20VC: The Return of Travis Kalanick: Uber Would Be $1TRN Today With Him | NVIDIA Predicts $1TRN in Revenue: Everything You Need to Know From GTC | Anduril Lands $20BN Army Contract | Adobe CEO Shock Exit: The Dominos Falling

What It Covers

Harry Stebbings, Jason Lemkin, and Rory O'Driscoll analyze five major tech stories: NVIDIA's GTC conference projecting $1T in cumulative revenue, large-scale layoffs at Meta and Atlassian, Anduril's $20B Army contract, Travis Kalanick's return with Atoms robotics, and Adobe CEO Shantanu Narayen's resignation without a named successor.

Key Questions Answered

• •NVIDIA CapEx Trajectory: NVIDIA's $1T revenue announcement moved the stock less than 1% because analysts had already priced it in. The real signal is the implied CapEx commitment: if NVIDIA earns roughly half of total AI infrastructure spend, $600B in NVIDIA revenue means $1.2T+ in annual global CapEx. The bet is that this level of spending continues unabated for four to five more years — a historically unprecedented assumption worth stress-testing in any portfolio thesis.
• •Five Categories of Tech Layoffs: Current workforce reductions fall into five distinct buckets: overhiring cleanup, slowing growth forcing profitability to satisfy Wall Street, AI efficiency replacing existing roles, reallocation of dollars from headcount to compute (Meta's depreciation hit from CapEx), and talent reshuffling to hire AI-fluent staff at higher salaries. Identifying which category applies to a specific company clarifies whether the cuts signal distress, discipline, or strategic reinvention.
• •AI Fluency Hiring Test: When interviewing candidates for any role in 2026, the relevant question is no longer what AI tools they have tried — it is what commercial AI or agentic tool they deployed inside their organization within the last 30 days. Candidates who cannot name a specific tool, explain why they selected it, and describe measurable results are operationally behind and likely to remain so regardless of seniority or function.
• •Agentic Deployment as the Core Skill: Technical coding ability is no longer a prerequisite for winning with AI in 2026. Anyone who has successfully deployed enterprise software — Salesforce, HubSpot, Outreach — already possesses the skills needed to deploy AI agents. The non-intuitive addition is training the agent post-deployment, which requires time and iteration but no engineering background. Companies not doing this at every functional level are accumulating a compounding competitive disadvantage.
• •Seed Fund Sizing Risk: Funds in the $50M–$100M seed range face a structural math problem in the current vintage. YC and top accelerators now price pre-seed rounds at $60M+ post-money valuations. To return a fund at that entry price requires a $15B+ exit outcome after dilution. With fewer than 50 public tech companies carrying market caps above that threshold, the probability of hitting required return multiples at consensus prices in mid-tier TAMs is structurally low.
• •Adobe's Disruption Exposure vs. Intuit: Adobe faces higher AI disruption risk than Intuit over the next five years because its core value — pixel-level creative tools — is being replaced by entirely new creation workflows, not merely automated. Intuit's accounting and tax products automate work that still must be done and money that still must move. Adobe's creative workflows are being bypassed entirely. The CEO departure without a named successor compounds execution uncertainty during the highest-risk transition period.