Academics Founded 27% of China's Newly Funded AI Startups

TL;DR: A count by Investment China Jiachuan, reported by 36Kr on October 10, 2026, found that 49 of 182 newly funded AI-chain companies, about 27%, were founded by university faculty, researchers or faculty-student teams. For US angels, it's a signal about where early AI talent sits, not a deal to chase.

What happened?

The count comes from Investment China Jiachuan, and 36Kr reported it on October 10, 2026. It looked at AI industry chain companies established after January 1, 2026 that have received investment, and found 182 of them. Of those, 33 were clearly founded by official faculty members or researchers from universities and research institutes. Another 16 came out of university or institute technology transfer, or from faculty-student teams. Together that is 49 companies, about 27%.

Group	Companies
Newly funded AI-chain companies counted	182
Founded by faculty or researchers	33
Spun out of university work or faculty-student teams	16
Academic-linked total	49 (about 27%)

I'd read the 27% with one caveat in hand. According to the same report, the 182 companies include general AI large models, AI applications and embodied models, but also AI chips and brain-computer interfaces. It's a wide net, not a count of frontier model labs.

What's the example behind the headline?

The article opens with a large-model startup founded only two months ago that has taken investment from Legend Capital, CAS Star and SenseTime. It then profiles Hypersphere Technology, which, per the same report, was founded on August 5 this year, with the financing disclosed at the end of September. The company says on its recruitment page that it has investment from top-tier US dollar funds and a valuation of over 100 million, and the report doesn't name the currency of that valuation. That valuation is the company's own claim, and I haven't seen a filing or an investor statement confirming it. Verify before you trust.

Why should a US angel care?

If you're writing $25K to $250K checks into early AI companies, the useful part isn't China. It's the pattern. A quarter or more of the newest funded companies in one market start inside a university or a research lab, and they get funded within weeks of incorporation. That leaves you very little time to do your diligence.

The US picture looks different, and the differences matter for how you screen. CNBC reported that Antler found the age of AI unicorn founders fell from a peak of 40 in 2021 to 29 in 2024. IFP's analysis of the 2025 Forbes AI 50 found that 25 of the 42 US-based companies were founded or cofounded by immigrants. Each market produces a different kind of founder, and none of these facts works as a filter on its own. What they do tell you is to screen on what the team has done, not on who they are.

A professor-founded company usually brings deep technical credibility and little go-to-market experience. I spent years as an innovation scout at Hartford Steam Boiler vetting startups for a Fortune 500 parent, and the gap between a working prototype and a company that sells was the gap I saw most. If you've read my checklist on AI startup due diligence, you know where I'd push first: who sells this, who pays for it, and what the team has shipped to a real customer. Downside first.

What can't this count tell you?

Not much about returns. It counts companies that received investment, not companies that survived, and 2026 is a partial year, so the number will move. It says nothing about valuations beyond the one self-reported figure above, and it doesn't tell you how many of these companies will raise a second round. Bigger early checks don't guarantee later ones. I've written about that gap in seed rounds getting bigger while Series A gets harder. The count is a snapshot of who got funded, nothing more.

What mistakes do investors make with a number like this?

The most common one is treating 27% as a founder-quality score. It's a share of a funded sample, and that sample mixes chips, applications and models, so it can't rank anyone. Close behind is assuming the same share holds in the US, when nothing in this report measures US companies.

The costly one is chasing the headline. A professor on the cap table is a starting point for questions, not an answer to them, and if you wire money before you know who on the team has sold to a paying customer, you've skipped the one question that matters. If you're weighing the enterprise side, my piece on AI startups selling demos instead of workflow change covers what to ask.

What am I watching next?

I'm watching whether Jiachuan publishes the full list, because the split between model companies and chip companies would change how I read the 27%. I'm also watching whether any of these companies reach a priced follow-on round. That would show whether early academic-founder funding turns into durable companies.

FAQ

Are 27% of all new AI companies founded by professors?

No. The figure covers 182 newly established AI-chain companies that received investment, according to 36Kr's report of the Jiachuan count. It doesn't cover unfunded companies or companies outside that sample.

Does this include US companies?

Nothing in the report says it does. It's a count by Investment China Jiachuan, so I'd treat it as evidence about one market and not as a US statistic.

Does a professor founder make an AI startup a better investment?

The report doesn't test that. It counts who founded the companies, not how they performed. Technical depth is useful, but you still need to diligence the business.

If you want this kind of read on private-market news each week, with the downside first, sign up for the free AIN briefing.

Educational content only. Not investment, tax, or legal advice. Not an offer or solicitation to buy or sell securities. Past performance does not guarantee future results. Private-market investments are illiquid and involve risk of loss, including total loss of capital. Consult qualified advisers. Angel Investors Network is not a broker-dealer or investment adviser.