AI funding in Q1 2023 is absolutely bonkers. Let me show you the numbers.
$12.4 billion in AI startup funding in Q1 2023 alone. That's more than all of 2020. I broke it down by category, stage, and geography. Generative AI is 73% of the total.
I keep a spreadsheet of AI funding rounds. I've been doing this since 2021. It used to take me an hour a month. In Q1 2023, it took me a full weekend to get through all the deals.
$12.4 billion. In one quarter.
Let me put that in context.
The year-over-year trend
| Year/Quarter | Total AI funding | Notable context | |-------------|-----------------|-----------------| | Full year 2019 | $26.6B | Pre-GPT era | | Full year 2020 | $9.9B | COVID, AI winter fears | | Full year 2021 | $28.7B | GPT-3 hype building | | Full year 2022 | $21.3B | DALL-E 2, ChatGPT launch | | Q1 2023 alone | $12.4B | GPT-4, generative AI mania |
Sources: Crunchbase, PitchBook, CB Insights, and the Stanford HAI AI Index.
Q1 2023 is 59% of all of 2022. If this pace continues (it probably won't), annual 2023 AI funding would hit $49.6 billion. That seems absurd, but so does $12.4 billion in a quarter.
Where the money went
I categorized every round over $10M by primary focus:
| Category | # of rounds | Total raised | % of total | |----------|-------------|-------------|------------| | Generative AI (LLMs, image, etc.) | 47 | $9.1B | 73.4% | | ML infrastructure & tooling | 18 | $1.4B | 11.3% | | Vertical AI (healthcare, legal, etc.) | 23 | $1.2B | 9.7% | | AI safety & alignment | 4 | $0.5B | 4.0% | | Other/general AI | 8 | $0.2B | 1.6% |
Sources: Deal data from Crunchbase and PitchBook, my categorization.
73% of the money is in generative AI. That's a staggering concentration. In Q1 2022, generative AI was about 25% of AI funding. The shift happened in roughly 9 months, driven almost entirely by ChatGPT's launch in November 2022.
The mega-rounds
The top 10 rounds account for most of the total:
| Company | Amount | Stage | Category | |---------|--------|-------|----------| | OpenAI (Microsoft investment) | $10.0B* | Growth | LLMs | | Anthropic | $450M | Series C | LLMs / AI safety | | Cohere | $270M | Series C | LLMs | | Adept AI | $350M | Series B | AI agents | | Inflection AI | $225M | Series A | Conversational AI | | Runway | $141M | Series C | Video generation | | Jasper | $125M | Series A | AI writing | | Typeface | $100M | Series B | Enterprise content | | Character.ai | $150M | Series A | Chatbots | | Mistral AI | $113M | Seed | LLMs |
*The Microsoft/OpenAI deal structure is unusual (investment, not traditional VC round). I include it for completeness but it skews the total significantly.
Sources: Company announcements, Crunchbase, press reporting.
Without the Microsoft/OpenAI deal, Q1 2023 AI funding is $2.4 billion. Still a lot, but a very different number. One deal is 80% of the quarter's total.
Mistral AI raising $113M at seed stage is worth flagging. A seed round that large for a company with no product yet is unusual even by AI standards. It signals investor belief that the LLM race is still open and that a European challenger can compete with OpenAI and Anthropic.
Geography
| Region | # of rounds | Total raised | % | |--------|-------------|-------------|---| | United States | 62 | $10.8B | 87.1% | | Europe | 11 | $0.8B | 6.5% | | Israel | 5 | $0.3B | 2.4% | | Canada | 6 | $0.3B | 2.4% | | Asia | 4 | $0.2B | 1.6% |
The US dominates. 87% of the funding. Europe's share is growing (Mistral, Stability AI) but it's still a fraction. The Asian numbers are surprisingly low for a quarter, likely because Chinese AI funding flows through different channels that don't show up cleanly in Crunchbase data.
Stage distribution
| Stage | # of rounds | Avg round size | Total | |-------|-------------|---------------|-------| | Seed | 24 | $18.2M | $437M | | Series A | 19 | $78.4M | $1,490M | | Series B | 12 | $115.3M | $1,384M | | Series C+ | 8 | $273.8M | $2,190M | | Growth/Corporate | 3 | $3,533M | $10,600M* |
*Dominated by the Microsoft/OpenAI deal.
The average seed round in AI is now $18.2 million. That's absurd. For context, the typical seed round in SaaS is $3-5 million. AI companies are raising at 4x the normal rate, at earlier stages, with less product validation.
What I'm watching
Three patterns concern me:
1. Concentration risk. The top 3 deals (OpenAI, Anthropic, Adept) are 88% of total funding. If those three companies stumble, the "AI funding boom" story changes overnight.
2. Category crowding. I counted 47 generative AI companies raising rounds in Q1 alone. Most are building products on top of the same foundation models (GPT-4, Claude, open source). When 47 companies are competing to build "AI writing tools" or "AI chatbots," most of them will fail.
3. The seed round inflation. $18M average seeds mean investors are pricing companies on potential, not traction. That works in a bull market. In a correction, these valuations will look silly.
My prediction for Q2 2023: the funding pace will slow slightly (the Microsoft/OpenAI deal was a one-time event), but the number of individual rounds will increase. More companies raising smaller amounts. The total might be $4-6 billion, still historically high but more sustainable.
The numbers are bonkers. That's the technical term.
If you found this interesting, you might also like:
- I counted every AI startup that raised money in Q1 2021. The numbers are strange.
- Anthropic just raised $580M. Let's talk about the AI safety funding numbers.
- The GPT-3 API waitlist is 6 months long. Here's what the early data looks like.
- My 2021 AI data roundup: the 10 numbers that mattered most
- The cost of running an AI startup in 2022: a data breakdown
-- dataku