State of the deep tech entrepreneurship in the UK and the emerging trends
Natalya Vilyavina22:362022
Women-only teams make up 28% of AI startups but get 1.5% of the funding
For every pound of UK venture capital, all-female founding teams receive less than a penny, even as deep tech, by her account, enters a golden age. Natalya Vilyavina, a venture manager and AI stream lead at the University of Oxford's Creative Destruction Lab, maps where deep tech money is flowing across the UK and Europe. So why do investors keep missing the women founders building it?
Learn why deep tech is riskier, where European capital is flowing and what holds women founders back
Natalya Vilyavina, a venture manager at Oxford's Creative Destruction Lab, explains what makes deep tech different from ordinary startups and why she believes it is entering a golden age, with the UK and Europe well placed to lead. She covers university spin-out equity terms, where European deep tech money is going, and the double bias that leaves women-led deep tech teams with a tiny share of venture funding.
“There is a lot of bias in the VC industry that women can do beauty, but they cannot maybe do deep tech.”
“And as we all know, all the money in VC, this is about pattern recognition.”
What you will learn
Distinguish deep tech, built on scientific discovery, from new software or apps built on existing infrastructure.
Explain why deep tech startups need patient capital and must turn a broad technology into one product before becoming a platform.
Assess how university spin-out equity stakes and later dilution affect a scientist founder's incentive to keep building.
Spot where European deep tech capital is going, including the rise of climate and energy investments she calls deep tech for good.
Recognize the two overlapping biases that keep venture money away from women-led deep tech startups.
Key takeaways from the speaker
We are approaching the "golden age" of deep-tech innovation.
AI is at the frontier of new technologies.
Trends in AI deep-tech innovation trends include application of GPT3 models, industrial AI and augmented workforce, adversarial AI, security and trust of AI systems
One of the challenges for deep tech entrepreneurs is lack of follow-on funding
Female founders of deep tech AI startups face challenges of access to funding and perception of their leadership role.
Details
Level
Beginner
Format
Talk
Technical depth
Non-technical
For
Founder
Length
22:36
Published
July 2022
Questions this talk answers
What is deep tech, and what is it not?
Deep tech is new technology based on scientific discovery whose commercial applications are still being developed. It is not an application of existing technology, such as a new app built for pre-existing digital infrastructure.
Why is deep tech riskier than a typical startup?
On top of market risk it carries R&D and commercialization risk, often starting from a technology in search of a use case, and it needs patient capital through long R&D phases. Early money therefore often comes from grants and government funding.
Why does Natalya Vilyavina call this a golden age of deep tech?
Foundational technologies such as AI, quantum computing and CRISPR make things better, faster or cheaper. That creates a basis for new technologies and for enabling technologies, such as security for the quantum age.
How do university spin-out policies affect deep tech founders?
UK universities such as Oxford traditionally took 50 to 60% of a spin-out, compared with 3 to 5% in the US, leaving founders heavily diluted by later rounds. She says Oxford's default is now about 20%, which still gives founders a good incentive.
Why do women-led deep tech startups get so little venture funding?
She describes two overlapping biases: that women are less suited to business and that they are less likely to do deep science. Because deep tech cycles are long and there are few successful female-founded spin-outs, VCs have not yet learned to recognize the pattern.
About this talk
The UK has been a leader in deep-tech innovation that emerges from its world-class universities and research centers. Since 2019 investment in deep tech in the UK increased by over 17%, which is the fastest rate among the developed countries. In 2021 University of Oxford Spin-Out Oxford Nanopore was valued at $4.6 Billion in London IPO. AI is at the frontier of deep-tech innovation, and the early-stage startups are paving the way. Trends in AI include the application of Open AI GPT3 models, industrial AI and augmented workforce, adversarial AI, security, and trust of AI systems. Female founders of deep tech AI startups face challenges in fundraising and proof of leadership.
How university equity stakes hold back spin-outs (70 s)
Less than a penny per pound for women founders (84 s)
Why VCs miss women deep tech founders (83 s)
Natalya Vilyavina: My topic today is the state of deep tech entrepreneurship in the UK and emerging trends. My name is Natalya Vilyavina. I'm a venture manager and AI stream lead at Creative Destruction Lab at the University of Oxford. Creative Destruction Lab, for those of you who don't know, is a deep tech program for startups. Natalya Vilyavina: It started at the University of Toronto in 2012 and since then it's grew to 11 sites globally. And our aim is to help founders commercialize their deep tech and increase the probabilities of success for deep tech founders.
