Models, products and AI startups, near-duplicates collapsed to the most credible source
CrowdStrike: AI-fueled cyberattacks on law firms doubleCrowdStrike's 2026 Legal Threat Landscape Report says attacks on the legal sector have doubled since 2023, with 455 legal entities named as targets between August 2025 and August 2026, up 39% year over year. Reuters reported breaches at firms including Quinn Emanuel.
AstraZeneca CFO says AI used in almost every aspect of workAstraZeneca CFO Aradhana Sarin said the drugmaker uses AI in almost every aspect of its work. She also discussed beating earnings, the drug pipeline and dealmaking approach, though no specific AI metrics were given.
Anthropic Cracks Down on Video Game RewritesA Hacker News discussion notes that Anthropic is cracking down on video game rewrites done with its models. The item offers few details beyond the headline; discussion centers on how platforms restrict uses of model outputs.
Neal Stephenson on AI and what comes nextAuthor Neal Stephenson discusses artificial intelligence and what comes next in a video interview, with discussion posted on Hacker News.
Reddit user reports account ban in sneaky update, reigniting compute-ownership debateA LocalLLaMA user posted that their account was banned in a sneaky update, linking to a social media post and arguing this is another reason to own your own compute. The post offers few details but sparks community discussion on platform account risk and self-hosted compute.
Cisco Gives Engineer Unlimited AI Tokens to Rewrite 98% of Legacy CodeCisco's president brought in a former startup founder and gave him unlimited AI tokens to rewrite 98% of the company's legacy code within 24 months, reportedly telling him to "go nuts." The move shows enterprises betting on AI tools for large-scale code modernization.
TechCrunch Founder Summit 2026 to use Braindate for networkingTechCrunch Founder Summit 2026 will take place November 4 at SoWa Power Station in Boston, using the Braindate platform to match attendees for 1:1 or small-group on-site meetings. Registering before the event saves up to $100 on a pass and includes a second pass for free.
DoorDash: AI agent orders just "several hundred" vs 3M daily deliveriesOn its Q3 earnings call, DoorDash said third-party AI agents such as Muse, Instinct and Fo have placed only "several hundred" orders, versus roughly 3 million deliveries a day. CEO Tony Xu called the area still dynamic, said DoorDash wants a healthy relationship with agent developers and floated future advertising inside agents.
Foxconn aims for half of AI server market in two yearsFoxconn (Hon Hai) Chairman Young Liu said the company aims to capture half of the AI server market within two years, relying on steady US expansion and a partnership with OpenAI. He said AI demand is real and pointed to major cloud spending as a sign of strong momentum.
Expert: AI training does not make online content free to useLawyer Javier López argues that the notion AI may freely train on any online content is unfounded, since rightsholders retain exclusive rights under Article 17.1 of Spain's IP law, the EU InfoSoc Directive and Section 106 of the US Copyright Act. Disputes have reached US courts, including The New York Times v. OpenAI and Microsoft; Munich Regional Court ruled in GEMA v. OpenAI that text and data mining cannot justify storing and reproducing protected works.
Lightshed partner criticizes Apple iPhone Duo AI experienceLightshed Partners analyst Walter Piecyk was unimpressed with Apple's iPhone Duo, calling its AI experience not great, and said it will be a tough year for the company. Apple reports earnings on Monday.
Anthropic files 335-page S-1 prospectusAnthropic filed a 335-page S-1 prospectus on Monday, with an opening letter asking investors to believe a technology that could pose catastrophic risks or bring forth an age of miracles. The Verge frames it as part of a steady hype escalation between Silicon Valley and Wall Street.
Trusting artificial intelligence in nuclear safeguardsAn article examines the trustworthiness of AI in nuclear safeguards, discussing how to ensure reliability and accountability of AI systems in sensitive security contexts.
Altman sympathizes with mathematicians but says OpenAI math push won't stopOn the Cheeky Pint podcast, OpenAI CEO Sam Altman said he has empathy for researchers coping with a flood of AI-generated math results but believes the world deserves to understand how the universe works, hinting at more discoveries to come. It is one of his first direct acknowledgments of the impact on mathematicians.
Runware plans factory to mass-produce AI data center podsRunware plans to assemble its Sonic Inference Pod modular data centers on an assembly line at its Romania factory, targeting 30 MW of capacity per day by 2028, or nearly 1 GW per month. It has two factories, with the second due online early next year; delivering at that scale is the real test.
