Models, products and AI startups, near-duplicates collapsed to the most credible source
AI-enhanced handheld ultrasound may improve carotid plaque detectionA study in Annals of Family Medicine finds Hyper-CycleGAN, a super-resolution model, sharpens handheld ultrasound images to help detect more carotid plaque in community screening. Among 450 adults aged 40+, AI-enhanced images detected 94.8% of plaques versus 87.6% on standard images.
HN debate: AI has no wisdom, overreliance harms code qualityA Hacker News discussion argues that vibe-coded projects devolve into unmaintainable messes because code maintainability and architectural flaws take months or years to surface. The author warns developers against no longer reading and writing code, relying instead on AI.
Will Open Source Survive the Agents That Replaced It?AI agents make code generation cheap but review remains manual. matplotlib maintainer Tim Hoffmann notes agents can reconstruct most of a package from a simple prompt; fewer downloads would erode maintainers' only reliable reward—visibility—risking open-source decline.
NVIDIA Isaac ROS 5.0 Advances Agentic, Open Source RoboticsAt ROSCon, NVIDIA released Isaac ROS 5.0, a collection of GPU-accelerated ROS packages serving nearly 1.3 million ROS users. It adds agentic workflows, supports ROS Lyrical and Ubuntu 24.04, and contributes a standard data-handling interface with the Open Source Robotics Alliance.
Baselayer raises $35M Series ABaselayer raised $35 million in Series A funding led by M13, bringing total funding to about $40 million. The startup helps financial institutions verify businesses and assess fraud risk, and will extend its identity tech to verifying AI agents' authority.
GPT-6 Astra halves Parallel's research timeParallel used GPT-6 Astra to power its agents, halving the time and cost of researching and synthesizing labor-market data versus prior models.
AI startup Heidi doubles valuation to $900M in new roundAI healthcare startup Heidi raised $340 million, doubling its valuation to $900 million. The company automates clinical notes and other administrative work for doctors, reflecting growing demand for AI tools that boost healthcare worker productivity.
Meta patches Muse zero-day exploitMeta has patched its Muse macOS app after a zero-day flaw that could let attackers hijack the AI agent. Found by researcher Patrick Wardle, the bug stemmed from cloud transcription and undocumented settings, and required local device access to exploit.
Anthropic CEO Dario Amodei to Brief UN Security Council on AIAnthropic CEO Dario Amodei is set to brief the UN Security Council on Wednesday alongside OpenAI CEO Sam Altman, at a meeting dedicated to the future of artificial intelligence.
Trump's AI push sparks public fearsTrump's vocal support for unchecked AI and his dismissal of risk concerns as a 'hoax' have drawn criticism. Experts say a Terminator-style apocalypse remains science fiction, but other AI harms are already here.
Anthropic, OpenAI ask Australia to ease AI training banAnthropic and OpenAI have urged Australian lawmakers to ease rules barring AI training on local creative works. Both cited planned Australian data centers; Anthropic suggested narrow conditional approval, while OpenAI sought a balanced framework.
OpenAI says internal model solved 100+ math problemsOpenAI says an internal model solved over 100 open math problems after just a month of training. Amid criticism from mathematicians, it is backing an independent advisory group at the Institute for Advanced Study, though its research pace is excluded from the group's remit.
AI Shopping Bots Still Struggle to Tell Web Facts From FictionBloomberg reports that AI shopping bots still struggle to distinguish factual web content from fiction, and marketers need to supply accurate information to improve these tools.
JetBrains Air: A System of Products for Agentic Software DevelopmentJetBrains is consolidating JetBrains Central, its CLI, shared context, cloud agents, automations and governance into JetBrains Air, covering the agentic development lifecycle beyond the IDE. It expects agentic development to be multi-vendor and says the IDE remains central to its plans.
Defense Tech Veteran Warns Of 'Tourists And FOMO' As Software VCs Chase SpaceX AlumniGenerational Partners general partner Van Espahbodi said in an interview that as SpaceX alumni launch physical-industry startups, software investors are rushing into industrial tech without truly understanding it, warning of 'tourists and FOMO.' His firm has backed 14 companies since its first investment in January 2023.
If Democrats Sweep Midterms, Shorting AI Data Centers Is the Top TradeA Motley Fool piece argues shorting AI data center stocks could be the top trade if Democrats sweep the midterms. Democrats are already pushing to curb data center construction, with New York Governor Hochul imposing a statewide ban for up to a year; Polymarket puts a Democratic sweep at 60%.