Natalya Vilyavina: So first of all, let's think about like what is deep tech. Let's have some definitions. So deep tech is a technology Natalya Vilyavina: based on scientific discovery. This is new technology, which commercialization application are just being developed and what deep tech is not deep tech is not an application of existing technology. This is Natalya Vilyavina: Not a new piece of software or an app built for pre existing digital infrastructure.
Natalya Vilyavina: So how deep tech is different. Natalya Vilyavina: So first of all, deep tech is a higher, high risk. And in addition to Natalya Vilyavina: usual market risk, there is a risk of commercialization, R&D risks. And I just, so in the deep tech, what we usually have is the Natalya Vilyavina: commercialization usually comes in a normal startup.
Natalya Vilyavina: You would go with the what you will start with. Why? Why we're creating startup because there is a problem that we want to solve in a deep tech. Sometimes it comes in a different way. So sometimes we see that there is a research being done. Natalya Vilyavina: Interesting technology being developed, but the commercialization strategy is not really clear.
Natalya Vilyavina: And this is why it's usually it's often we can see that in a deep tech, you come with the Natalya Vilyavina: First, you have the already the technology and then you need to figure out what a commercial strategy for it. Natalya Vilyavina: This is why it's even more risky and the venture VC found VCs are often very Natalya Vilyavina: aware of it because they have to take higher risk. And sometimes we have founder that come to us and create a destruction lab and they can say, for instance, we are our technology is Natalya Vilyavina: broad technology that can apply to different areas. We developed it in the lab. It's like a calculator. It can do Natalya Vilyavina: Multiplication, division and subtraction and what we always tell them is that first your technology needs to be a product and then a platform. So you cannot start from the platform from abroad technology. You first need to find your unique use case and apply it to
Natalya Vilyavina: Other to different verticals. What is also very common for deep tech. It's always rely on IP Natalya Vilyavina: Usually it relies on IP and about 70% of the companies that we see have Natalya Vilyavina: Have developed their own IP and have secured this IP. Natalya Vilyavina: Another differentiation of deep tech is usually it comes in the intersection of technologies. So, for example, it can be AI and biology AI and material science and about 90 over 90% of the companies don't use like just one technology in deep tech. They use a combination of technology.
Natalya Vilyavina: And finally, the other differentiation of deep tech is that it requires patient capital. It means that the Natalya Vilyavina: The capital for deep tech is your development of the deep tech requires a lot of time. So the first time the first phase is R&D phase, which is can be many years, especially for like drug discovery when you can Natalya Vilyavina: Clear different regulatory hurdles. And that's why it's also very common for for deep tech startups to first have first money usually come from grants and government funding and these kinds of things.
Natalya Vilyavina: Okay, let's move on and talk about why deep white golden age of deep tech right now. Natalya Vilyavina: I believe it is because a lot of foundational technologies are being developed. And by foundational technologies, I mean technologies such as AI, quantum or CRISPR technology, which is create basis for other technologies being developed. So for example, if you Natalya Vilyavina: In CDL, for instance, we say that Natalya Vilyavina: Your technology needs to do something better, faster or cheaper. And technologies like artificial intelligence or quantum computing, they do make things better, faster or cheaper. For example, founder of creative destruction lab, Ajay Agrawal, he wrote a book that is called Natalya Vilyavina: Prediction Machine, where they argue that AI makes cost of prediction cheaper and by making cost of prediction cheaper, you can create other technologies based on
Natalya Vilyavina: Because the cost of prediction become cheaper, the other technologies can be developed based on this foundational technology, which is artificial intelligence and same things for quantum computing. Quantum computing makes the computing makes the computation faster and more efficient. And therefore, other technologies can be developed based on this foundational technology. So this is like one thing. So these foundational technologies in deep tech create opportunities for other technologies that use them as a basis. Also, foundational technologies create opportunities for enabling technologies.
So what it means is, for example, if we take quantum computing and in quantum age, every task can be done really, really quickly. What it means, it means that, for instance, the cyber security in quantum age is really vulnerable because all our passwords by default can be really quickly solved by quantum computers. So we need to rethink and startups have to rethink what is the security in quantum age look like and develop respective technology. So therefore, these foundational technologies, they both create the background for new technologies to be developed.
So the next part of my presentation is about Europe and the UK. So why deep tech in Europe and in the UK? So first of all, a quarter of European VC funds go to deep tech. So the quarter of the money go to deep tech. There was research that UK deep tech investment was 17% up in 2020. And this was the highest growth rate globally. And this is also due to a high rate of top educational institutions in the UK in Europe.
