AI Reading for Wednesday October 15

BlackRock-led consortium including GIP, MGX, Nvidia and Microsoft agreed to acquire Aligned Data Centers from Macquarie for $40 billion and said it will expand the company’s campuses to meet rising AI computing demand. - Yahoo Finance

Nscale signed an up-to-$14 billion deal with Microsoft to deploy about 104,000 Nvidia GB300 chips in Texas within 18 months and 12,600 GPUs in Portugal, and its CEO said the company targets a public listing late next year. - Financial Times

A Giant New AI Data Center Is Coming to the Epicenter of America’s Fracking Boom: CoreWeave and Poolside announce partnership for a data center built on a sprawling ranch in West Texas - The Wall Street Journal

Freshly appointed dual chief executives Clay Magouyrk and Mike Sicilia push back on concerns over Oracle’s margins as investors fear AI bubble. - The Wall Street Journal

Trump calls wind and solar a 'joke.' JP Morgan says it's crucial for making AI work. The bank says the U.S. will struggle to meet growing power demands—and stay energy-independent—without renewable projects spurned by the administration. - Fast Company

The Frothiest AI Bubble Is in Energy Stocks, Nukes. Concept stocks with no revenue have soaring valuations. OpenAI CEO Sam Altman has backed the biggest of these energy companies. - The Wall Street Journal

AI economics are brutal but demand keeps soaring. - The Wall Street Journal

U.S. business investment has surged as companies invest in data centers, AI chips, automation and cloud infrastructure, driven by AI adoption, Trump-era tax cuts and consumer spending, boosting semiconductors, enterprise software and construction; tariffs and immigration crackdowns pose risks. - The Washington Post

The IMF raised U.S. growth forecasts to 2.0% (2025) and 2.1% (next year), warned tariffs are dimming prospects, said AI investment has materially offset tariff drag and likened the surge to the late-1990s tech boom. - Fortune

ASML reported a surge in orders for its advanced chip-making machines as artificial intelligence drives demand. - Bloomberg

Jensen Huang said 100% of Nvidia's software engineers and chip designers use Cursor, name-checked 5 other AI startups—OpenAI, Harvey, OpenEvidence, Replit, and Lovable - Business Insider

Nvidia DGX Spark runs large models pretty fast: Hands-on with the mini AI workstation powered by the GB10 SoC (Blackwell GPU and 20‑core Arm CPU), featuring 128 GB unified LPDDR5x, 6,144 CUDA cores and 1 petaFLOP sparse FP4, priced $3,000–$3,999. - The Register

Goldman Sachs warns the US may be entering 'jobless growth' as AI boosts GDP and productivity while job growth outside healthcare fell, citing ADP −32,000 jobs, a 17.2% year‑over‑year drop in openings and Challenger hiring plans at lowest since 2009. - Business Insider

Goldman Sachs told staffers it would limit hiring, and to expect additional job cuts as the bank takes advantage of AI and seeks further savings, but still expects to finish the year with an increase in overall headcount. - Bloomberg

Hunter Biden warns of possible job ‘mass extinction event’ in US — says AI could destroy 3.5M jobs across country - Yahoo Finance

OpenAI has drawn up a five-year business plan to meet $1 trillion in spending pledges, targeting new revenue lines, debt partnerships and further fundraising without providing detailed revenue breakdowns or timelines. - Financial Times

New paper reveals OpenAI wants to stop ChatGPT from validating users’ political views by mirroring their framing. - Ars Technica

OpenAI unveils “wellness” council; suicide prevention expert not included; eight-member Expert Council on Wellness and AI, will guide ChatGPT mental-health upgrades and advise on parental controls, conversation design, and youth developmental risks after a lawsuit alleging a teen 'suicide coach.' - Ars Technica

Sam Altman said they nerfed ChatGPT personal advice chats over mental health criticisms, will gradually relax restrictions, make restrictions more specific, add age checks, and allow adult chats. - Reddit

Specifically 'In December, as we roll out age-gating more fully and as part of our “treat adult users like adults” principle, we will allow even more, like erotica for verified adults.'

