The Bear and Bull Cases for OpenAI: A Deep Dive

This is John and Peter with the venture capital podcast VC fm. Peter's the vc, I'm the. Can I say I'm a former vc?

Sure.

Former vc, now founder.

I just work for John. I'm like his intern.

Yeah, you get paid so much. You're not my intern. I'm your intern.

I learn everything from John.

What have you learned from me?

Oh, man, so many great things of.

What not to do. No, let's wait. Let's go. Let's. Let's hear it. Let's hear it.

You know what I've learned from you? I have. I'm not. I haven't, like, fully incorporated it because I'm not that skilled, but I have learned how to be a better networker, human being and more empathetic by watching the way you treat other people.

I don't know if I'm the best at treating people. I feel like I should be better.

Shouldn't we all?

No, not everyone.

But I think. Look, I think everybody does things better in some areas of their life, even if it's occasionally okay. And I think you do many things better than I do in a lot of areas of your life, and those are areas for me to learn from.

You're more intellectual than I am, by far. You have a higher iq. You are a much better communicator.

I don't know if that's true. I think you're a pretty good communicator.

I think I'm scattered as I made conversations.

I think you're really good with people.

I think I'm good with people for sure.

Which requires good communication.

Can we do a tribute real fast?

Sure.

Justin Bingham, really good friend, passed away this past week. He was an amazing individual, and as I was at his funeral, my limited encounters with him made me feel like I was like his younger brother. And when we went there, there were 12, 000 people who watched the live stream that day. Wow. There was, I don't know, four or five hundred people there.

Yeah.

And like, he was on an entirely. On a completely different level.

That's amazing. So, such. So sad, though, for his family. Yeah. Well, look, we all have somebody. You know, there's. There's. There's always something to aspire to.

What are my flaws?

That doesn't diminish the goodness in us.

What are my flaws?

What are your flaws?

What do you think holds John Bradshaw back?

What is this, like, therapy session podcast part of?

I don't know. It's all. We. We cover all sides of the VC world. What do you See in me that you see in other founders that hang hangs us up and you can pick on me instead of the other founders.

Look, this is something that I struggle with as well, but I think focus is a big one.

Chat GPT would agree with you. I had a therapy session with ChatGPT this week and it said we think you're doing a lot of things.

I think you're doing a lot of things. You should just do one.

I think I've. I'm at the point in my life, in my career where I'm learning how to delegate a lot more.

Yeah. Which is good.

And that's at least for me, not a very easy or intuitive thing to do. But I think Chachi PD gets confused because it thinks I'm doing everything.

Yeah. But that is my other piece of feedback is that you have control issues.

Yeah.

And you like to control everything.

Isn't that what a lot of founders have though that are good or. No, I think I'm an average founder.

Is that characteristics of a great founder that they're very controlling? I don't know that it is.

Okay.

I think it's so great founders like one of the few things that is true across all great founders in my opinion is that they are very controlling and very set and one like one minded focus on their vision. Right. They may have lots of routes to get there, but they do not compromise on where there actually is. Right. And you know, I think Henry Ward over at Carter Carta said this best when he was, when he talked about how like hey look, I gotta swap out C suite executives, right. As the company grows. Cause we need to level up and we need to bring on people to have the right experience. And Ben Horowitz talks about this too, right. And he's like, well then why don't I as the CEO get the boot and level up. Right. And he's like, the reason why is because it's my job to hold the torch of the vision. And I think really effective founders do that incredibly well. And the very best do it their entire lives basically. Right. You look at the Jeff Bezos and the Mark Zuckerbergs and the Larry Pages and the, the Bill Gates of the world, right. Those are all like prime examples of people that like hold that torch of the vision. But at the same time they are not micromanaging. They're letting other people come in and help them and build. If they were truly, truly all about being controlling, they wouldn't raise money from VCs, they wouldn't give stock to their co founders or even to their employees, right? They'd be like, I want to control all of this. I want to have it all. But I think, like, the beauty of, like, startups and this idea of sharing equity is that by going out and getting other smart people involved and partnering with you, you're able to make one plus one equal three or four. Right. And the best founders, that's one of the other things that really denotes a really great founder is somebody who can attract those incredible people and get the most out of them. Right. I think love or hate him, if you look at Elon Musk, that is what he does incredibly well. He brings in amazing people and gets the most out of them.

But he's also been able to be lucky enough to have exits, been lucky enough to be at exits, to ride the top of this. Well, because I feel like. I feel like the point I'm at in my career, it's, yeah, I'm a doer and I'm having these. These oscillating moments where you can start having a team to start delegating more, but you're not fully there. I've got one foot in and one foot out.

But like, this podcast is an example where I offer to help and you don't let me.

Okay, you can. You can publish this podcast. How about that? Sounds good.

