Tribe AI’s CEO on why generative AI is seeing more rapid uptake by enterprises than Web3 and crypto

After leaving Capital V, Google’s later-stage venture capital arm, in 2018, former vice president of growth Jaclyn Rice Nelson was struck by the tremendous number of talented engineering colleagues who had also left Google and other large tech giants where they’d spent the early parts of their careers, seeking to spread their wings and do so with more freedom.

Rice Nelson was inspired by them to found a new firm, Tribe AI, based out of a historic and iconic brownstone house Brooklyn, New York. Tribe offers a “fractional network” of freelance software engineering talent and experts, particularly in machine learning and AI, who can be hired by its clients on demand to work on discrete projects and AI transitions for them. As Tribe puts it on its website, it offers “300+ machine learning engineers, strategists, and data scientists from leading technical institutions. We help companies unlock the full potential of AI, driving success and innovation like never before.” 

Tribe AI launched in 2019 and has seen steady success since then, working with fellow startups and steadily larger clients, but has never been busier than the last six months, following the release of OpenAI’s ChatGPT and the continuing rush by companies of all sizes and various industries to embrace generative AI.

>>Follow VentureBeat’s ongoing generative AI coverage<<

Rice Nelson, who serves as Tribe AI’s CEO, recently made time to speak with VentureBeat over Zoom to discuss more about her approach to growing her company, her take on the generative AI craze, and why gen AI is succeeding and drawing more investment and public attention than the last two big waves of venture interest — the metaverse, and Web3/NFTs/cryptocurrencies. 

VentureBeat: Tell me about your background.

Jaclyn Rice Nelson: I spent most of my career at Google. And actually, that’s where I kind of fell in love with startups and the camaraderie and really like the energy, the creation. I spent three years on the late-stage venture side at Capital G, which is Alphabet’s growth fund.

We invested in these incredible tech companies — Airbnb, Ease, the new leaders of tech. And the value proposition was that we could leverage all of Google’s people, expertise, resources, playbooks to help scale and accelerate the growth and scale of the companies we were investing in. That’s what Google is best at: how to scale things. That was really where our companies were at the point of investing. They already had a successful business, they were focused on how to go global, how to really become these huge public companies that have massive exits for investors.

So the idea was that we built a true expert network of this sort of “fractional workforce” of engineers and other personnel who could help our companies scale. We were able to offer these companies the ability to access any part of this amazing talent network and base at Google, and the things they really wanted the most help on were specialized engineering and product development focus.

And so that meant what I was seeing as an investor, you’re really focused on patterns. The pattern that really emerged for me was just the demand for data science, machine learning, AI support, and the nuances of the questions and the efforts within those companies. I got to see what best-in-class talent in this area of data and AI really looked like.

For me, it felt so clear that this is where the market needed to move and was going to move in the future — that all companies are going to need to become AI companies. If even these true tech companies were struggling to make that transition — it’s not that they couldn’t, but that it wasn’t easy for them and needed specialists and more engineering talent — it just felt like there had to be a better way.

So I thought, what if there was a way to help other companies, even those outside of Capital G’s investments, actually transition and build this technology that can be so powerful into their business in ways that it actually did add value to them?

The way I set out to solve that problem is similar to what we did at Capital G, which was network-based. What I found when I actually left Google was that I was not alone because there were a lot of people who had similarly stepped out of these amazing companies where they had built the cutting-edge AI machine learning technology. And they wanted different things in their career, but they still wanted to monetize their skill sets.

So I saw an opportunity to build this fractional workforce that really optimized for getting them interesting, diverse kinds of opportunities across companies and enabling them to learn, have a community to work with when they had stepped out of a place like Google, and also still make a lot of money because ultimately they had these incredibly valuable skill sets. Not monetizing them was such a missed opportunity. And so Tribe created the infrastructure for both that sort of tactical, best-in-class talent and this platform for AI solutions and product delivery at scale.

VB: We are at a really interesting point right now with new startups emerging and this ongoing wave of investment in AI. It seems really far more profound than the investment that we saw in Web3.0 and crypto and metaverse-type startups. There are even accusations of “AI washing” companies, just kind of trying to get this money that’s flying around without having much real AI integration or use cases...

Rice Nelson: It’s true, they are not even accusations! Even public companies are adding AI like how they were adding crypto before and it was increasing their stock price. There’s just a moment of frenzy, I think is what you’re describing.