Japan's cabinet minister Minoru Kiuchi asked OpenAI to adjust Sora 2 to avoid copyright infringement on 'irreplaceable' manga and anime content - Engadget

New York judge orders OpenAI to hand over Slack messages tied to a LibGen-sourced book dataset - Futurism

WPP struck a $400 million deal with Google for access to the Gemini model and Veo video generator, enabling large-scale generative-AI tools for automated creative and ad production. - Financial Times

Google is adding Gemini-powered 'Help Me Schedule' to Gmail, which detects meeting plans, creates an inline widget with suggested times from the sender's calendar, and lets recipients pick a time to auto-schedule; group scheduling isn't supported at launch. - Ars Technica

Advocacy group SAVE-US, founded by Tesla Takedown alum, urged states to require clearer system-limit disclosures, public crash reporting and mandatory multi-sensor hardware, explicitly calling out Tesla. - WIRED

Waymo will launch a fully driverless robotaxi service in London in 2026 and will begin human-supervised data-collection runs with safety drivers in the coming weeks, offering commercial rides via its app using driverless Jaguar SUVs. - The Verge

(interesting test b/c roads seem somewhat orderly unlike NYC chaos, but narrow and very complicated to navigate)

Johnson County Sheriff's Office used Flock Safety's 83,000-device ALPR network to run a nationwide AI-enabled search for a Texas woman alleged to have self-administered abortion medication; EFF and 404 Media documents say deputies opened a 'death investigation' and discussed prosecution. - Crooks and Liars

California Governor Gavin Newsom signed laws including SB 243—requiring chatbot guardrails, crisis referrals, annual reporting and a private right of action—plus AB 56 social-media warning labels, a device age-verification law and new privacy rules, while vetoing two tougher bills. - Gizmodo

Ten foundations, led by the MacArthur Foundation and Omidyar Network, launched Humanity AI and pledged $500 million over five years to fund five human-centered AI priorities, pooling grants through Rockefeller Philanthropy Advisors to initial grantees including AI Now. - AP News

EdX launched a three-week experimental Foundations of Generative AI course taught by an AI avatar and graded by AI, created by Georgia Tech's David Joyner and MIT's Anant Agarwal - Fortune

An essay warns that large language models' 'cognitive abundance'—providing instant fluent thought—risks producing 'anti-intelligence' (fluency without understanding), devaluing routine expertise, raising demand for scarce human skills, and requires structuring roles and purpose to avert collapse. - Psychology Today

information that was like gold throughout human history is now free … hard for kids today to understand the value an artifact like an LP or a book had even in the 80s

Graphite found about 52% of newly published English-language articles were AI-generated in May 2025 after rising from ~10% in late 2022, based on analysis of 65,000 articles from Jan 2020–May 2025. - Futurism

Ukrainian startup Dropla's Blue Eyes AI detects landmines and drones in real time (up to 130 fps), relaying coordinates to Ukraine's battle‑management systems after combat tests since early summer. - Business Insider

The Army plans AI to scan battlefields and direct artillery and air-defense targeting, including AI-enabled fire control to reduce manpower and cognitive load, but Maj. Gen. Frank Lozano says current language models lack necessary spatial reasoning and real-time situational awareness. - Defense One

Walmart struck a deal with OpenAI to enable purchases directly through ChatGPT via OpenAI's Instant Checkout, CEO Doug McMillon said, and the announcement sent Walmart shares up nearly 5% to a 52-week high, though launch date and terms remain undisclosed. - CNBC

Orion's AI-enabled mattress cover adjusts temperature throughout the night, targets wearable users and hotels, uses a subscription model, and projects potential mass-market pricing as low as $500. - Fast Company

ChatGPT can analyze selfies to assess skin tone, eye and hair color, face shape and features, then generate personalized palettes with hex codes, color charts, mood boards and shopping recommendations, enabling DIY color analysis and potential e-commerce integration. - CNET

American Airlines launched an AI-powered destination recommender in a trial to about 50% of users that provides instant destination, flight, fare, hotel, restaurant and activity suggestions but can mislabel one-stop flights as nonstop and yield inaccurate or expensive results. - Business Insider

Augment hikes price for AI code editor, users protest; possible day of reckoning coming for big token consumers in many apps. - The Register

Train your own LLM for $100 - Simon Willison’s Weblog

Paper2Agent automatically converts research papers into interactive AI agents for reproducibility and question answering. - NextBigFuture.com

"It is as if you are making hammers in a hammer factory and one day the hammer that comes off the line says, 'I am a hammer, how interesting!' This is very unusual!" - THE DECODER

Disney is integrating AI into its parks, including a September patent for AI-driven projection animatronics and Walt Disney Imagineering trials of Meta AI glasses, while AI-generated art and merchandise prompted public backlash and some removals. - Inside the Magic

Signal migrated to quantum-ready encryption. Not a bad idea since otherwise you can capture encrypted chats, and then in 10 years when current encryption is broken, have it all decrypted. Ofc you will still be able to do that for phone calls and texts and whatnot. - Ars Technica

Follow the latest AI headlines via SkynetAndChill.com on Bluesky