If you want me to. But.

The team in India is ready for you to share the files with them after this. You're going to have to sync all the files, submit it to them, let them edit.

Now John's getting upset because I gave him some harsh feedback, just like our last episode. Getting defensive.

I'm not defensive.

Fine, you do it all.

All right, so let's.

We should probably cut this part out. Nobody wants to hear John and Peter bicker like a married couple.

That's what all of our comments are. We should keep it in. So. But let's talk about OpenAI.

All right, let's talk about OpenAI.

You ready?

I'm ready.

So OpenAI did something that's never been done before. They just raised 6.6 billion, which, if I'm correct, that's the largest fund ever raised.

Largest funding round. Yeah.

By a startup.

By a startup.

And my guess is it will be the second largest funding ever raised this year. They'll only be beaten by themselves because my guess is this year they'll raise, or in the next 2412 months, they'll raise another round even larger than the 6.6 billion.

Potentially. Yeah, potentially. That round was, like, massively oversubscribed to. Like, they could have raised more. There was demand to raise more, and.

I think they will raise more.

They probably will. No, yeah, they probably will. So I just thought it'd be interesting to talk about the bear case and the bull. And the bull case and the bear case.

Okay.

As we see it, what is your perspective? Are you bullish or bearish on OpenAI?

Bearish.

Bearish. Are you bearish, like, just in general or just bearish at this valuation?

I am bearish at the company in general.

In general. Okay, so lay out for us your bear case.

My bear case. I think there are other companies that have far more strategic advantages.

Like what?

Like Microsoft.

Even though Microsoft's, like, a major owner of this business.

They are a major owner of this business. But what makes I think Microsoft powerful is they have all these deep integrations. They have a sales team, they have a brand. They've got Bing.

They have Bing.

I think Bing is becoming more powerful. Like, they've got, like, a team that can support something like this.

Okay.

They have a lot of AI, will be in a lot of their of their products, from Microsoft Dynamics to their Office Suite.

Yep.

And I look at AI generally as a feature. That's a free feature that most companies will be expected to have. I was listening to the Sasser podcast this week, and they said. And Jason Lemkin was talking about his perspective, which is that founders will have less funding than they had before. They have to cut things back. But they will also then have to be able to somehow afford an AI team, because that's just the next thing you're going to have a CRM. And if you don't have an AI team who can get rid of all the manual work, you're going to lose to someone else. I think OpenAI.

How is that a bear case on OpenAI?

A bear? I think unless OpenAI can become, like, the next Google.

Huh.

I think they'll just disappear in five to ten years.

Interesting.

So I think they're winning with the younger generation, but they're not winning with entrenched players like they. Like they can or should, in my opinion.

So why, like, what evidence do you have that that's true?

What evidence? You're just asking my general opinion that.

Well, but your general opinion is they're not winning with, like, I haven't looked at larger companies, and I'm just curious.

Why I don't think they're not winning.

Why you believe that?

I think they're making integrations. They obviously have strong partnerships with Google. They've got strong partnerships with Microsoft, but Microsoft has this entire ecosystem.

Sure.

Same with. Same with Google.

Yeah.

So I think if they're going to win, they'll be winning like a Snap Snapchat or like a. Like a social network with a younger generation. I don't use Google. I use Open our chatgpt.

Interesting. So it's really just a branding play. That's all they really got.

I don't. I think it's not. Yeah. Cap. Capturing the minds of the next generation for where do I find answers online.

Interesting.

And I think that's a really hard thing to do.

Yeah. Okay.

Microsoft couldn't do it.

Couldn't do what?

Compared to Google?