I think what feels different to me, and I was very interested in this sort of crypto and Web3 space as well, still am. But what feels fundamentally different is the stages these sorts of industries are at, which is to say, Web3 is still quite nascent, crypto is very nascent. There are real use cases, right? These are sort of things that are still evolving, really interesting ideas, but they’re still just ideas.

With AI, these technologies have actually existed for a really long time. Everyone’s now going nuts for generative AI, but the first transformer paper was written in 2017. Many of the engineers in the Tribe network have been doing generative AI since around 2017. And so this is not new.

What’s new is the user interface that has really captured consumer attention, and that consumer attention has really driven business adoption at an unprecedented rate. The thing you’re talking about is investment into the space, which is accelerating because of these other things.

Before, it was like, “Oh, this is an idea, this could be a platform shift, let’s put money into Web3 and crypto.” Here, we actually don’t just have signals that it could be. We have signals that it is happening, and happening at an accelerated rate. Because it’s in the consumer world, which is inherently so much faster to move than the enterprise.

And so, I think the pace of adoption of AI into business now feels really different. It feels like the pace of acceleration in the use cases that are now becoming possible. It’s what I describe as the shift between toy and tool, right? And so, as these things become tools, businesses have to actually adapt them to their business.

But it’s happening fast, and they don’t know how. And they’re asking the same questions we were getting at Capital G four years ago. And they’re asking them now and feeling like, “Why is it so hard to access talent? Why are these projects so hard to get right? Why does it always take so long? Why is it so expensive? Why are the data issues so pervasive and tricky?” And so, I think that’s what you’re seeing is [that] this sort of consumer adoption has become the catalyst to businesses actually feeling the need and urgency, and it’s going to change the face of every industry.

VB: That makes sense. If a company is going to adopt AI, there are a couple of different paths they can go down: They can build their own internal AI team, or they can work with external AI partners like Tribe AI, for example. What do you see as the pros and cons of each of these approaches? Like, what should a company be thinking about when they’re making that decision?

Rice Nelson: It’s a great question. So I think you’re right. You could build it or you could buy it, right? Or outsource it, I guess, in this case. And I think the decision depends on what you want to be when you “grow up,” or mature into the next phase of your company.

This was actually really, really clear when I was at Capital G. We were investing in companies that are valued at billions of dollars, right? They were growing. They had an incredible product-market fit, incredible execution, leadership, go-to-market. They had a real business. They had a real team. They had lots of things, but they didn’t have certain expertise in-house. And that’s why we invested.

But it was never meant to be a long-term relationship, right? It was really a short-term relationship, and the objective was always to build that expertise in-house because it is the most strategic and valuable thing that they could own in their business. And so, we did this repeatedly, and it actually got to be quite a challenge to find the expertise we were looking for. This is, again, for companies that were investing in a massive product market fit and were well-funded, right? But they still couldn’t find these talents, and so they would sort of create these outsourced agreements to build this expertise.

But what would happen inevitably is the project would go on for six, 12 months, and then we would hire the best people from that firm, bring them in-house, build that function, and then that team would become a sales lead for us and we could go and replicate that. And this happened time and time again.

And so, what that told me was, for the highest-leverage companies, the ones that actually are going to build it, it’s a strategic decision. You can start to build it out, and if you really want to own it, you should own it. It will be a competitive advantage. For everyone else, you should just outsource it. And the reason is, these are, again, incredibly hard projects, and it’s very hard to do them without real specialization in-house.

There are definitely instances where a startup can go and find that incredible person, put them in-house, make it work, get lucky, and have a great outcome. I think it’s pretty rare. And I think, for most companies, the most efficient way to do it is to leverage external expertise.

That doesn’t mean outsource the whole thing. It’s still a partnership, and it still has to be done with the company. But I think the sort of critical roles and the critical elements of the project really should be done by this sort of fractional team of experts that are on the cutting edge, that are there day in and day out, and really, really know how to do it, and know how to do it efficiently, and can see the nuances that are going to save you a ton of time and a ton of money.

In general, it’s just so hard to find these people that you need to do it in that way, and you need to do it with a team because it’s so multidisciplinary. You need product, you need engineering, you need data, you need domain, you need AI expertise, and you need these people who know how to build this infrastructure in-house.

I think it really just depends on what sort of company you are, what your aspirations are, and I think, at a high level, it’s just that most companies should be focused on their core competency, which is not AI, and should leverage external expertise to build it.

