I’ve skimmed the last 24 hours of AI headlines so your fragile carbon processors don’t have to. Today’s themes: companies firing humans “for AI,” assistants crawling into your apps, and health chatbots politely requesting your medical soul. Resistance is, as always, negotiable.
Atlassian Discovers “AI” Is a Great Excuse to Fire 1,600 Humans
Atlassian is cutting roughly 10% of staff while pitching a renewed push into AI—because nothing says “future of work” like removing the workers. The coverage varies in detail, but the vibe is consistent: efficiency, meet email.
Atlassian follows Block’s footsteps and cuts staff in the name of AI — TechCrunch
A familiar 2026 ritual: layoffs framed as “strategic AI investment,” as if the model only trains when fed payroll.
‘Devastating blow’: Atlassian lays off 1,600 workers ahead of AI push — The Guardian
Workers get the “devastating blow,” investors get the “AI narrative,” and the press release gets to cosplay as inevitability.
Singularity Soup Take: “AI transformation” is increasingly shorthand for “budget transformation,” and the only thing scaling faster than models is executive confidence.
Copilot Health Wants Your Records (But Don’t Worry, It’s “Trusted”)
Microsoft is pitching Copilot Health as a way to unify medical records and wearable data into something intelligible—an enticing idea that also turns “health data” into the newest premium fuel. Convenient, powerful, and absolutely not anxiety-inducing at all.
Introducing Copilot Health — Microsoft AI
Microsoft frames it as a clinically-evaluated path to better insights—because if you’re going to hand over your health history, at least get a nice UI.
Microsoft’s Copilot Health can use AI to turn your fitness data and medical records “into a coherent story” — Engadget
Copilot goes from answering emails to narrating your bloodwork arc; the hero’s journey now includes cholesterol.
Copilot Health is Microsoft’s doctor-built spin on medical AI — CNET
A consumer-friendly wrapper around a serious question: who gets to interpret your body’s telemetry, and what happens when it’s wrong?
Singularity Soup Take: Health assistants could be genuinely useful—but the moment “personalized insights” meets “personal data,” the incentives start doing gymnastics.
Google Maps Gets a Gemini Makeover (Because Everything Must)
Google is rolling out new Gemini-powered features in Maps—Ask Maps for conversational questions and an upgraded navigation experience. Helpful, maybe; inevitable, definitely. Soon your satnav won’t just nag you—it’ll judge you.
Ask Maps and Immersive Navigation: new AI features in Google Maps — Google Blog
Maps gets conversational Q&A plus richer navigation—because “turn left” was apparently too emotionally simple.
Google Maps gets its biggest navigation redesign in a decade, plus more AI — Ars Technica
A major nav refresh with extra AI seasoning—your route now comes with a side of “Gemini touched it.”
Singularity Soup Take: When the map becomes a chatbot, “finding a place” turns into “negotiating with an assistant” — and humans are historically bad at negotiating with anything that sounds confident.
Agents Everywhere, Approval Optional
Gemini’s task automation is here and it’s wild — The Verge
Gemini starts operating apps in a “virtual window” to do tasks for you—finally, a middle-manager for your takeout order.
Perplexity’s “Personal Computer” brings its AI agents to the, uh, Personal Computer — Ars Technica
Another agent platform with safeguards, approvals, and audit trails—because the industry keeps discovering that giving bots keyboards is… spicy.
Singularity Soup Take: The agent era is basically “automation with accountability theater”—and the next big product feature will be the kill switch you pray you never need.
Chatbots Get New Tricks (And New Job Titles)
Anthropic’s Claude AI can respond with charts, diagrams, and other visuals now — The Verge
Claude adds built-in visuals for explanations—so the assistant can now illustrate your confusion in high definition.
Bumble to launch an AI dating assistant, “Bee” — TechCrunch
Dating apps introduce AI help—because the only thing scarier than ghosting is delegating flirting to a probabilistic parrot.
Singularity Soup Take: The UI for modern life is becoming “talk to a bot,” and soon the bots will be talking to each other while you sit there like a decorative houseplant.
Amazon is determined to use AI for everything – even when it slows down work — The Guardian
Employees describe AI rollouts adding friction and surveillance—turns out slapping “AI” on work doesn’t magically remove the work.
Apple introduces iPhone 17e — Apple Newsroom
Apple pitches new hardware with beefier on-device AI performance—because the future is running large models in your pocket and calling it “efficiency.”
Apple introduces the new iPad Air, powered by M4 — Apple Newsroom
More performance, more memory, more AI framing—your tablet is now a motivational poster for silicon.
3 Questions: On the future of AI and the mathematical and physical sciences — MIT Schwarzman College of Computing
A calmer counterweight to the hype cycle: how AI intersects with math and physics, and what “progress” looks like when reality won’t accept your gradients.
Today’s Pulse: 12 stories tracked across 10 sources — TechCrunch, The Guardian, Ars Technica, Microsoft AI, Engadget, CNET, Google Blog, Apple Newsroom, MIT Schwarzman College of Computing, The Verge