🚨: The Big Story

Larry Ellison borrowed $100 billion for AI. Now 30,000 employees are paying the tab.

Oracle is planning to cut 20,000 to 30,000 employees (up to 18% of its 162,000 workforce) as early as this month, per Bloomberg. The company is drowning in over $100 billion of debt from its AI data center buildout. It cancelled a flagship AI data center in Texas, froze cloud division hiring, and disclosed a restructuring plan expected to cost $1.6 billion in severance alone. Earnings drop today (Tuesday, March 11).

Why it matters:

Oracle isn't cutting because AI replaced those workers. It's cutting because it spent too aggressively building AI infrastructure for companies like OpenAI, Meta, and xAI, and the cash isn't flowing back yet. Wall Street projects negative free cash flow until 2030. US banks have pulled back from financing, doubling Oracle's borrowing costs. So tens of thousands of people might lose their jobs to fund servers that haven't turned a profit.

Our take:

Oracle posted $523 billion in contracted future revenue. Cloud infrastructure revenue grew 66% year over year. GPU revenue grew 177%. The business is exploding. But Larry Ellison bet the company on an AI buildout so massive that banks are walking away from the financing. The solution? Fire up to 30,000 people to free up $8 to $10 billion in cash. Record demand, record contracts, record layoffs. And Oracle's stock is down 54% from its September 2025 high, so employees with equity comp are getting hit twice: potential layoff plus cratered stock.

What to watch:

• Oracle reports fiscal Q3 earnings today. Expect layoff details and restructuring guidance on the call.

• If Oracle executes the full 30,000, it would be one of the largest single layoff events in tech history.

• Other companies burning billions on AI infrastructure could face the same squeeze. Microsoft already cut 9,000 amid its own AI spending surge.

Your move:

• If you work at Oracle (or any company burning cash on AI data centers while freezing hiring), update your resume now. Don't wait for the all-hands.

• Oracle's cloud competitors (AWS, Azure, GCP) still need people who understand enterprise cloud. Your skills transfer directly.

📡  Developing Stories

Anthropic sues the Pentagon over its blacklist. Nvidia writes tens of billions in checks to AI startups.

Anthropic vs. The Pentagon, Chapter 3:

The saga we've been tracking escalated again. Anthropic filed two federal lawsuits Monday against the Department of Defense and other agencies, calling the "supply chain risk" designation "unprecedented and unlawful." The company says it violates their First Amendment rights and exceeds the Pentagon's statutory authority. Quick recap: Anthropic refused to let the military use Claude for mass surveillance of Americans or fully autonomous weapons. The Pentagon wanted unrestricted "all lawful purpose" access. Negotiations collapsed. Defense Secretary Hegseth blacklisted them. Trump ordered all federal agencies to stop using Anthropic's tools.

What's new: Anthropic's CFO told the court the designation could cost billions in 2026 revenue. Enterprise customers are already pulling or shortening contracts. But here's the twist: dozens of scientists from OpenAI and Google DeepMind filed a personal amicus brief supporting Anthropic, arguing the blacklist harms US competitiveness and chills public AI safety discussions. Meanwhile, Claude is still being used in active military operations during the phaseout period. The company that the Pentagon calls a "supply chain risk" is simultaneously supporting the Pentagon's war effort. TechCrunch

Nvidia Is Writing Checks Like It's the Federal Reserve:

Two huge moves this week. First, Nvidia made a "significant investment" in Mira Murati's Thinking Machines Lab, the AI startup founded by the former OpenAI CTO. The deal includes access to at least 1 gigawatt of Nvidia's next-gen Vera Rubin chips (the Financial Times estimates the chip supply is worth tens of billions). Thinking Machines has raised $2 billion since launching 13 months ago and now has around 120 employees. CNBC

Second (announced today): Nvidia invested $2 billion in Nebius Group, an Amsterdam-based "neocloud" company, to build AI data centers with 5+ gigawatts of capacity by 2030. Nebius shares jumped 15%. This follows Nvidia's $2 billion into CoreWeave in January and $30 billion into OpenAI last month. Bloomberg

Why jobseekers should care: Every one of these investments means hiring. Thinking Machines grew from 30 to 120 employees in a year and is still scaling. Nebius, CoreWeave, and the neocloud operators are all building infrastructure that needs people to design, deploy, and manage. Critics note Nvidia is financing its own customers (circular investment), but the hiring at these companies is real and happening now.

📉 Layoff Report

Oracle, Morgan Stanley, Dow, and more. Record revenue keeps producing record cuts.

