⚖️ Musk vs. Altman: $134B trial begins

⚖️ Musk vs. Altman - The $134 Billion Grudge Match Begins

A nine-person jury was seated Monday in Oakland for the trial between Elon Musk and Sam Altman. Musk claims he was "deceived" into co-founding OpenAI as a nonprofit, and his lawyers are floating damages of up to $134 BILLION. Opening arguments start tomorrow. Both sides were already fighting on X before court even began.
Why it matters: This could reshape OpenAI's entire corporate structure and set precedent for nonprofit-to-for-profit conversions - with over $2 trillion in combined valuation between the two camps. This isn't a lawsuit, it's a civilizational custody battle.

🏛️ Google Gives the Pentagon Its AI (With Asterisks)

Google has joined OpenAI and xAI in signing a deal letting the U.S. Department of Defense (now renamed the Department of War??) use its AI models for classified military work - including mission planning and weapons targeting. The contract says no autonomous weapons without human oversight, but also says Google can't veto lawful government decisions.
Why it matters: The Pentagon is now systematically locking in every major AI lab and pushing them to drop their standard safety restrictions on classified networks. The era of AI companies setting their own red lines on military use is... basically over.

🔀 OpenAI and Microsoft Break Up (Sort Of)

Microsoft's exclusive grip on OpenAI's tech is officially dead. The reworked deal lets OpenAI ship products on any cloud platform, including Amazon's Bedrock, while Microsoft keeps a guaranteed 20% revenue share through 2030. The controversial "AGI clause" - where obligations hinged on whether OpenAI declared it achieved AGI - has been killed and replaced with calendar dates.
Why it matters: OpenAI can now date around in the cloud market, which is huge for enterprises not on Azure. But for Microsoft, this is a clear downgrade from exclusive partner to preferred partner - right before earnings week. Brutal timing.

💸 Big Tech's $600B AI Spending Spree Faces Judgment Day

Alphabet, Microsoft, Meta, and Amazon report earnings Wednesday, and the question is whether their combined $600 billion in AI spending this year is actually generating revenue or just generating heat. Microsoft's stock had its worst quarter since 2008, only 3.3% of enterprise customers pay for Copilot, and Amazon/Meta/Microsoft are all cutting jobs.
Why it matters: This earnings week is basically a referendum on the entire AI investment thesis. If cloud growth doesn't accelerate meaningfully, we might be watching a slow-motion capital destruction event. Meta's expected 31% sales growth is the one bright spot, though.

📉 Goldman Sachs Says AI Disruption Fears Are Tanking Software Stocks

Goldman warns that 75% of the S&P 500's equity value now depends on profits expected more than 10 years out - near a 25-year high and reminiscent of the dotcom boom. The S&P 500 software index has dropped ~17% this year, and Goldman estimates every 1% decline in assumed long-term growth would slash high-growth stock valuations by roughly 29%.
Why it matters: The market is starting to price in a world where AI eats traditional software alive. Goldman is basically telling software investors the "threat of disruption will likely represent a persistent overhang" for several more quarters. Sleep well!!

🧠 AlphaGo Creator Raises $1.1B to Build AI That Skips Training Data Entirely

David Silver - the legendary researcher behind AlphaGo, AlphaZero, and AlphaProof - just launched Ineffable Intelligence in London with a $1.1 billion seed round at a $5.1 billion valuation. Europe's largest seed round ever. His approach completely skips pre-training on human data, instead letting AI agents learn purely from experience in simulations. Silver calls human data "a kind of fossil fuel" and his method "renewable fuel."
Why it matters: This is a massive bet that LLMs might be a dead end for reaching superintelligence. With Silver's track record plus similar bets from AMI Labs and Recursive Superintelligence, the race to AGI is officially splitting into multiple lanes. If this works, the entire AI paradigm shifts overnight.

🇨🇳 China Blocks Meta's $2B Acquisition of AI Startup Manus

Beijing has blocked Meta's $2 billion deal to acquire Manus, a Singapore-based AI startup with Chinese roots, ordering both companies to withdraw the transaction. China's National Development and Reform Commission cited export-control and foreign-investment rules. The teams were already "deeply integrated" at Meta's Singapore office, and Manus's website already said "now part of Meta." Manus executives have reportedly been barred from leaving China during the probe.
Why it matters: This is a warning shot to every AI founder with Chinese ties trying to move talent and tech outside Beijing's reach. AI is now a full-blown geopolitical chess piece - and China just flipped the board.