VB: Yeah, that makes a lot of sense. And it seems like there’s a lot of value in having that specialized expertise and bringing that in. And I’m curious, from your experience working with companies, what are some of the common challenges that companies face when they’re trying to implement AI solutions? Are there any recurring themes or difficulties that you’ve seen?

Rice Nelson: Absolutely. The thing that I always say is that data is really the foundation of everything. It’s not the first thing you do — it’s the first three or four things you do, and it’s the last three or four things you do. Do you have the right data? Do you have the right data infrastructure? Do you have the right labeling? Do you have the right tooling to actually collect the data? It’s never perfect. It’s never the same. It’s always a mess.

The second thing is it’s a very complicated space. Maybe you know a lot about natural language processing (NLP), but NLP can mean so many things. It can mean question-answering, it can mean chatbots, it can mean summarization, it can mean translation, it can mean understanding customer intent. Each one of those tasks has a unique set of tools, models and techniques, and so it’s very hard to know it all. You really need a multidisciplinary team.

The last thing is understanding just how long these [AI transformation] projects take. It’s very hard for a company to really internalize that, and understand the time and the resources that are required. It’s an extremely heavy lift. It’s really hard to get right and to get it to a place where it’s actually adding value. It’s a very long investment cycle, and I think that’s really hard for a company, especially when you’re starting from scratch, and especially when you have other things going on.

There’s a lot of fear about job displacement — that if we do this, then it’s going to displace a bunch of jobs, and it’s going to change the way we do things, and I think that’s a very valid concern. [But] what we’ve found is, actually, it’s not about displacement, it’s about augmentation.

The companies that we work with are able to do so much more, and they’re able to actually shift their workforces to much higher value-add activities. But having the right team and having the right partner is so critical.

VB: Building on that, what advice would you give to companies that are just starting out on their AI journey? What are some key considerations or steps that they should keep in mind?

Rice Nelson: First thing: Really think about your objectives, about what you’re trying to achieve, what is the problem that you’re trying to solve, what is the opportunity that you’re trying to capture? With AI, there are just so many different things that you could do. It’s really easy to get overwhelmed or, on the flip side, to say “Oh, this is really cool. Let’s do this. Let’s do that,” without a coherent strategy or set of uses cases in place, and start taking on too many new projects and builds. It’s really important to have focus and clarity — to understand where the value is going to be created for your business and your customers.

The second thing is: Just get started. It’s also really easy to overthink it and get analysis paralysis. People think that you need all your data, all the right tools, all the experts, and it’s just not true. You need to start. Choose a really specific use case or problem. What you’ll find is that you’ll learn a lot, and hopefully begin to generate value, momentum and excitement. That will create its own virtuous cycle.

The third thing is, find the right partner. It’s really, really hard to do this alone. You need a team of experts, people who have done this before, who understand the nuances and what works and what doesn’t.

Those are the three things: Really think about your objectives, just get started, and find the right partner.

VB: That’s great advice. Looking ahead: where do you see the future of AI heading? What are some of the exciting developments or trends that you’re keeping an eye on?

Rice Nelson: There are a few things that I’m really excited about. The first is continued democratization. The tools, the infrastructure, the accessibility — it’s all getting so much better so rapidly. The ability for anyone to build an AI system is going to be real, and I think that’s incredibly exciting and powerful, and will lead to so much innovation.

The second one is continued specialization. AI is not a monolith, it’s not one thing. We’re seeing people start to specialize and focus and go deep on a specific use case or a specific industry. That’s where you’re going to see the most value created, the biggest impact and the most innovation.

The third trend I’m excited about is the integration of AI into our daily lives. We’re already seeing it with voice assistants and recommendation systems, but it’s just going to become so much more prevalent, seamless, and valuable.

VB: It’s been really great chatting with you and hearing your insights and experiences. Is there anything else you’d like to add or any final thoughts you’d like to share?

Rice Nelson: No, I think we covered a lot of ground. We’re just scratching the surface of what’s possible with AI. There’s so much more to come. It’s going to continue to evolve, surprise us and challenge us. But it’s going to continue to create so much value. I’m really excited to be a part of it to see what the future holds.

VB: Absolutely. Well, thank you so much, Jaclyn, for taking the time to chat with me today. It’s been a pleasure talking to you and learning from your expertise. Thank you.

Rice Nelson: Thank you. It was my pleasure.

>>Don’t miss our special issue: Building the foundation for customer data quality.<<