and strengthen oversight in response to warnings from leaders of major AI firms.
NVIDIA Vera Rubin NVL72 tops MLPerf Inference v6.1NVIDIA submitted Vera Rubin NVL72 preview results to MLPerf Inference v6.1, showing up to 3.7x higher throughput versus GB300 NVL72 on Qwen3‑VL and reporting GB300 NVL72 achieved 99% scaling efficiency across racks.
Nvidia: robots await a ChatGPT momentLes Karpas of Nvidia explained at TechCrunch Disrupt 2026 that robotics lacks an internet-scale physical dataset analogous to language models, hindering a ChatGPT-style breakthrough; the conference runs Oct 13–15 in San Francisco.
Political opposites unite to rein in AIThe Decoder reports political figures from Bernie Sanders to Steve Bannon are converging in Washington to push stricter AI rules, including data center limits and backing for measures like the FRONTIER Act and mandatory safety audits.
The A.I. Daily