And let's look at the next slide and see what exactly in the top research and universities are happening in Europe. So first of all, high share of highly cited research publication, about 43% go to UK and Europe. So European students are more likely to do STEM degrees as compared to their US peers. So you can see that in Germany, for instance, 35% of students are actually do STEM. In the UK, this is 26%. And in the US, it's only 18%.
Also, you can see that in Europe, universities that are at the top of the rank of the computer science, it's out of 20 universities, seven universities are allocated in Europe. And out of this seven, four are actually in the UK. But if you see the size of Europe and UK, especially UK, you can see that UK actually lifts much above its own weight. So the size of the economy and size of the country is not that big as the US, for instance, but it has a lot of research and computer science opportunities.
So next, I'd like to share with you the fact that deep tech investment and spin-outs in the universities are growing exponentially. So this map is, we will not read this map, but I just want to share it with you so that you see, this is a map of Oxford University spin-outs in the university. Oxford University spin-outs from 1959 to 2021 of the Oxford University innovation, which is tech transfer office of the university, which is responsible for the spin-out startups. And you can see that over the course of these years, the number of the spin-out startups grow, grow really exponentially.
And there is a slight concern and trick here that UK and Europe in general, traditionally didn't have very favorable spin-out policy for the founders of the deep tech startups. This means that in the US, it means that in the US it is traditional that the university could take like three to 5% of the share of the startup when it spun-out of the university.
And in the UK, it's sometimes in Oxford, for instance, it used to be like, can be like 50% or 60%. So in this scenario, it's really difficult for the scientists who have developed some technology that can be commercialized to have an incentive to actually move this technology forward. Especially this is relevant for the technology that requires a lot of funding because the further you go down in the series A, series A, series B, you have to give up a lot of shares of your company.
And if you start with a relatively small number of shares, let's say 40%, then the next round you get with a really small percentage because of the dilution of your own shares. And it makes it really hard for the scientists and startup founders to be incentivized to continue building these technologies for a long period of time. Luckily, this policy is changing in the UK specifically and in Oxford specifically.
So I think now the default rule is that University of Oxford is taking about like 20% for deep tech spin-outs, which still gives a good incentive for founders to continue building. In the next slide, I'd like to talk about like European deep tech investment scene in general. This is an atomic report and we can see that in this is a report of like just nine months of 2021.
And it's you can see that capital investment in the startups in Europe in 2021 in the deep tech startups in Europe almost doubled in the just in the first nine months of 2021 from 9.4 billion dollars. So we see that the trend is going up so this means that deep tech technology will continue to develop and grow.
It's a lot of interest in the VC community to continue investment and in the deep tech. These two stars on the right represents the average around of the funding. It says that over 50, about 50% of the startups in deep tech received the funding of less than 5 million. And in the again, in 2021, it was only 12 deals that received more than 250 million dollars in one round, which obviously, which is maybe obvious because it's at the first at first you VC VC money go to like small, larger proportion, but the smaller amount of checks.
And further down the line, there are there were less successful startups and the follow on rounds are getting larger, but the number of those rounds are getting quite small. Okay, so next I'd like to give you a quick overview of the emerging trends of the deep tech in Europe that we observe. These data comes from pitch book and also like atomic report.
So at the first glance, if you see at this slide, it looks like by the number of deal count, the largest, by far the largest investment, largest investment money go into quantum computer computing, digital twins, AI powered for drug discovery, neurotechnology. However, if you look at this gray boxes at the bottom, you can see some areas like battery technology, carbon capture, hydrogen energy, waste to energy, long duration energy storage.
So to me, it looks like this comprises a big chunk of the investment money in Europe. And I would call this deep tech for good investments. So it looks like there is a trend being developed in not only deep tech, but also in investing deep tech for good. So I found it interesting and I wanted to share this with you as well.
So finally, I'd like to quickly stop at the deep tech gender gap. And here we all know that there are very few percentage of money in the VC go to female led startups and especially small amount go to only female led startups. So here is some stats for you only 15% of deep tech startups are founded or co-founded by women and only 6% have all women founding teams for every one pound in venture capital in the UK in general.
Female founder, female founded teams get less than one penny and all male founded teams get 89 pennies and mixed gender teams 10p. So again, you see like if at least like one co-founder, it's only 11% pretty much of venture money that goes to teams where at least like one female founder and to all female teams. It's only 1% it's general for the VC for the VC money at the same time, if you look at female startups in the industry, such cosmetics, shoes, home decor, fashion, flowers, wedding, beauty, 22% of the money go to female teams.
