The Signal — March 13, 2026
Atlassian cuts 1,600 jobs citing AI. The Supreme Court says AI cannot be a copyright author. Julia Angwin sues Grammarly.
THE SIGNAL
Future Shock Daily — March 13, 2026
The "AI ate my workforce" playbook just got another high-profile case study, the Supreme Court put a period on one of AI's longest-running legal sagas, and a Pulitzer Prize-winning journalist is suing Grammarly over what it does with your writing.
Atlassian Cuts Roughly 10% of Its Workforce, Blames AI Pivot
Atlassian is laying off about 10% of its workforce, roughly 1,600 people, as CEO Mike Cannon-Brookes reorients the company around AI. He posted a four-minute video explaining the decision, framing it as a structural shift rather than a cost-cutting exercise.
The move follows Block's similar AI-justified layoffs earlier this year and adds to a growing pattern: companies using the AI transition as cover for headcount reductions that might have happened anyway. To be clear, some of these cuts probably are genuine restructuring. AI tools really are changing which roles companies need. But the "we're pivoting to AI" explanation has become so convenient that it's getting harder to tell restructuring from rebranding.
Atlassian had about 13,800 employees at the end of fiscal 2025. Reuters, TechCrunch, and Business Insider all covered the layoffs independently, and Cannon-Brookes' video statement serves as the primary source.
Sources: TechCrunch · Reuters · Business Insider
Supreme Court Refuses to Hear AI Art Copyright Case
The U.S. Supreme Court declined to hear Stephen Thaler's appeal in his yearslong fight to get copyright protection for art generated by his AI system, DABUS. The cert denial ends the case for good.
Thaler argued that AI-generated works should be eligible for copyright even without a human author. The Copyright Office, federal courts, and now the Supreme Court have all said no. The legal principle is clear: copyright requires human authorship. Full stop.
This doesn't resolve every question about AI and copyright — there's still plenty of unsettled territory around AI-assisted works where a human contributes creative direction. But on the pure "can an AI be an author?" question, the answer from every level of the U.S. legal system is no.
Julia Angwin Files Class Action Against Grammarly
Julia Angwin, the investigative journalist behind ProPublica's Machine Bias series and co-founder of The Markup, filed a class action lawsuit against Grammarly. The suit alleges the company mishandled user data, though the specific claims haven't been fully detailed in early reporting.
Angwin isn't a random plaintiff. She's spent her career investigating how tech companies collect and misuse personal data, won a Pulitzer Prize for it, and literally wrote the book on digital surveillance (Dragnet Nation). When someone with her track record files suit, it tends to be well-researched.
Grammarly processes a staggering amount of user text: emails, documents, messages, draft after draft. The question of what happens to all that data, especially as AI features multiply across the product, is one a lot of people have quietly wondered about. Now there's a lawsuit asking it out loud.
Sources: TechCrunch
On the Editor's Desk
Three of the council's top five picks from yesterday didn't appear in today's editor review at all — the Iran-linked Stryker cyberattack, the Telus 1-petabyte breach, and the Washington Post's analysis of AI money flowing into 2026 midterms. These stories may have aged out of the ingestion window or lacked sufficient independent sourcing for our pipeline to pick them up. We're watching all three.
NVIDIA's reported $26 billion open-weight AI model investment got held. The Nemotron 3 Super model launch is confirmed and real, but the $26B SEC filing figure traces back to a single blog post and hasn't been independently verified. The Decoder covered it, but the dollar amount needs qualification before we'd run it. When (if) NVIDIA confirms the investment figure through official channels, this becomes a lead story.
The Anthropic-Pentagon story continues to develop — The Verge and Zvi Mowshowitz both covered new angles this week. But we've already given it significant coverage on March 10 and March 11, so today it rests. The story isn't going anywhere; we'll pick it back up when there's a genuinely new development rather than running commentary.
The pipeline's significance scorer also had a rough day. An Arduino edge AI development board got scored a 5 out of 5 — the same tier as GPT-4's launch. It's a fine product announcement. It is not redefining the field. Calibration continues.
