Welcome to tl;dr ai

Your daily AI-powered summary of what matters.

News today:

OpenAI Subpoenas Meta for Zuckerberg-Musk Coordination Docs on $97 B ChatGPT Bid

TechCrunch •

OpenAI subpoenaed Meta for any documents showing coordination between Mark Zuckerberg, Elon Musk and xAI on Musk’s unsolicited $97 billion bid for the ChatGPT maker. OpenAI’s team alleges Musk discussed financing with Meta, prompting a legal fight over relevant evidence. Meta has pushed back, calling the request irrelevant and pointing to Musk’s camp for answers.

Read original ↗

Microsoft, NFL Extend Partnership to Bring Copilot AI to Sideline Surface Tablets

The Verge •

Microsoft and the NFL have extended their partnership, adding AI smarts to the familiar Surface tablets on the sidelines. Coaches and players can now tap GitHub Copilot–powered features that sift through play options, while 365 Copilot dashboards help analysts crunch game data faster on over 2,500 Surface Copilot Plus PCs. Future AI plans include scouting prospects and boosting teams’ business operations.

Read original ↗

Wired and Insider Retract AI-Generated Freelance Articles by “Margaux Blanchard”

The Guardian •

Wired and Business Insider have removed multiple freelance articles by “Margaux Blanchard” after discovering they were AI-generated. At least six outlets yanked pieces—including a Minecraft wedding feature at Wired and parenting essays at Insider—once Press Gazette flagged authenticity issues. Blanchard’s unverifiable sources, like a Chicago “digital celebrant” and a secret Colorado forensic training town, sparked the probe. The incident highlights mounting challenges in spotting AI-produced journalism and enforcing editorial standards.

Read original ↗

NHTSA Audits Tesla for Months-Long Delays in Autopilot Crash Reporting

The Verge •

Tesla’s under NHTSA investigation for months-long delays in reporting crashes involving its Level 2 Autopilot and Full Self-Driving features. Federal rules require reporting within 1–5 days; Tesla blames data glitches now fixed. The agency has launched an audit to ensure compliance with a 2021 standing general order that has seen Tesla account for nearly all reported fatal ADAS crashes.

Read original ↗

CBA to Rehire 45 Staff After AI Voice Bot Fails to Cut Call Volumes

Ars Technica •

Commonwealth Bank of Australia reversed course and offered to rehire 45 veteran customer-service staff after its AI ‘voice bot’ failed to reduce call volumes as claimed. A union tribunal found CBA’s redundancy justification false, prompting an apology and options to return, redeploy or exit with a payment. The bank still plans further AI partnerships but insists there’ll be no more cuts.

Read original ↗

Amazon’s AGI Lab Head David Luan Unveils ‘Gyms’ for Reinforcement-Learning AI Agents

The Verge •

In a wide-ranging Decoder interview, Amazon AGI lab head David Luan—an OpenAI and Google AI veteran—explains why “agents”—AI systems that learn by trial and error in simulated workflows—are the next big step beyond chatbots. Amazon is building “gyms” of digital work environments (from CRM to CAD) for reinforcement-learning agents to master real tasks with high reliability. Luan left his startup Adept for Amazon’s massive compute scale, betting that training true action-driven AI agents, not just language models, will redefine computing and unlock vast new productivity gains.

Read original ↗

Google AI Mode Expands to 180 Countries, Adds Restaurant Booking & Personalized Suggestions

The Verge •

Google is expanding AI Mode in Search to 180 countries, still in English. Subscribers can now use agent skills to book restaurants—specifying date, time, party size, location, and cuisine—and link directly to reservation platforms like OpenTable and Resy. US testers in Labs can get personalized dining suggestions based on past searches. Plus, all US AI Mode users can share their chat sessions via a public, manageable link.

Read original ↗

Experts Warn Google’s Gemini AI Efficiency Claims Underplay Data Center Impact

The Verge •

Google says its new Gemini AI uses only about 0.26 ml of water (five drops) and 0.24 Wh of electricity per text prompt—roughly nine seconds of TV viewing—resulting in just 0.03 g of CO₂. But experts warn these figures underplay the true environmental impact of AI data centers.

Read original ↗

South Park S27 Premiere “Sickofancy” Skewers ICE Raids, AI Craze and Trump’s DC Police State

The Guardian •

South Park’s 27th season kicks off with “Sickofancy,” a sharp satire of ICE raids, Trump’s martial DC makeover and our AI obsession. After border agents raid Randy Marsh’s Tegridy Weed farm, he turns to ChatGPT, launching “Techridy,” an AI-powered pot venture. Meanwhile Trump transforms Washington into a dystopian police state. The episode skewers AI, tech bros (hello, Tim Cook and Zuckerberg) and Trump’s photo-op politics, ending with Randy forced to sell his farm and retreat to suburbia.

Read original ↗

10 Free AI Tools to Supercharge Your 2025 Instagram

Mpost •

Instagram in 2025 is more competitive than ever, but these 10 free AI tools help creators, brands, and businesses automate content creation, scheduling, and editing. From Instagram’s native Meta AI assistant and the standalone Edits video app to third-party generators like Adobe Express, Hootsuite’s OwlyGPT, and Ahrefs’ caption tool—plus schedulers Buffer and Meta Business Suite Planner, and versatile editors Fotor and Picsart—these solutions streamline captions, visuals, and posting, boosting engagement and creativity at zero cost.

Read original ↗