AI Reading for Monday October 6

OpenAI and AMD have reached a deal that could see Sam Altman’s company take a 10% stake in the chipmaker. OpenAI will deploy up to 6 gigawatts of AMD Instinct GPUs over multiple years, beginning with a 1-gigawatt rollout in 2026. - CNBC

Coming right after a $100b OpenAI deal with Nvidia, it makes you wonder if this was the plan all along and they were transparent. Or if AMD made an offer they couldn't refuse to crash their party. Maybe they should do a deal with Google and AWS too, to make one giant Big Tech trust. MAGNA will be a real company, or OMAGNA (OpenAI, Microsoft, AMD, Google, Nvidia, Amazon)

Do We Need to Break Up Big AI Before It Breaks Us? - TIME

America is now one big bet on AI, seen as the magic fix for every threat to the US economy - Financial Times

Odd Lots on what the data center scale-out ends up looking like - Bloomberg

xAI buying another $18b worth of Nvidia chips. - The Wall Street Journal

A Meta alum is walking the earth looking for Stargate locations - CNBC

Hedge funder warns of economic risk due to rapid obsolescence of AI infrastructure, comparing AI buildout to historic capital-heavy projects like railroads, with potential for huge capital destruction. AI data centers depreciate in 3 to 10 years, increasing the required revenue to justify them. - MarketWatch

Apple is seeking a new AI chief to replace John Giannandrea and internal candidate Mike Rockwell is now seen less likely to succeed him. The company is said to be considering a senior AI executive from Meta, marking a role reversal from Meta's historic recruitment of Apple's AI employees. - Apple Insider

Apple may have tried to hire a top AI exec from Meta, John Ternus seen a frontrunner to succeed Tim Cook. - Bloomberg

Could the Jony Ive device be Friend 2.0? Key challenges include perfecting the device's voice, personality, engagement timing, addressing privacy concerns of continuous listening, and sufficient computing resources to support the AI functions. - The Verge

Sora, the 'TikTok for deepfakes,' now allows users to limit how their AI-generated doubles can be used, including restricting appearances in political videos or usage of certain words. - The Verge

AI Slop Is Everywhere. What Happens Next? It's not just cat soap operas anymore. - The Wall Street Journal

Sora said to have banned South Park videos because people were making full fake episodes - Reddit

I feel like AI might generate soap operas and South Parks and pro wrestling as good or better than the real thing.

Adobe's Firefly AI Video Generator can now transform still images and illustrations into lifelike, animated videos using text prompts, with customizable video styles from cinematic to anime, integrated in Photoshop and Premiere Pro. - Tech Times

Animating Van Gogh, art or abomination? - Kuriositas

OpenAI's GPT-5 and GPT-oss models were found to have been trained on data that includes phrases from adult websites, GitHub repos, political sites, and spammy content, revealed through analysis of their tokenizer's embedding matrix and specific glitch tokens. - Fi-Le

AI won't just eliminate millions of jobs. It will also create millions - Fast Company

A Stanford University study reveals artificial intelligence works best when used to automate dull workplace tasks. - Fast Company

ChatGPT changes how we write professionally, polished, formal, or stylized language previously considered professional is now often viewed with suspicion, prompting some to intentionally degrade their writing for perceived authenticity. - Business Insider

Microsoft's Copilot is helping workers perfect the ancient art of doing sweet f all and generating corporate work slop - The Register

GotBanned3rdTime says it so it must be true, top performers get bullied for not using AI enough, CEO wants to replace seniors with juniors + AI. - Reddit

The real deadline isn’t when AI outsmarts us — it’s when we stop using our own dang minds, and human cognitive skills decline, leading to deteriorating deep thinking abilities and functional illiteracy. - The Argument

Principles of AI Agent Economics, what Alfred Marshall would write today? - arXiv.org

Microsoft and the Department of Energy's Pacific Northwest National Laboratory used AI and the Azure Quantum Elements platform to identify a new solid-state electrolyte, NaxLi3−xYCl6, reducing lithium use by about 70%, potentially enabling safer, high-density solid-state batteries. - TechSpot

Chinese team uses AI to find antibiotics that are effective against multiple drug-resistant bacteria. - Nature

The Earth BioGenome Project (EBP) aims to sequence the genomes of all 1.8 million named eukaryotic species by 2035, enabling unprecedented insights into evolution and biodiversity. - IEEE Spectrum

AI is accelerating nuclear fusion research by improving plasma control, addressing instability issues like Edge Localized Mode (ELM) that disrupt experiments. - Yahoo

Inside the U.S. Air Force's plans to operate drones piloted by artificial intelligence, including loyal robot wingmen, alongside aircraft flown by humans. - CBS News

Bot Auto completed the first real-world driverless autonomous trucking run in Houston without a backup driver or remote operator, managing lane merges, traffic interactions, and logistics timing fully autonomously. - BGR

North Korean agents pretending to be IT guys and using AI to fool Western companies have funneled up to $1 billion into Kim Jong Un's nuclear program - Fortune

Homeowners in Singapore are increasingly using generative AI tools to plan and visualize home renovations, enabling easy furniture rearrangement and design customization via mobile devices, enabling cost savings, flexibility and control over home design. - South China Morning Post

The Otonomus Hotel: "AI doesn't make our hotel feel cold" despite being mostly AI-powered and robot-run - usa.boutiquehotelier.com

Using AI to forecast bird migrations - Hacker Noon

AI girlfriend apps simulate better-than-real adult interactions with virtual companions with customizable traits and appearances. Platforms claim to reduce exploitation associated with adult content. Critics highlight ethical concerns about harmful stereotypes and user behavior and representation. - the Guardian

AI gap? AI will be a major factor in the 2026 elections, with Republicans leveraging AI for personalized messaging, persuasion, and campaign strategy. Trump's AI-generated memes and influence over AI development illustrate a strategic political realignment toward Republican values in Big Tech. - Schneier on Security

AI political ads increasingly seen as inauthentic. - AS USA

The anonymous 'No Moore' social media accounts, using AI-generated images to disparage Maryland Gov. Wes Moore, have formed an anonymous federal political action committee (PAC) but remain unregistered at the state level, raising legal, transparency, and accountability concerns. - The Banner

Follow the latest AI headlines via SkynetAndChill.com on Bluesky