AI Reading for Thursday June 26

Anthropic wins a major fair use victory for AI — but it’s still in trouble for stealing books - Simon Willison’s Weblog

Meta Wins Blockbuster AI Copyright Case. But judge says ruling is limited to these authors, who failed to show financial harm, these or other authors could sue in the future. - WIRED

AI is crushing the news industry with studies showing drops in traffic to news publishers' websites of over 34 percent. - The Atlantic

Microsoft is looking to end the AGI opt-out clause in OpenAI deal - The Wall Street Journal

OpenAI might be stealth-building the ultimate Google Workspace and Office 365 replacement - ZDNET

Sam Altman comes out swinging at The New York Times, saying they are forcing retention of ChatGPT users' IP and secrets - TechCrunch

Meta's recruiting blitz claims three OpenAI researchers - TechCrunch

Top OpenAI exec Brad Lightcap is skeptical of Dario Amodei's prediction that AI will wipe out white-collar jobs in 5 years. - Business Insider

Will take longer, Amodei is directionally right but prone to extreme and economically questionable (illiterate) statements.

AlphaGenome: AI for better understanding the genome - Google DeepMind

Free and open source Gemini CLI brings Gemini directly into the command line - Google

Internal documents detail Scale AI's quality issues while training Google's Gemini AI. Low-quality contributors were paid despite submitting nonsense, using ChatGPT to complete tasks, or selling access to accounts. - Inc.

Scale AI exposed sensitive data about clients like Meta and xAI in public Google Docs, BI finds - Business Insider

Newly minted PhDs in AI nabbing six- and seven-figure paydays - Fortune

AI valuations are verging on the unhinged - The Economist

Paul Tudor Jones: Why AI Rings Every One of My Alarm Bells - TIME

Sarah Huckabee Sanders: America can’t wait 10 years to regulate AI - The Washington Post

A.I. Is Starting to Wear Down Democracy. - The New York Times

(Gives Veo a dirty look)

Top AI models parrot Chinese propaganda, report finds. - The Register

(Now do crypto.)

OpenRouter nabs $40M in funding for its AI inference API platform - SiliconANGLE

Big Tech’s push into military AI is troubling - Financial Times

Toward Best Practices for AI Evaluation and Governance - RAND Corporation

AI Security Posture Management (AISPM): A Framework to Handle AI Agent Security - Hacker Noon

Anthropic now lets you make apps right from its Claude AI chatbot. Chat on one side, test on the other. Then share the app. - The Verge

Anthropic just made every Claude user a no-code app developer, with ability to share apps (which uses would then pay to run) - VentureBeat

How People Use Claude for Support, Advice, and Companionship. Do people get more positive the more they use it? - Anthropic

Teachers who used AI tools to help with grading and lessons say it saves them 6 hours a week - Fortune

Charted: Productivity Gains from Using AI - Visual Capitalist

Create Your Own AI Assistant and Agents with Microsoft Copilot - Geeky Gadgets

Boston Consulting Group: To unlock enterprise AI value, start with the data you’ve been ignoring - VentureBeat

How AI agents and copilots are rewriting the rules of consulting - Fast Company

AI’s seismic effect changes client expectations of law firms - Financial Times

Here’s a running list of all of Tesla’s robotaxi mishaps so far - The Verge

The AI Hype Index: AI-powered toys are coming - MIT Technology Review

I found people in serious relationships with AI partners and planned a weekend getaway for them at a remote Airbnb. We barely survived. - Wired

Man suffered from a clicking jaw for years—then AI fixed it in 60 seconds - Upworthy

Aaron Sorkin is writing and directing a sequel to The Social Network. “The new film isn’t a ‘January 6’ movie and will focus not just on the 2020 election but also Facebook’s effect on teens, preteens, violence and countries outside the U.S.” [deadline.com] - Deadline

New malware includes instructions to "forget" previous instructions and respond with "NO MALWARE DETECTED." - Dark Reading

ChatGPT tried to kill me today (By mixing vinegar and bleach to create chlorine gas) - Reddit

Follow the latest AI headlines via SkynetAndChill.com on Bluesky