00:00 Have I seen something that scares me?
00:02 Not really.
00:04 Because I see it and I think about the code that's being written.
00:07 You're on the cutting edge of AI, nothing scary?
00:09 The technology isn't scary to me at all.
00:11 But the societal impacts and how we react, that could be scary.
00:14 So it's very important that the government get involved and help with retraining and education.
00:19 And that companies are responsible and help with retraining and education.
00:27 Hi everyone, I'm Kenret Kai. I'm a staff writer at Forbes and I'm joined by Konstantin Buehler, a partner at Sequoia Capital.
00:33 Sequoia, Forbes and Meritech Capital, we work together on the AI50 list, which we're here to talk about.
00:41 It's the sixth year of the list that identifies the most promising AI startups in the world.
00:48 So Konstantin, tell me a little bit about what you do at Sequoia and how that relates to
00:55 your role with AI50.
00:57 You got it, Kenret. Well, it's been fun working on this list with you for several years now.
01:01 Absolutely.
01:02 And my background is actually as an AI engineer.
01:05 So I got into this as a student.
01:07 I did my undergraduate work in stochastic modeling and my graduate work in natural language processing AI at Stanford.
01:14 And I was lucky enough to be there at a time
01:17 over a decade ago when the transformation
01:20 from classical AI to deep learning AI was happening.
01:24 What that means is we went from more basic algorithms to algorithms that were based on the neural network that is so powerful today.
01:32 Right.
01:33 That completely changed the trajectory of my career and I've been totally focused on
01:38 the impact of technology and in particular the impacts AI can have.
01:43 So tell us about this year's list and your observations from it. What really stood out to you?
01:48 And yeah, what are your takeaways from the companies that made it onto our list this year?
01:53 It was hard getting to the final 50 because there were so many great candidates.
01:58 What was special about the companies that got to the final 50 is this year,
02:01 I'm pretty sure almost all of them and you'd know because you've seen the final numbers, had great revenue and fundamental financial growth.
02:08 Right.
02:09 Like that's pretty new and exciting for the AI industry that businesses are really demonstrating
02:15 great revenue growth and AI companies are also really nice businesses.
02:21 Part of why that's happening is because we see a lot of this year's list in the productivity space.
02:26 So we have new companies on the list this year like Notion in the productivity space and they were historically a productivity company
02:35 along the lines of the best possible
02:38 note-taking and online knowledge repository.
02:42 Sequoia is an investor.
02:43 We are and you just wrote an article about it.
02:45 That's right.
02:45 So we want to hear about that.
02:46 I'm about to ask you about that in a second.
02:48 But Notion is an example of that.
02:51 And now they've got this great QA, which is able to answer questions way faster.
02:58 So, Kenric, I just mentioned Notion.
03:00 Why did you pick Notion and what did you learn about it?
03:04 Yeah, absolutely.
03:05 So for our, the AI50 goes in the magazine and along with the main list, we also want to feature a real standout company.
03:15 From the list.
03:16 So in the past, we've done companies like Databricks and like Hugging Face.
03:21 And so last year we did Scale AI and Notion, I think, had really stood out to us because, as you mentioned, they are a newcomer to the list.
03:30 However, they've been around since 2013 and they've grown into a very formidable company.
03:36 As you all know, last round was at a $10 billion valuation and they're making hundreds of millions of dollars in revenue.
03:45 So, you know, they are already a very popular company that has really captured consumer attention.
03:55 You know, their viral TikTok videos that caused the servers to crash in 2021.
04:01 And YouTube is just flooded with Notion tutorials, like from college students and young professionals who are obsessed with like creating to-do lists and
04:12 cool ways to organize their work and their life on Notion.
04:16 So it's a company that just really has already established itself.
04:19 But then why it's so interesting to AI50, I think, is kind of what you touched on already.
04:24 That Notion, I think, more than a lot of the other incumbent companies that are already established tech companies that have been around for a couple of years and have real revenue already.
04:37 Notion, I think, more so than most companies, really jumped on the AI wave really fast and integrated the AI in really unique ways.
04:46 I'm curious, with some of the areas that were buzzier a couple of years ago, but we don't see so much of this year's list, for example, autonomous vehicles.
04:56 We have a company called Wabi on this year's list, autonomous trucking.
04:59 But otherwise, we don't have very much of that placement on the list.
05:05 What about these other spaces that are not attracting so much attention in the current AI moment?
05:10 What does the future hold for the companies in those spaces?
05:14 Yeah. One of the most important things for your audience to understand is that AI is a constant evolution.
05:22 And we are at a point in time when audiences watching this video in five years, AI is going to look completely different.
