Editor's Note: This article is part of an ongoing series in which CrunchBase News interviews active AI investors. Read previous interviews with Felicis, Battery Ventures, Bain Capital Ventures, Menlo Ventures, Scale Venture Partners, Costanoa, Citi Ventures, Sierra Ventures, Andrew Ng of AI Fund, and more interviews in 2023.
Most corporate ventures avoid investing in rivals. However, GV (formerly Google Ventures) is not a majority of companies. Not only did Google back up Slack while having competing products, it is now investing in AI companies that compete directly with the efforts of parent company Alphabet's own in the artificial intelligence race.
The GV model – a single LP, independent decision – allowed us to stay fast and agile during the most explosive moments of AI. Managing partners Dave Munichiello and Tom Hulme are helping the company across chips, compilers and applications, making early and late bets. The GV bet is clear. The AI era rewards those who move fast, trust founders, and don't turn their eyes away from ratings.
Recently we spoke with Silicon Valley-based Municielo and London-based Hulme, a globally co-leading digital investment practices, about GV's AI investment approach.
Munichiello was previously at Kiva Systems, but now it's now Amazon Robotics. Hulme has a background in physics and was a design director for European IDEOs and an active angel investor before joining the then Google Ventures in 2014.
The Western fanatic AI gold rush
“It's definitely a wild time on the West Coast of San Francisco. Every weekend, it's five or six walks with founders who are flooded with AI terminology sheets,” Munichero said. “The founders are in a home full of other AI founders. The topic is tough for the world's meta and openly poached people.
Munichiello and Hulme both wrote, digested and summarised complicated ideas when they had middle and high school elderly people using technology in their daily lives.
There is a dramatic push from companies spending to win this “AI Magic” from widespread consumer recruitment.
The talent competition gets hot
“We believe the stack is as low as chips and infrastructure and as high as application-level, and all the different industries where application-level investments are occurring,” Munichero said.
Early GV investments in AI include lattice data purchased by Apple in 2017 from Power Siri, video generator Synthesia, and Data Labeling Company Snorkel AI. He was recently a record seed round investor for Machines Lab, a $2 billion seed round at a $12 billion valuation.
GV invests independently of Alphabet, its parents, and only investors. This means that they may support companies that are considered competitors of alphabet-owned products. For example, Thinking Machines Lab is competing with Alphabet's Gemini when GV invested in Slack and Google had a competitive internal product.
As for why they invested, Munichiello said he heard from their internal recruitment team that former Openai CTO Thinking Machines CEO and co-founder Mira Murati considers him to be one of the best recruiters of deep and technical AI talent. “Amazing companies are prices that sometimes make you feel very uncomfortable, and that was certainly the case with thought machines,” he added.
The competition among AI researchers is fierce, and the performance of models can be quantified. So, “For the first time, many of the big companies are competing in the same game. So, when you look at meta, alphabet, or Openai, they're all competing for having cutting-edge basic models,” says Hulme.
“We've never seen a company grow as quickly as we see today,” he said. “In the Foundation model, we don't see talent move as often as we see at present.”
The team pays attention to where the smartest researchers work. “They're not going to go to big tech. They're going to startups and they're all going to San Francisco,” Munichero said.
App layer
“We've been slow to invest in application tier companies until we know we have real traction,” Munichiello says. “If you look at Harvey or OpenEvidence, you didn't lead Series A. You led the later rounds.”
Munichiello said the question they are trying to address is, “Does these dollars seek use cases that are looking for interesting applications in their company or are they trapped in production?
To assess whether revenue is experimental, teams spend time understanding customer use. The product is beyond proof of concept, but is it used every day?
In 2024, the company led the Series C of AI Legal Tech Startup Harvey at a $1.5 billion valuation. The company has since raised two rounds and is now valuing $5 billion.
More recently, GV Co-LED, along with Kleiner Perkins, is a Series B in OpenEvidence, AI-powered platform that helps physicians and other healthcare professionals more easily aggregate and process peer-reviewed medical literature at a $3.5 billion rating.
GV is also jointly showing London-based online legal platform Lawhive's $40 million Series A investment with New York-based TQ Ventures.
“When you look at companies coming to procure, revenue execution rates are insane. These companies are growing faster than ever and faster than ever,” Munichiello said. “And it's very difficult to spend a lot of time looking at AI application companies and going back to looking at other companies.”
“In our portfolio we have businesses like Bolt.New. [a product of StackBlitz] – They built the technology, then turned on monetization, going from zero to $40 million in 12 weeks,” Hulme said, “I think that's the new normal.” GV was a co-leading startup seed round in 2022. The company launched Bolt.new in October 2024 and started the code using a text prompt in a web browser.
Infrastructure-level investment strategies
For the infrastructure layer, the company calculations are different, Munichiello said. “We need to start a deep technical job, there's a universal compiler, and it takes a lot of money to get there, and we're going to lead the first round.
That's why GV led the $30 million seed round in Modular. This is trying to build a unified calculation layer that interfaces with AI hardware. The company has created a Universal Compiler, a competitor for CUDA. This allows AMD's GPUs to run as fast as Nvidia's GPUs. Modular has raised a $100 million Series B, led by General Catalyst.
Multi-stage, multi-sector, multi-geo investors
GV has been investing for 15 years and manages $10 billion in assets. The company is headquartered in Silicon Valley, with offices in Cambridge, Massachusetts, New York and London. Over time, the biggest exits include Uber, Nest Labs, Slack, Gitlab and more.
CEO and managing partner David Krane leads a team of 21 partners. Alongside Krane, Municheello and Hulme, the fourth managing partner is Krishna Yeshwant, based in Cambridge, and is co-led with general partner David Schenkein to lead life science practices.
Companies meet at least one other partner through the investment process, usually pitching the entire partnership. Partners can share ideas and ideas with leads and then make investment decisions. The team was able to move quickly and close the deal within a week of their first meeting with the founder.
“A and B are sweet spots,” Munichiello said. “Our job is to look at A and B when we find an area we are excited about. And then we say, 'Is this the best way to invest in this category?” ”
If the right company is in Series D or Seed, the team will make the investment. Team members are not limited to Locale, they can invest in the best companies around the world.
Hulme thinks from a city perspective rather than a country. “For every great founder, we need five or ten great operators, and in my opinion, San Francisco, New York and London have their global pockets left. These are the cities we are most excited about.
Analog Business
“We're a venture capital business, but we think of it as a human capital business. We're as good or bad as the people on the GV team. And, interestingly, they're as good or bad as the people who're investing.
Munichiello said the company's diverse approach and flexibility is important.
“GV is a place where you can get 30 people in 30 different directions, and we're investing in 30 different things. And we're not trying to promote consensus,” Munichiello said.
Related Crunch Base List:
Illustration: Dom Guzman
Clarification: This story has been changed since the original publication to update the company's life science practices.
Keep your CrunchBase up to date with recent funding rounds, acquisitions and more.