There is a lot of bias in the VC industry that women can do beauty, but they cannot maybe do deep tech. Female only led teams comprise 28% of all AI companies, but get only 1.5% of funding, which is 53% of a dollar as well. So this gives you an idea about where we stand with the investment and gender bias in the deep tech investment.
And it seems that there are two biases here. Like one is that in general, there are much less money go to female teams, which basically translates into the language that VC say we don't think that females are really good in doing business. But there is another bias as well, which says that it seems like female females are less likely to do deep tech science and therefore it's another bias that basically overlaps with the first one.
So it creates this really tricky environment for deep tech founders. I also spoke to teams at creative destruction lab, female lab teams, and they often tell me that when they lead the company as a chief executive officer, they often hear from the VCs that they need to hire a professional CEO. So basically giving them this idea that maybe they need to step down and maybe have a secondary role in startup, not a CEO role.
So this is also really hard for the female founders to have energy and the strengths to overcome these biases and perception. So I also think that in deep tech, why it's so little money goes to deep tech female founders, because I think that VC don't recognize this pattern. So there is not enough spin out companies founded by VCs that have funded by females that have become already successful.
And also the development cycle of deep tech is so long. So it's really hard for VC to see the pattern that actually I know the female founder and I know the startup that worked and I know this can be successful. And as we all know, all the money in VC, this is about pattern recognition. And as soon as we teach venture capital industry to recognize the pattern and to see that female founders can be successful, the more money go into the deep tech for female founders.
And then the second bias about the opportunity, opportunity for women to actually do deep tech. I want to leave you with one slide here, which is the successful female scientist in the past century. And you can see here the Marie Curie, who got the Nobel Prize for in 1903 for researching of radioactivity. You see the Sarah Gilbert, who was one of the researchers at the University of Oxford and the founder of Oxford AstraZeneca vaccine successful.
You see Emanuele Carpenter, if I pronounce it right, and Jennifer Doe, who also won the Nobel Prize in 2020 for CRISPR technology, gen editing technology. So you see all these female founders, female scientists that are actually there and female can do science and female can do deep tech. So I want to leave you with this and thank you very much for your attention.
Apologies for my not perfect use of technology. And I wish you a really pleasant day today. And I hope to see you at the Hopin platform. Thank you.
Transcribed automatically; names checked against the speaker record. Select a time to jump there.
Article based on this talk
Why deep tech is booming in the UK while women founders are left out
Natalya Vilyavina of Oxford's Creative Destruction Lab explains what makes deep tech different, where the money is going and why women-led teams still struggle to get funded.
In short
Natalya Vilyavina argues that deep tech is entering a golden age, driven by foundational technologies such as AI, quantum computing and CRISPR, and that the UK and Europe are unusually well placed to lead it. Yet the same ecosystem directs only a sliver of venture capital to women-led teams, because of two overlapping biases and a pattern that investors have not yet learned to recognize.
Key takeaways
We are approaching the "golden age" of deep tech innovation.
AI is at the frontier of new technologies.
Trends in AI deep tech innovation include the application of GPT-3 models, industrial AI and the augmented workforce, adversarial AI, and the security and trust of AI systems.
One of the challenges for deep tech entrepreneurs is a lack of follow-on funding.
Female founders of deep tech AI startups face challenges in access to funding and in how their leadership role is perceived.
What makes deep tech different
Vilyavina, a venture manager and AI stream lead at Creative Destruction Lab (CDL) at the University of Oxford, starts with a definition. Deep tech is technology based on scientific discovery, whose commercial applications are still being worked out. It is not a new app or piece of software built on existing digital infrastructure.
That difference brings extra risk. A typical startup begins with a problem to solve; a deep tech venture often begins with a technology from the lab and then has to find a commercial strategy. Founders sometimes arrive at CDL describing a broad technology that can do many things, and the advice is always the same: become a product first, then a platform, by finding one unique use case before expanding into other verticals. As the speaker cited from CDL's own portfolio, about 70% of the companies they see have developed and secured their own IP, and over 90% combine several technologies, such as AI with biology or material science.
Timing is the other difference. "And finally, the other differentiation of deep tech is that it requires patient capital." (4:19) R&D phases can last many years, especially in fields like drug discovery with regulatory hurdles, so first money often comes from grants and government funding.
Why this is a golden age
"I believe it is because a lot of foundational technologies are being developed." (5:03) At CDL, she explains, a technology has to make something better, faster or cheaper. She points to the book Prediction Machines by CDL founder Ajay Agrawal, which argues that AI lowers the cost of prediction, opening the door to new technologies built on top of it. Quantum computing does the same for computation.
