1Mby1M Virtual Accelerator AI Investor Forum: With Naganand Doraswamy, Managing Partner and Founder at Ideaspring Capital (Part 1)

Naganand Doraswamy, Managing Partner and Founder at Ideaspring Capital, discusses how he is thinking about investing in AI. There’s a lot of confusion around AI right now. The hype is astronomical. Amidst this noisy environment, we’re trying to bring you some signal, some wisdom.

Sramana Mitra: Let’s start by level setting a little bit. Since we talked last, where are you with the fund? Of course, the industry has been evolving quite significantly. Give us a little bit of an introduction of where you are with Ideaspring and then we’ll dive into the modern age of what’s happening in the age of AI.

Naganand Doraswamy: Thanks a lot for having me on this show again, Sramana. Congratulations on your 646^th roundtable. I’m looking forward to the day where it will be thousand and it will be awesome to hit that milestone as well.

In terms of Ideaspring Capital, we are an early stage fund based in Bangalore, investing on innovation-based startups. I use the term innovation because it’s not necessarily software, but we also do life sciences and hardware to a certain extent. We focus on India-based entrepreneurs who are building products from India for the Indian market and for the world market as well; because as soon as you say you’re building a software product, particularly for the enterprises, then the market has to be global.

We have had some very interesting investments since we last spoke. We have invested in a quantum computing company, which was started by a professor from RISE. He moved back to India and started this startup. We have also invested in a startup building or manufacturing proteins as a sugar substitute.

Another company we invested in recently is a startup based in California. It provides highly contextual, personalized learning graphs for anybody who is trying to learn anything, such as corporate skill development.

In India, the key change that has come in is there are more and more funds and more and more people. Everybody talks about deep tech, although I would say, deep tech is still 20% -25% of the total startups that even our own fund invests in. I think the probability is going up for good products coming out of India.

Also now talk on semiconductors and AI, so we’ll talk about that.

Sramana Mitra: Since I started working on the Indian ecosystem back in 2005, the big shift that I see is that there’s no ambiguity about Indian products in the global market anymore, which is a major accomplishment.

Naganand Doraswamy: I agree with you a hundred percent on that.

Sramana MItra: Where we are today, in the popular consciousness of the entrepreneurial ecosystem, everything currently is dominated by AI. Today, I wanted to invite you back here to discuss what is your investment thesis in the age of AI? How are you looking at the market? What are the trends, where are the investment opportunities, what to avoid? AI is happening in many different ways. What can you actually invest in and make money in?

Naganand Doraswamy: I think as an investor, I should frankly accept that I’m living in one of the most confusing times. When you look at AI, what can you invest in? How much can you invest in? At what stage do you want to invest in, whether you want to go horizontal at the foundation layer or you want to go into tools or you want to go into verticals, contextualizing or SaaS companies adding copilots. These are confusing times because opportunities exist, but things are moving so fast, it becomes very hard to say what will still be relevant three years down the road.

As an investor, whenever you want to invest, you want to be able to fast forward three years and see what will be relevant at that time; because what you’re building today will take three years to get some kind of a product to market. However, things are moving at such a furious pace that it’s hard to predict what’s going to happen next quarter, let alone what’s going to happen in the next three years.

So that’s why I say it’s very confusing times. We recently conducted research for our fund and based on that, the number of seed stage startups in AI in the last six quarters that have emanated in India is about 550.

Now, many of those are run up the mill or solving some problem that I am afraid one of the big guys will just sneeze and then it gets subsumed by that. I think the biggest challenge that we are facing in our own decision making process is how to steer very clear of what the big guys can do. We’re putting tens of billions of dollars investing and trying to do something and stay away from that, meaning don’t be in the foundation model, don’t be in the genetic stuff that they will absolutely subsume.

And also now with ChatGPT-4-o multimodal stuff, where you are saying, ChatGPT is textual, I will look at voice, video, and other aspects and make it multimodal, that’s also gone. That also gets subsumed.

You either play in the consumer space – providing intelligence to consumers based on their data. But we are not a consumer focused fund. That brings us to what can you do with enterprises. You can either give them tools for them to choose the LLMs they want or how best to gather all the data and how best to train the LLMs in an enterprise specific use case for them to basically make sense of deploying AI for converting tasks into workflow. That’s an area we are looking at.

In some cases, we are also saying, “Are there specific things you can do with enterprise specific data to build generic platforms that enterprises can use to derive benefits out of the data that they have?” Open AI is generic, but it doesn’t have enterprise specific data. Then the question is, how do you get the enterprise specific data and train it so that it’s meaningful?

So we are looking at these specific areas and not venturing out into either core tools or even the foundation layer. Our area of focus is quite narrow, and that’s where we are looking to invest in startups. We have invested in one in the AI space.

Most of our SaaS companies are also now expanding into using generic capabilities. That’s a trend. Five or six of our startups are now bringing on AI copilots and then trying to make best sense of the emerging AI sector. So this is kind of where we are in our thought process and our journey in looking at AI.

This segment is part 1 in the series : 1Mby1M Virtual Accelerator AI Investor Forum: With Naganand Doraswamy, Managing Partner and Founder at Ideaspring Capital
1 2 3 4