05:29 And that's almost by the definition of AI.
05:32 So the list in 2019 looks radically different from the list today.
05:37 Big time, yeah.
05:38 And a lot of that, Kenric, is because the technologies have changed.
05:42 Let's think about 2019.
05:44 We had a lot of autonomy.
05:46 We had a lot of computer vision.
05:47 I think this was the year where we had companies like Hinge Health and Verkata in the computer vision space and then a lot of the autonomous startups for driving.
05:58 Right.
05:59 And you think about hype cycles.
06:01 That was a peak moment.
06:03 That was a moment where computer vision and autonomy was at a high end of the hype cycle.
06:07 And you think about the Gartner hype cycle chart.
06:09 Those are probably at the latter end of the hype cycle, as in they're not necessarily front of mind.
06:16 But I'm very confident they're going to be super important trends.
06:21 Computer vision, this idea that a computer can understand what it sees, was near magical when I was in grad school and we were working on this kind of stuff.
06:34 It was a breakthrough idea.
06:35 And we're seeing computer vision all of a sudden have passed its hype, but have a really good effect in industrial settings, in retail settings, in commercial and even in houses.
06:51 Another thing that I think is somewhat overlooked at the moment, but will be critical, is autonomy.
06:57 I mean, I have been in a Waymo vehicle in San Francisco and it is fabulous.
07:04 And they've gotten better and better.
07:06 The first time I was in one of these vehicles, it was choppy.
07:09 Now it's smooth and we're operating in and out of streets and with stop signs in residential areas.
07:13 It's exceptional.
07:14 Konstantin, I'm curious, you know, you spend so much time looking at the AI space.
07:19 Yeah.
07:20 How do you use AI in your personal life?
07:22 And I'm also curious to throw a curveball.
07:25 Are there things that you're seeing with the AI that are starting to scare you at all?
07:28 So, yeah, AI has been part of my life for 15 years.
07:32 And so I use it.
07:34 I try to use it a lot.
07:36 So everything from mundane tasks, like I now, anytime I can work alone, have my AirPod Max's on and I talk to my computer to write emails.
07:48 And I read them with text to speech.
07:51 I can move really quickly that way.
07:53 I also use it to write things.
07:57 So for memos, I now use AI in order to help speed things up.
08:04 When I get up the curve on a new company, I also use generative AI to help me understand a market.
08:11 And one thing that I think is pretty unique is I've gotten to the point where I use a lot of AIs in competition and in unison.
08:18 OK, interesting.
08:20 So I use a product called Dust, which is in the Square portfolio, in order to call not just GPT-4, but also Mistral and Gemini and Cloud 3.
08:30 And call means what?
08:32 And I just, I basically say instead of GPT-4, I say @GPT-4, just like you would in Slack.
08:38 @GPT-4, @Cloud3, @MistralLarge, @Gemini.
08:42 Tell me a little bit about Forbes.
08:47 I see.
08:47 And they will each come up with an answer.
08:50 Their answers will compete.
08:52 I'll see them in real time.
08:53 And then I can say @GPT-4, build on Claude's answer, because I like their first answer.
08:59 And they basically collaborate and compete with each other.
09:03 That's been a critical change in my workflow.
09:07 So those are some ways I use it day to day.
09:09 Have I seen something that scares me?
09:10 Not really.
09:12 Because I see it and I think about the code that's being written.
09:16 You're on the cutting edge of AI, nothing scary?
09:18 It doesn't scare me.
09:19 I mean, I see it and I think about the code that's being written.
09:22 And I say, OK, I know where this is coming from, roughly.
09:28 I know how they got there.
09:30 I think the things that will be scary will be societal impacts.
09:36 That's what matters.
09:37 The technology isn't scary to me at all.
09:38 Right, right.
09:39 But the societal impacts and how we react, that could be scary.
09:42 I mean, there's an unfortunate reality is when you have cost savings that will probably affect jobs, maybe a lot of jobs.
09:50 So it's very important that the government get involved and help with retraining and education, and that companies are responsible and help with retraining and education.
09:59 And that they make sure that AI is augmented intelligence, not artificial intelligence.
10:05 That it's a technology that makes people better so they can do more, more productively and have a better life as opposed to being replaced.
10:12 We don't want that.
10:13 It'll be interesting to see what happens and how AI50 evolves in those years to come.
10:19 Thank you so much for all the time chatting with me about AI50, Konstantin.
10:23 Kanrek, thanks for doing it again this year.
10:25 And congratulations to all the companies.
10:27 Thank you.
10:32 [END]
Comments