• Atlassian – Atlassian CEO Cites AI Shift When Announcing Plan to Shed 1,600 Jobs. Bloomberg

• Oracle: Planning 20,000 to 30,000 cuts across multiple divisions to fund AI data center debt. Restructuring expected to cost $1.6 billion in severance. Cloud division hiring frozen. Bloomberg

• Morgan Stanley: Cut 2,500 (3% of workforce) across all three divisions after posting record $70.6B revenue in 2025. CEO Ted Pick got a 32% raise. One commenter on an industry forum: "Gotta fund Ted Pick's $40M payout somehow." Reuters

• Palo Alto Networks: Laid off 500 CyberArk employees after acquiring the identity security company. Cuts targeted overlapping roles from the merger; R&D was spared. InformationWeek

• Dow Chemical: Cutting 4,500 (13% of workforce), citing AI, automation, and "simplifying operations." Net sales were down 9% YoY in Q4. InformationWeek

• Gloat: The workforce intelligence platform (irony noted) cut 20% of staff on March 7. When the company selling AI workforce planning tools lays off its own people, that tells you something about where we are. Intellizence

Bottom line: The pattern this week is unmistakable: record revenue, then layoffs. Morgan Stanley posted its best year ever, then cut 2,500. Oracle's cloud grew 66%, then it froze hiring. Tech employee confidence hit a record low of 47.8% this month per Glassdoor. People aren't imagining this.

🧑‍✈️ Career CoPilot

How to prove "AI experience" when hiring managers don't know what it means yet.

Every job posting now seems to want "AI experience." But most hiring managers don't actually know what that means yet. That's your opening. You don't need a machine learning PhD. You need to show you can work alongside AI tools to get better results, faster.

• Add an "AI Tools" line to your skills section. List what you actually use: ChatGPT, Claude, Copilot, Cursor, whatever. Be specific. "Prompt engineering for code review" beats "familiar with AI."

• Quantify what AI helped you do. "Reduced report generation time by 60% using Claude for first-draft analysis" is a resume bullet that lands. Frame AI as a force multiplier for your existing skills.

• In interviews, volunteer your AI workflow. When asked about your process, mention where AI fits in. PwC data shows workers with AI skills earn up to 56% more than peers in the same roles. That premium is real.

• Build one public artifact. Write a LinkedIn post about how you used AI to solve a real work problem. Or push a small project to GitHub that shows AI-assisted development. Public evidence of AI fluency > a bullet point claiming it.

• Don't overstate it. Only 43% of US workers regularly use AI at work (Indeed Hiring Lab). Honest, specific experience beats inflated claims. Hiring managers can smell fake fluency.

📊 Trends & Data

47.8% confidence. 56% AI wage premium. The market is splitting in two.

• Tech employee confidence hit a record low of 47.8% in March per Glassdoor, down 7.1 points year over year. Tech is now the fastest-declining sector for worker morale in the US. Glassdoor via LatestLY

• 20% of 2026 tech layoffs (roughly 9,238 jobs) are directly tied to AI and restructuring, per RationalFX. The rest? Mostly pandemic-era bloat correction dressed up in AI language. TNGlobal

• 53% of US tech job postings now require AI/ML skills, up from under 40% a year ago. Data scientist roles are projected to grow 414%; cybersecurity analysts at 367%. IEEE-USA / Robert Half

• Analysts describe a new pattern: "forever layoffs." Instead of big one-time cuts, companies run a continuous trickle of smaller reductions designed to fly under the radar. Glassdoor's lead economist calls it "more weeks of winter for workers." Glassdoor

What this means for you: Morale is low, but the skills market is splitting fast. AI-adjacent roles are growing while everything else flattens. The billions Nvidia is deploying into infrastructure companies this week alone confirms the money is real. Position yourself where the spending is.

👀 Companies to Watch

Cloudflare, CoreWeave, Nexthop AI, GitLab.

• Cloudflare: While most companies cut entry-level hiring, Cloudflare is going the other direction: 1,111 paid internships in 2026 (up from ~60 in 2024). CEO Matthew Prince: "Many companies see AI as a way to replace jobs; we see this next generation as a way to bring the best uses of AI into Cloudflare." Also hiring across infrastructure, security, and developer platforms. Careers

• CoreWeave: AI cloud infrastructure company valued at $52 billion after a $2 billion Nvidia investment in January. Building specialized AI data centers through 2030. Named to TIME100 most influential companies. Growing fast across engineering and operations. Careers

• Nexthop AI: Raised $500 million at a $4.2 billion valuation to build high-performance AI infrastructure for large cloud operators. Founded 2024, based in Santa Clara. Small team scaling up. Careers

• GitLab: One of the world's largest fully remote public companies. 29% year-over-year revenue growth in FY2026. DevSecOps platform with native AI features keeps expanding. Remote-first with 1,500+ team members across 60+ countries. Careers

🧰 Jobseeker Tools

Track your apps, practice your interviews, know your runway.

• Teal: Job search CRM that tracks applications, contacts, and follow-ups in one place. Stops you from losing track of where you applied last Tuesday. Free tier available. Try it

• Interviewing.io: Anonymous mock interviews with engineers from top companies. Practice without the stakes, get blunt feedback. Paid, but worth it if you're rusty or targeting FAANG-tier roles. Try it

• Offboard - Financial Runway Calculator: Maps your savings against monthly expenses so you know exactly how long you can search before things get tight. Helps you decide which offers to take (or wait on). Free tier: 30 uses/mo. Try it