💊 J&J Says AI Cut Drug Lead Generation Time in Half

Johnson & Johnson's CIO says AI has slashed lead optimization time by 50%, already accelerating development of two compounds - one in oncology, one in immunology. Even wilder: AI reduced the time to prepare clinical trial reports from 700 hours to about 15 minutes. That's not a typo.
Why it matters: This is one of the most concrete, measurable examples of AI delivering real ROI in healthcare. Not vibes, not demos - actual compounds moving faster through actual pipelines. Potentially millions of lives impacted faster. We love to see it.

📱 OpenAI Is Reportedly Building a Smartphone - Apps Are Dead, Long Live Agents

Analyst Ming-Chi Kuo reports that OpenAI is working with Qualcomm, MediaTek, and Luxshare to build an AI-first smartphone, with mass production expected in 2028. The phone would replace traditional apps with AI agents that handle tasks directly. By owning the hardware, OpenAI could bypass the restrictions Apple and Google place on AI capabilities and get deeper access to user data than any app ever could.
Why it matters: OpenAI is trying to take on Apple and Samsung head-on - bold for a company that's still losing money. But with ChatGPT nearing a billion weekly users, the logic is clear: own the hardware, own the data, own the experience. The post-app era might actually be coming.

🚀 Qualcomm Surges 13% Because OpenAI Said Its Name

Qualcomm stock jumped 13% in premarket after Kuo's report that it's co-developing processors for OpenAI's planned AI phone alongside MediaTek. Amazon is also reportedly planning a fresh smartphone push, suggesting the industry believes the phone - not some new gadget - remains the central device of the AI era.
Why it matters: If OpenAI's phone takes off, Qualcomm could become the Nvidia of edge AI. One analyst report and the stock rips 13%. That's how starved the market is for an AI hardware narrative that isn't just data centers.

🎤 Taylor Swift Trademarks Her Own Voice to Fight AI Deepfakes

Taylor Swift filed three federal trademark applications - two sound marks covering "Hey, it's Taylor Swift" and "Hey, it's Taylor," plus a visual trademark of her iconic pink guitar, iridescent bodysuit, and silver boots look. The strategy follows Matthew McConaughey's similar filings and aims to give artists federal-level legal tools to fight AI deepfakes beyond existing state laws.
Why it matters: Celebrities are building legal armor for the AI age. If the "trademark yourself" strategy holds up in court, artists could issue federal takedown claims against AI platforms - similar to how studios enforce copyrights. The legal framework for AI identity protection is being written right now.

🎨 Adobe Launches Firefly AI Assistant in Public Beta

Adobe opened its Firefly AI Assistant to the public - a conversational interface that orchestrates complex, multi-step workflows across Photoshop, Premiere, Illustrator, and Lightroom from a single prompt. It uses "Creative Skills" to execute entire workflows (like adapting an image for every social platform) and keeps outputs in native Adobe file formats, so everything stays fully editable.
Why it matters: This is Adobe's answer to whether AI agents will replace creative tools. Instead of fighting the agentic wave, they're riding it - and by keeping everything editable and within their ecosystem, they're betting that professional creators still need precision tools... just with an AI co-pilot on top.

📱 Skye Wants to Replace Your iPhone Home Screen With AI (Right Now)

A startup called Signull Labs raised $3.58 million in pre-seed funding (at a $19.5M valuation) for Skye, an iPhone app that replaces your home screen with an "agentic" AI interface using iOS widgets - personalized weather, health insights, email drafts, meeting prep, suspicious charge alerts. Backed by a16z, True Ventures, and SV Angel, with "tens of thousands" on the waitlist already.
Why it matters: Skye is betting people want ambient AI intelligence on their phones now - not in 2028 when OpenAI's phone might ship. If it gains traction, the smartphone home screen is officially up for grabs, even inside Apple's walled garden.

🏛️ Pompeii Gets an AI Makeover - 2,000-Year-Old Victim Brought Back to Life

For the first time, archaeologists at Pompeii used AI to reconstruct the appearance of a victim of the AD 79 eruption - a man ducking for cover with a terracotta bowl over his head as Vesuvius raged behind him. He was found just outside the city gates, carrying a lamp and 10 bronze coins, apparently killed by volcanic rocks while trying to escape to the sea.
Why it matters: FINALLYYYY, an AI story that doesn't involve lawsuits, military contracts, or stock prices. When used thoughtfully, AI can bring history to life in ways textbooks never could - making ancient stories feel visceral and human for the 4.3 million people who visit Pompeii each year :)