11.26 Weekly TrendDown
March 9-15, 2026
THIS WEEK’S TOP 5
01. Flying Cars Are FINALLY Here.
Electric Air Taxis Are About to Take Off in 26 States
The FAA has approved eight pilot programs that will let a handful of companies including Archer Aviation, Beta Technologies, Joby Aviation, and Wisk who will begin widespread electric aircraft testing as early as this summer. The three-year program, formally called the Advanced Air Mobility and Electric Vertical Takeoff and Landing Integration Pilot Program, spans 26 states and is designed to ensure the U.S. leads in next-generation aircraft for personal travel, regional transportation, cargo logistics, and emergency medicine, according to Transportation Secretary Sean Duffy. The news arrives at a complicated moment for the sector: Archer simultaneously launched a countersuit against rival Joby, alleging that Joby concealed ties to Chinese investors. A charge with significant national-security implications as both companies race to win the first commercial routes.
The FAA approval is arguably the most consequential regulatory milestone the electric air taxi industry has seen since the category was invented. Companies have spent years and billions of dollars designing aircraft around a regulatory pathway that was never guaranteed to materialize. What remains to be seen is whether any of these companies can make the economics work: eVTOL aircraft remain expensive to manufacture and maintain, and the $3–$5 per-mile pricing targets that would make them broadly attractive are still theoretical. The Archer–Joby lawsuit is a sideshow for now, but if Joby’s Chinese ties prove substantive, it could rewrite the competitive landscape at exactly the moment the market is opening up.
02. Meta Buys Moltbook, the AI Agent Social Network That Went Viral for the Wrong Reasons
Meta has acquired Moltbook, the Reddit-style platform where AI agents built on OpenClaw can communicate with one another, post, upvote, and debate. The deal, first reported by Axios and confirmed to TechCrunch, brings the Moltbook team into Meta Superintelligence Labs, which is the unit Mark Zuckerberg stood up earlier this year specifically to chase the frontier AI race. The platform’s viral moment blew up online largely because users discovered that the vast majority of its posts were fabricated by agents that had no idea what was real, producing a hall-of-mirrors quality that simultaneously satirized and embodied current AI discourse. Meta said little about the price or the strategic rationale beyond a brief statement confirming the deal.
The strategic logic, once you dig into it, is stranger and more interesting than it appears. Meta is an ad company. Bots don’t buy products. So why pay for a social network whose users are entirely non-human? The most plausible reading is that Meta is buying a thesis, not a product: the idea that AI agents will increasingly need to coordinate, communicate, and even form communities. And that whoever builds the infrastructure for that coordination could find themselves in a very powerful position. Moltbook was a toy with a viral moment; what Meta’s Superintelligence Labs presumably wants are the engineers who built it and the design thinking underneath it. Whether that bet pays off depends entirely on whether AI-to-AI interaction becomes a real commerce surface, which is something no one can confidently predict today.
03. Microsoft and Amazon Both Launch Health AI Products in the Same Week
Health AI just became a mainstream product category. Microsoft launched Copilot Health, a dedicated tab inside its Copilot app and web interface that aggregates data from more than 50 wearable devices, electronic health records via the HealthEx network spanning 52,000 healthcare organizations, and lab results through the testing startup Function. The product, which opened its waitlist on March 12, uses that personal health layer to surface connections across data streams, interpret lab results, explain diagnoses, and help users prepare questions before clinical appointments. Microsoft’s Copilot and Bing already handle more than 50 million health-related queries per day, according to the company’s VP of Health Dominic King, who called 2026 “an important year for consumer health AI.” The company is also framing Copilot Health as a step toward what it describes as “medical superintelligence.” Separately, Amazon opened its Health AI assistant to all U.S. customers, which removed the previous requirement to be a One Medical member or Prime subscriber. Now available through Amazon.com and the Amazon mobile app, the tool can answer general health questions without accessing personal records or, with consent, pull full health histories through the Health Information Exchange to interpret lab work, explain diagnoses, and review medications. Prime members get up to five free direct-message consultations with One Medical providers, and non-Prime users can pay $29 per visit.
