1Mby1M Virtual Accelerator AI Investor Forum: With Benjamin Narasin, Founder and General Partner at Tenacity Venture Capital (Part 5)

Sramana Mitra: That is absolutely correct in the examples that you’re giving. There is one thing I want to point out for the audience. Remember when software was being sold by large companies to large enterprises. Back then, this whole category of software companies came in and priced very low SaaS offerings and started selling to SME.

That was a very different business, and their customers are really the no-name customers. That was the strategy that they followed successfully. So that is a valid strategy.

Benjamin Narasin: So there’s another place where you could see a reason someone could have a competitive advantage. This comes from the founder of Hippocratic, Munjal Shah, with whom I spent so much time.

His point was, there was a time, very similar to what you just said, where if you could sell me something 80% cheaper, I would be willing to tolerate 20% failure rate. That’s not viable in certain industries using LLMs. Examples would be legal or medical.

In legal, if you go in front of a judge and use something that turns out to be a hallucination, you’re not just going to lose the case. You could get disbarred. You do not have the ability to just be wrong 20% of the time.

In medical, when his LLMs are doing nursing calls and they hear that somebody has shortness of breath, they immediately divert that to a human being. You cannot take the risk of putting somebody’s life at stake because you did not understand that something was a risk factor.

So there will be an opportunity in industries and places where there is no ability to be wrong for highly reliable, bulletproof LLMs that can probably charge a lot more and will ultimately be the player because people will trust them. They don’t care about the fact that it costs more because the cost of failure is so extreme.

Sramana Mitra: There is still a go-to-market strategy and a positioning that is a price-based positioning going to the low end of the market.

Benjamin Narasin: Yes, it depends on where you are, right? There’s an old saying in the apparel business: Price, quality, speed; choose any two. If you’re in an industry where quality is paramount, you can’t afford to mess with price.

So I’ll give you an example. I was stunned that the Bloomberg box was never displaced by the internet. I met Mr. Bloomberg at the very beginning of the web. Somebody in the room asked him, “What do you think about the internet?” He said, “It’s going to be the biggest thing you’ve ever seen in your life, and I’ll believe that when you show me a CB radio in your car.”

I thought, “Wow, this is an incredibly brilliant man. How can he have this huge blind spot? His business is more at risk than anybody’s business on the planet. His business just delivers data to you in a box on your desk for a stunning amount of money.”

The primary reason for that he never got displaced is probably because the Bloomberg box is also your email. You are not giving up your email. You have decades of legacy communication and experience, historical records, financial transactions, and everything else on your Bloomberg box. By the way, for somebody that’s making millions of dollars a year personally, which is probably the case for many of the people using a Bloomberg box, you’re not going to take the risk just to save 10 grand or 150 grand or whatever it costs.

There are places where you will pay what you need to because the downside is too material for you for an error or to go with a cheaper product. There are other places where that’s not the case.

Then there’s a third instance, which I find quite interesting and slightly disturbing. One of the first companies in Geneva I almost funded, but didn’t, was a Y Combinator company that pitched that they would use LLMs to take orders at drive-throughs in fast food restaurants. I thought that’s great. How hard is it for an LLM to say, “Do you want fries with that?” The turnover of those team members is incredibly high – typically teenagers, not particularly skilled. I thought this could be really interesting for several reasons. One, it’s such a nichey, weird idea. I doubt anybody’s gonna attack them anytime soon. Although somebody just pitched the same thing at YC literally two weeks ago, but it’s been two years.

The second reason was I thought on this first-to-brand market, if they could land McDonald’s, then they’d be good to go because no one’s gonna buy. If you’re the one that’s running this for McDonald’s, then I’m gonna trust that you do it right. If you’re the one that’s running this for Big Boy, I’m probably gonna be a little less convinced.

At the end, McDonald’s announced that they were gonna build their own team to build their own LLM, to do exactly this. I didn’t invest, by the way, the price was too high.

Sramana Mitra: There’s a lot of that going on in Fortune 500 right now. People are building their own stuff.

Benjamin Narasin: So stupid, it’s insane. What a waste of time.

This segment is part 5 in the series : 1Mby1M Virtual Accelerator AI Investor Forum: With Benjamin Narasin, Founder and General Partner at Tenacity Venture Capital
1 2 3 4 5 6 7