Report: AI startup funding hit ~$93.5B in 2021, more than double 2020's figure, but the number of newly funded startups fell from 1,051 in 2019 to 746 in 2021
That shape held. By mid-2025, AI startups were taking ~66% of all US VC funding, and by H1 2026 the share had reached 86% of a $412.7B US total. What looked like a 2021 anomaly was the start of a durable redistribution of venture capital toward fewer, larger AI bets.
First-order effects
Founders outside the AI winners' circle face a shrinking on-ramp: with 305 fewer startups funded than two years earlier despite doubled dollars, first checks are going disproportionately to established teams and repeat founders.
Second-order effects
Non-AI startups compete directly against AI deals for the same LP-backed pools, so every incremental dollar into mega-rounds raises the effective bar for seed and Series A companies in adjacent categories.
Third-order effects
If the concentration pattern holds — from 2021's doubling through 2025-26's supermajority shares — venture becomes a barbell market: a handful of capital-intensive AI platforms absorbing most dollars while the long tail of new company formation thins, reshaping which categories produce independent companies at all.
The trend: Venture capital is consolidating into ever-fewer, ever-larger AI companies, turning startup funding from a broad pipeline into a concentrated bet on scale.
Just released: The #AIIndex2022. This year's report provides new and updated metrics across all aspects of AI: research and development, technical AI ethics, AI policy and governance, diversity in AI, and more. Read the report: https://t.co/ESTJ49xHYv https://t.co/zJFrITpYW6
For years, people have said the AI Index should try and do more work on AI ethics. Well, now we have an entire chapter. Some of the highlights include explorations of trends in toxicity and bias in language models, analysis of workshop topics, and more! https://twitter.com/...
Our analytics and research partner @CSETGeorgetown provided valuable data on AI journal, conference, repository publications and patent data in the research and development chapter of this year's report. Read it here: https://hai.stanford.edu/.... #AIIndex2022 https://twitter.com…
Congrats to the @StanfordHAI team behind the #AIIndex2022! This massive report highlights advances in AI in education, technology, policy, and more. https://hai.stanford.edu/...
JUST RELEASED: THE 2022 AI INDEX REPORT Check out the latest AI progress in research and development, technical performance, technical AI ethics, the economy and education, AI policy and governance, and more. Read #AIIndex2022 here 👉 https://t.co/PRCfma9iL5 @StanfordHAI https://t…
the 2022 AI Index report is out! 🌻 i'm so excited that this edition features a new chapter on measuring fairness and bias in AI, co-authored by me and Ellie Sakhaee. as capabilities increase, it becomes critical to track and measure potential harms as well https://twitter.com/...
According to #AIIndex2022, 2021 saw the globalization and industrialization of artificial intelligence intensify, while the ethical and regulatory issues of these technologies multiplied. Read the main takeaways of the report: https://hai.stanford.edu/...