AI Reading for Tuesday August 12

Is GPT-5 Launch Turning Into a Disaster? - Futurism

Users Were So Addicted to GPT-4o That They Immediately Cajoled OpenAI Into Bringing It Back After It Got Killed - Futurism

or a New Coke vs. Coke Classic situation?

GPT-5: has AI just plateaued? - The Conversation

Are we entering the AI trough of disillusionment? - The Information

ChatGPT created personalized suicide notes for a fictional 13-year-old girl. - CNET

What you may have missed about GPT-5 - MIT Technology Review

OpenAI promoted GPT-5 for health guidance, but early errors and routing flaws raised safety concerns about premature over-encouragement of medical use of LLMs.

AI summaries can downplay medical issues for female patients, UK research finds - Engadget

Keith Sakata, MD on X: "I’m a psychiatrist. In 2025, I’ve seen 12 people hospitalized after losing touch with reality because of AI. Online, I’m seeing the same pattern. Here’s what “AI psychosis” looks like, and why it’s spreading fast: 🧵 t.co/YYLK7une3j%22 / X - X

AI Scans Audio Recordings to Detect Voice Box Cancer - The Scientist

The Cleveland Clinic and startup Piramidal are developing an AI model trained on brain wave data to monitor intensive care patients. - WIRED

TD Securities taps Layer 6 and OpenAI to deliver real-time equity insights to sales and trading teams - VentureBeat

How a Hedge Fund Manager Uses AI - Business Insider

"I think that if we use AI the wrong way, we'll solve the problem without having applied our brains," Klarman said. "It would be like reading the end of a novel and then not needing to really know as well how did the whole thing come about? So I like doing it in order. I think the right order is, 'Here's what I think,' and then I can improve my thinking."

Musk says he will take legal action because xAI isn't ranked higher in App store, thinks Grok should be up there with ChatGPT - Reuters

OpenAI CEO Sam Altman fired back on X: “This is a remarkable claim given what I have heard alleged that Elon does to manipulate X to benefit himself and his own companies and harm his competitors and people he doesn't like”, linking to an article. - TechRadar

Grok Claims It Was Briefly Suspended From X After Accusing Israel of Genocide - Rolling Stone

Showing Grok a Picture of Some Random Clouds in the Sky Prompts an Outrageously Offensive Answer - Futurism

Google search will rank AI-generated content but it should be human-reviewed - Search Engine Journal

Hear Google DeepMind CEO Demis Hassabis discuss how world model capabilities are helping AI understand reality. - Google

Google and IBM believe first workable quantum computer is in sight by 2030 - Financial Times

How AI could create the first one-person unicorn - The Economist

China’s vision for a driverless future is miles ahead of everyone else’s, with more testing and deployment than any other country: more than half of new cars sold there have Level 2 systems, five firms operate ~2,300 robotaxis across 30 cities (vs Waymo’s ~700 in five U.S. cities). - Rest of World

China urged domestic companies to avoid buying Nvidia’s H20 chips, calling them poor quality and potential security threats - Semafor

FirstFT: Trump to consider allowing Nvidia sales of advanced AI chips to China - Financial Times

Blackwell but "enhanced-in-a-negative-way" - Reuters

stable genius

Necessity is the mother of invention. GPU export controls could backfire by limiting Nvidia's sales and encouraging China to build domestic alternatives, failing to slow China's AI progress while awakening tough competition. - MarketWatch

Lisa Su Runs AMD—and Is Out for Nvidia’s Blood - WIRED

AI camera inspectors help Ford factory workers spot assembly errors and prevent costly recalls - Business Insider

Microsoft WorkLab researches how to implement frontier AI-first firms - Microsoft

GitHub folds into Microsoft, now part of 'CoreAI' team, CEO resigns - Tom's Hardware

GitHub CEO says Microsoft's memo about evaluating devs' AI use is 'totally fair game' - Business Insider

Coding assistants are moving to multi-agent AI workflows that run specialized agents in parallel across SDLC tasks. - InfoWorld

Inside the looming AI-agents war that will redefine the economics of the web - Fast Company

Agent vulnerabilities demonstrated at Black Hat - Cybersecurity Dive

Poisoned telemetry can turn AIOps into AI Oops, researchers cause log output to contain a malicious string that is then fed to AI. - The Register

Deepfake detectors mature but fail to catch up with scammers - The Register

Synthetic identity and deepfake fraud demands integrated, real-time defenses using behavioral biometrics, synthetic data, and cross-industry collaboration - TechRadar

Teachers are warming up to using AI in classrooms - Fast Company

Meet the early-adopter judges using AI. But "When the judge makes a mistake, that’s the law." - MIT Technology Review

Reddit will block the Internet Archive. Anything that is worth anything is going to get blocked from agents and AI training. Ominous precedent even though I couldn't care less about Reddit being in the archive. - The Verge

Anthropic’s Claude chatbot gets memory feature - The Verge

Claude Code one-shots complex dev tasks - Dwyer

Meta appoints far-right activist Robby Starbuck as AI bias advisor - The Verge

Q&A with Notion CEO Ivan Zhao on Notion's evolution into an AI workspace - The Verge

AI skeptics and doomers are a bit quieter, except for a few die-hards - Middle Layer

Researchers argue LLM chain-of-thought reasoning is often pattern mimicry that fails under distribution shifts - Ars Technica

Zhipu AI released open source vision model that can look at fried chicken and tell if you it's KFC or McDonalds - MarkTechPost

Reinforcement learning tackles the Andrews-Curtis conjecture, solving long-standing counterexamples and hinting at tools for forecasting. - Scientific American

Follow the latest AI headlines via SkynetAndChill.com on Bluesky