Foundational technologies also create demand for enabling technologies. Her example: if quantum computers can quickly break today's passwords, startups need to rethink what security in the quantum age looks like. "So these foundational technologies in deep tech create opportunities for other technologies that use them as a basis." (6:10)
Why the UK punches above its weight
Vilyavina shares figures she presented on slides: a quarter of European VC funds go to deep tech, and research showed UK deep tech investment rose 17% in 2020, the highest growth rate globally. She adds that about 43% of highly cited research publications come from the UK and Europe, that 35% of students in Germany and 26% in the UK study STEM compared with 18% in the US, and that seven of the top 20 computer science universities are in Europe, four of them in the UK. "But if you see the size of Europe and UK, especially UK, you can see that UK actually lifts much above its own weight." (9:03)
She shows how Oxford University spin-outs have grown exponentially since 1959, but flags a catch. US universities traditionally take 3 to 5% of a spin-out, while Oxford used to take as much as 50 or 60%. With later rounds diluting founders further, scientists had little incentive to keep building. "Luckily, this policy is changing in the UK specifically and in Oxford specifically." (11:48) She says the default is now about 20%, which still leaves founders a good incentive.
Where the money is going
Citing an Atomico report, Vilyavina notes that capital invested in European deep tech startups almost doubled in the first nine months of 2021. "So we see that the trend is going up so this means that deep tech technology will continue to develop and grow." (12:58) But the shape of that money matters: about half of deep tech startups raised less than $5 million, and only 12 deals in 2021 exceeded $250 million in a single round. Follow-on rounds get larger, but there are far fewer of them.
Drawing on PitchBook and Atomico data, she sees most deals going into quantum computing, digital twins, AI for drug discovery and neurotechnology. Yet battery technology, carbon capture, hydrogen, waste to energy and long-duration energy storage form a big share of the money. "And I would call this deep tech for good investments." (15:31)
The deep tech gender gap
The final section is the sharpest. As the speaker cited, only 15% of deep tech startups are founded or co-founded by women and 6% have all-women founding teams. For every pound of UK venture capital, all-female teams get less than a penny, mixed teams 10p and all-male teams 89p. Female-only teams make up 28% of AI companies but receive only 1.5% of funding, while in categories such as cosmetics, fashion and beauty, women's teams get 22%. "There is a lot of bias in the VC industry that women can do beauty, but they cannot maybe do deep tech." (17:33)
She describes two biases stacked on top of each other: doubt that women are good at business, and doubt that they do deep science. Women CEOs at CDL tell her VCs often suggest they hire a professional CEO. Her explanation for why so little money reaches them is pattern recognition: few female-founded spin-outs have yet succeeded, and deep tech's long cycles make the pattern slow to appear. "And as we all know, all the money in VC, this is about pattern recognition." (20:11)
She closes by pointing to Marie Curie, Sarah Gilbert of the Oxford AstraZeneca vaccine, and CRISPR Nobel laureates Emmanuelle Charpentier and Jennifer Doudna. "So you see all these female founders, female scientists that are actually there and female can do science and female can do deep tech." (21:41)
Watch the talk
"State of the deep tech entrepreneurship in the UK and the emerging trends" by Natalya Vilyavina, 23 minutes: https://www.womentech.net/video/state-deep-tech-entrepreneurship-in-uk-and-emerging-trends-natalya-vilyavina
About the speaker
Natalya Vilyavina is an Executive Board Member of the Oxford Entrepreneurs Network and an AI Stream Lead and Venture Manager at the University of Oxford Creative Destruction Lab. After more than a decade in law she became a deep tech venture builder, and she holds an LLM from the University of San Diego and an MBA from the University of Oxford.
Questions this talk answers
What is deep tech, and what is it not?
Why is deep tech riskier than a typical startup?
Why does Natalya Vilyavina call this a golden age of deep tech?
How do university spin-out policies affect deep tech founders?
Why do women-led deep tech startups get so little venture funding?
Drafted with AI from the talk transcript. Editor review pending.
Frameworks and ideas mentioned
Better, faster or cheaper test (Creative Destruction Lab)
Product first, then platform
Resources mentioned
Prediction Machines (book by Ajay Agrawal)
Cite this talk
Natalya Vilyavina. "State of the deep tech entrepreneurship in the UK and the emerging trends." WomenTech Network, 2022. https://www.womentech.net/video/state-deep-tech-entrepreneurship-in-uk-and-emerging-trends-natalya-vilyavina