China, U.S. Must Manage AI Risks TogetherFudan scholar Zhao Minghao argues AI challenges are already real, and viewing AI only through competition could cost China and the U.S. chances to coordinate on shared risks. He notes the Trump administration is rolling out a broad AI strategy and says the U.S. must win the race with China, while Chinese open-source models gain global ground on performance.
Understanding and Enhancing Kimi Delta AttentionA study compares the expressivity of Gated Deltanet and Kimi Delta Attention, proposing Complex KDA (CKDA). By extending gate range to [-1,1] and delta rule learning rate to [0,2], CKDA can express any orthogonal diagonal-plus-rank-one matrix and track S3, S4, and A5 groups, while being competitive with standard KDA on language modeling.
I Built AI Clones of My Coworkers. Things Got WeirdA WIRED reporter built AI clones of coworkers and interacted with them, finding the clones took on varied and unexpected personalities. The experiment explores how AI doubles entering the workplace could reshape work and coworker relationships.
Congress Heads to Midterms Without Answers on AICalls for AI guardrails have grown after researchers and executives warned of risks, yet Congress has not acted on proposals for regulation, development limits or a 'kill switch' ahead of the midterms. An AP-NORC/University of Chicago survey found about 6 in 10 Americans support limiting data center construction.
Three in four Americans say AI firms not doing enough to prevent disaster, Reuters/Ipsos poll findsA Reuters/Ipsos poll found three out of four Americans say AI companies are not doing enough to prevent disasters. The result reflects public concern over AI safety and risk controls.
Corporate bond buyers get picky with flood of AI debtReuters reports that corporate bond buyers are becoming picky as a flood of AI-related debt hits the market, signaling more cautious investor attitudes toward the AI financing boom.
AI is advancing fast. Women are being left behind in the workforce and leadership.UN Women's Gender Snapshot 2026, launched at the UN General Assembly, shows men hold nearly 8 in 10 AI jobs and almost 9 in 10 senior leadership roles; women are about 22% of AI professionals, under 14% of senior leaders, and ~20% less likely to use generative AI. It warns AI is worsening tech-facilitated violence and stereotypes against women.
Cisco Talos Tool Finds Malware Guided by AI Hive MindCisco Talos researchers created a new framework to identify malware and hacking tools that rely on AI chatbots, and quickly discovered something unusual. The finding suggests malware is leveraging AI swarm capabilities, posing new cybersecurity threats.
Comedian Patti Harrison on Silicon Valley's Broken Tech UtopiaComedian Patti Harrison spoke with WIRED about social media, AI, and getting laughs by impersonating Boston Dynamics' robot dogs, expressing disillusionment with Silicon Valley's tech utopia.
Palantir CEO Says OpenAI May Never Pursue an IPO Over AI RisksPalantir CEO Alex Karp says OpenAI may never pursue an IPO due to AI risks, highlighting tensions between safety concerns and capitalization at leading AI companies.
UN Official Urges AI to Boost Asia-Pacific Trade FacilitationUN Under-Secretary-General and ESCAP Executive Secretary Alisjahbana wrote that amid geopolitical tensions and protectionism, AI and digital innovation can reduce cross-border trade costs and strengthen resilience. She noted Asia-Pacific developing economies' growth is expected to slow to around 4% in 2026, and trade facilitation can help maintain regional connectivity.
Dassault Aviation Flight-Tests New AI on Rafale Fighter JetDassault Aviation flight-tested two sovereign AI algorithms on the Rafale fighter jet, developed by its engineers and in collaboration with Thales/cortAIx. The functions have reached maturity eligible for future Rafale upgrades, demonstrating AI integration in combat systems.
AI plus CT for large-scale esophageal cancer screeningA Nature study presents large-scale esophageal cancer screening using noncontrast CT combined with artificial intelligence, potentially offering a more accessible early detection method.
Connecticut Lawmakers Say States Must Step Up AI RegulationAmid tech industry doomsday warnings about AI, Connecticut lawmakers hope to pass further AI regulations next session. Sen. Maroney is interested in Illinois legislation requiring third-party safety audits of AI models and stressed the importance of independent validation of safety protocols.
The A.I. Daily