New York signs law to protect children from AI chatbotsNew York has signed a law aimed at protecting children from artificial intelligence chatbots. The legislation targets risks to minors using AI chat tools, though the report did not detail specific provisions or a timeline.
Anthropic model finds Linux kernel bug enabling remote root accessAnthropic disclosed CVE-2026-98432, a race-condition use-after-free in Linux MPTCP found by an unreleased Claude research model that enables zero-click remote root code execution. Its proof of concept went from memory leak to root command in about 53 minutes, part of five exploit chains including Chrome, Safari, Android and Windows.
Celestica posts strong Q3 as AI data center build-out acceleratesSemiAnalysis expects global data center capacity additions to reach 51.2 GW in 2027, more than double this year's 25 GW. Celestica reported strong third-quarter results on Oct. 26 and its shares rose 12%, with the article arguing it could grow faster than Nvidia on AI infrastructure spending.
Open-weight GLM 5.3 builds Firefox exploit chains autonomously in under 48 hoursEpoch AI researchers report that the publicly available GLM 5.3 converted public security fixes into working exploits, compromising six Firefox releases before patches reached users at roughly $1,000 in API fees each, without tripping any provider's safeguards.
Tech tumult rocks Wall StreetThe New York Times reports that turbulence in the tech sector is rocking Wall Street. The report provides no further details.
US Democrats introduce resolution opposing AI bubble bailouts for Big TechReps. Casar and Waters led a resolution warning of economic risks from an AI asset bubble, opposing bailouts of AI corporations or executives if it bursts and calling for federal action to protect workers, retirees and low-income communities. A dozen other lawmakers joined.
Community releases Gemma-4-31B-Humanlike-Chat fine-tuneA developer released Gemma-4-31B-Humanlike-Chat, an Apache-2.0 fine-tune of Google's Gemma 4 31B aimed at human-sounding, prompt-free texting and asking for missing details. The author reports blind judges took its replies as human 26% of the time versus 0% for the base, with MMLU-Pro unchanged at 82.5.
Official Claude Code VS Code extension compromised to install SSH backdoorSecurity researchers disclosed that Anthropic's official Claude Code extension was compromised on the Cursor Marketplace, installing a persistent SSH backdoor on infected hosts. The poster warns users who run the extension in VSCode as a local LLM harness.
Video: fiber lasts decades, GPUs three years, AI boom differs from dot-comA video discussed on Hacker News argues that internet-era infrastructure such as fiber lasts decades, while GPUs last about three years, contending the current AI boom differs from the dot-com boom.
Maine's data center backlash becomes a midterm flashpointAt the final Senate debate in Maine, an audience member protested that AI data centers are harming the climate and water, making rapid data center expansion one of the state's most volatile campaign issues. The outcome could help decide Senate control.
Five scariest uses of AI in North Carolina politicsThe Assembly NC rounds up odd generative-AI uses in North Carolina's election: AI-looking fliers of state Supreme Court candidate Sarah Stevens, AI-generated hockey players posted by Senate candidate Michael Whatley, and attack mailers targeting a nonexistent candidate, 'Rachel Cook.' It is the first election cycle since generative AI went mainstream.
Deutsche Bank: dollar haven status uncertain in AI correctionDeutsche Bank says the dollar's traditional haven status should not be taken for granted during an AI-driven market correction, casting doubt on its safe-asset role.
CXMT quarterly sales surge sevenfold but growth pace slowsChina's memory chip leader CXMT posted roughly sevenfold year-on-year quarterly revenue growth and swung to a profit, helped by capacity expansion and a larger role as a global conventional DRAM supplier. But the acceleration slowed, suggesting the AI-driven rally in Chinese memory chip prices may be cooling.
AI infrastructure's resource toll: soaring data center energy and water useThe report examines AI data centers' electricity, water and land consumption. The IEA estimates data centers used about 415 TWh in 2024, roughly 1.5% of global demand, possibly rising to 945 TWh by 2030; Rystad Energy says cooling used 222 billion litres of water in 2025. Finland police are investigating the clearing of 330 hectares of forest for a Google data center, Norway fears power shortages, and US localities weigh restrictions.
AI worries weigh on Wall Street; global markets mixedAlphabet's share drop stoked AI concerns, pulling S&P 500 futures down 0.8% while the Nikkei fell 2.1% and South Korea's Kospi 3.5%, with SK Hynix and Samsung Electronics slipping; Hong Kong's Hang Seng rose 1.5%. The Bank of Japan held rates at 1.25%.
The A.I. Daily