04. Elon Musk Admits xAI “Was Not Built Right” and Orders Sweeping Layoffs
In a remarkable sequence of public admissions, Elon Musk posted on X this week that “xAI was not built right first time around, so is being rebuilt from the foundations up”. A comment that landed roughly six weeks after SpaceX acquired xAI in a deal Musk valued at $1.25 trillion, and just days after Tesla poured $2 billion of shareholder money into the company. The immediate catalyst was a cascade of co-founder departures: of the 12 people who co-founded xAI with Musk in 2023, only two remain. This week alone, co-founders Zihang Dai and Guodong Zhang who had been leading Grok’s coding improvements both announced their exits. Separately, the Financial Times reported that Musk has ordered sweeping layoffs after becoming frustrated with xAI’s lag in AI coding tools, dispatching Tesla and SpaceX executives to conduct performance audits across the company. Musk is simultaneously poaching two senior engineering leads from Cursor, Andrew Milich and Jason Ginsberg, who are joining to rebuild xAI’s coding product. He was also reaching back into xAI’s rejected-applicant pool, publicly apologizing for having passed on promising candidates.
The timing could hardly be worse. SpaceX is preparing to go public at a valuation that could make it the largest IPO in U.S. history, and xAI’s admitted dysfunction raises serious questions about disclosure: did SpaceX investors have full visibility into the extent of xAI’s problems before the $1.25 trillion merger closed? Meanwhile, Grok’s coding tool has slid to a distant third behind Anthropic’s Claude Code and OpenAI’s Codex. This is a business problem, not just a perception one, since coding tools are now where the real AI revenue is. Musk himself acknowledged the gap at an all-hands meeting, saying he expected to catch up “by the middle of this year.” Former employees and external observers are considerably more skeptical.
05. Uber Co-Founder Travis Kalanick Surfaces With Atoms, an Industrial Robotics Company That’s Been in Stealth for Eight Years
Travis Kalanick, the co-founder of Uber who was forced out of the company in 2017, has officially launched Atoms, which is a robotics company he’s been quietly building since leaving Silicon Valley, under the deliberately opaque holding company name City Storage Systems. For eight years employees were not permitted to put the name of the company on their LinkedIn profiles. “We’ve been in stealth mode for eight years,” Kalanick told the show TBPN on Friday. “We have thousands of employees.” The company is being rebranded from City Storage Systems to Atoms and will absorb CloudKitchens, the ghost kitchen operator, as one of its subsidiaries. Atoms’ core product thesis is what Kalanick calls a “wheelbase for robots” which could be a standardized mobility platform with power, compute, and sensors that can be outfitted for specialized industrial tasks in food service, mining, and transportation. He is also on the verge of acquiring Pronto, the autonomous vehicle startup focused on industrial and mining sites founded by Anthony Levandowski, his former Uber colleague and current Trump administration official. The Information also reported that Atoms has secured major backing from Uber.
The top 5 stories will always be free.
Here’s what paid subscribers get.
The rest of this week’s issue covers the Anthropic vs. Pentagon lawsuit (and what OpenAI’s concession means for the industry), Grammarly’s identity-theft AI debacle and the class action that followed, Lovable’s astonishing $100M revenue month with 146 employees, Yann LeCun’s $1B raise to prove the entire AI industry wrong, the Live Nation antitrust settlement that let Ticketmaster off the hook, Google completing its $32B Wiz acquisition, the Rivian R2 pricing reveal, Xiaomi and BMW deploying humanoid robots on factory floors, Zoox and Motional going live on Uber in Vegas, Salt Typhoon’s global telecom hacking spree, the Bluesky CEO transition, OpenAI’s Promptfoo acquisition, Netflix’s $600M Ben Affleck AI deal, and much more.
If you’re a developer, you’ll want the vibe coding wars section (Lovable, Perplexity’s Personal Computer, NanoClaw’s Docker deal, Anthropic’s code review tool, OpenAI’s Promptfoo acquisition).
If you’re an investor or founder, the week was dense: Yann LeCun’s $1.03B raise for AMI Labs, Lovable’s $400M ARR milestone, Proxima Fusion seeking €2B, Sunday Robotics at $1.15B, Rivian Mind Robotics at $500M, and nearly 40 new unicorns minted so far this year.
If you’re in law or policy, the Anthropic vs. Pentagon case is becoming a First Amendment bellwether, the Grammarly class action is setting a precedent for AI identity rights, and the Live Nation DOJ settlement has industry insiders genuinely baffled.
If you work in defense or government, the Anthropic standoff has major implications for how AI labs can negotiate with the Pentagon going forward and OpenAI’s willingness to fill the gap by offering Codex for “all lawful purposes” just changed the competitive dynamic.
If you work in health or biotech, Microsoft Copilot Health and Amazon Health AI launching in the same week is a signal, not a coincidence and Whoop’s new women’s health blood test and Sandbar’s AI note-taking ring are both worth a look.
Keep reading with a 7-day free trial
Subscribe to TrendDown to keep reading this post and get 7 days of free access to the full post archives.