Oh, yeah. What they're still numbers couldn't displace Google. The flip side though is I work with a lot of young people, a lot of students, and they're like, I don't even go to Google anymore. I just go to ChatGPT. I think the bear case, the bull case for ChatGPT is one they've gone from basically 0 to 2.7 billion in a year. You don't do that off of a bunch of Gen Z and millennials that are using it for a search engine and aren't paying for it. Right. You do that by selling it to large corporations. So clearly they're getting enterprise sales and they're doing it at a pace that is like unheard of. Like to go to grow revenue that fast. I mean, it's the fastest ever been done and not by a little bit. It's not like they're growing like 10% faster than Google did or Facebook. Like, they're growing like an order of magnitude faster than those companies did. Turning on revenue. They are also the verb. Right. To a certain extent, like, you know, I'm going to like ChatGPT. This is what people say. They don't go like, I'm going to Gemini this. Right. I'm not going to anthropic this. Right. Like when you think AI or you think ChatGPT, they're almost synonymous right. At this point. So they just have this incredible brand as it relates to AI and even to the point where like I think one of the bare case points that's been laid out there by others that I would tend to agree with and I think is the biggest threat personally, is that in the world of AI, and we've talked about this before, like, know kind of our, our thought process of, of where the opportunities and risks reside in AI is that AI is going and is leading to a Proliferation of stuff. That stuff could be content, could be apps, could be all kinds of stuff, right? It just makes it so much easier to create. And when that happens and you have basically free distribution through the Internet, it becomes really challenging to solve the discoverability issue, right? You have to spend a ton of money on marketing and sales. And that's why I believe that incumbents are going to be the true winners of most AI technology because they don't have to get customers. They already have customers. They already have them, right? People come to them, they have attention. It's an upsell, it's an upsell, it's an upsell or it's just part of what they do, right? And because AI makes it so easy to integrate and build solutions and so forth, it's going to be really easy for incumbents to catch up to any startup that is building something on the AI front and implement it and offer it. And you're starting to already see this, right? Like if I go onto Google and I Google search something, what are the very first results? It's all Gemini results, right? It's all AI results and they look very similar to ChatGPT. And it's all built in, right? Facebook's gonna do the same thing, Microsoft's gonna do the same thing, Amazon's gonna do the same thing. All these companies is gonna, are gonna do the same thing in their various niches and in their various ways that makes sense with customers. And because Chat GPT doesn't have necessarily that same reach, it's a major, it's a massive threat, right? Because people have to like change their behavior to use ChatGPT over what they had been using traditionally, whether it's Meta apps or Google Apps or Apple apps or what have you. Uh, I think the contrary argument to that though is that there is a great number of people that are spending like they, they're foregoing those other things and going straight to Chat chatgpt. So I think that's one, I think the other is the contracts that ChatGPT has where they're integrating the solution with lots of different companies. They are getting embedded in those companies in their AI solutions and they are becoming the AI solution on the back end. Which is why I've always believed that the top foundational models will actually be an interesting place to invest because they're so far ahead of all the other models. You can't afford as a business to use like a second rate model. You really want to use the best. Because to your earlier point, if you're not Using the best, you're falling behind and you won't be able to keep up. Right. And so it won't be enough to just be like, oh, well, we could just use Gemini if Gemini is not really the best that's out there. Right. And these models just get better and better. The more people that use them, the more data it can ingest, the more it can train. Right. So I don't know. And then from an innovation standpoint, you know, they continue to lead in innovation when it comes to releasing new product and are definitely ahead of all the other major players. Now, will those other players catch up? Maybe. I don't know. Right. But there tends to be an advantage given to a lot of companies that have a substantially like new breakthrough technology and are continue to imp. Like to improve it over time that it's hard for others to catch up in a meaningful way. So.

And I don't think it's that hard to catch up.

And you disagree. You just don't think it's that hard to catch up.

I think their best, the best thing they have going for them is their brand of being a Google alternative for search.

Yeah, okay, that's, that's. But I don't think that that's interesting. Is it really an alternative to search though? Because I know a lot of people that have basically said I think it's.

More than an alternative to search.

Like, why do you search? At the end of the day, you search to solve problems. And so in my mindset, the thing that's interesting about ChatGPT is that it does a better job solving problems, which is your underlying reason you go to Google. And so why would you go to Google if even, even when Google, like kind of layers in a little bit of chatgpt, it doesn't actually necessarily solve your problem in the same way that ChatGPT does. Right. Like, I have friends that they're like, I didn't know how to program and, or I didn't know how to use this like, complex software. So I just worked with ChatGPT and it taught me how to do the whole thing and it did a big chunk of it too, for me. Right. Like, Google doesn't do that. Google might point to you, point you to some YouTube videos, it might point you to some articles and you can maybe to some apps that you can kind of use to do those things, but it doesn't actually do it for you in the same way that ChatGPT does.

But I don't think, I think the fight at the end of the Day is about owning the mines. I don't think it's about the technology. I don't think Google's LLM or anyone else's is going to be that much.

Better than anyone else's or that much worse.

That much worse. I think the benefit ChatGPT has now is not only is it a search engine, it's a tutor. And Google is still, I believe, fighting the space of where a search tool. If you ask me right now, that's where I think Google is losing. But I think they'll shift.

You think they'll figure it out eventually. And you don't think it'll be too late?

I think. I don't think it's too late. I don't think the company has to view itself not as a search engine more, but we're a problem solving tool.

Yeah. You know, potentially. I mean, if there's somebody that'll pull it off, it would be a Google. But I think it's really hard to move ships when you've been going in one direction for a long time. I mean, that's a lot of momentum to shift and frankly, they haven't been doing a very good job of it today.

I think there's pros and cons. So I don't know, we could see. My guess is, I mean, again, I would be looking at the data of the next generation. Like your students. What are they doing?

And I told you what my students are doing.

They're using ChatGPT, which is what I'm watching.

Yeah.

And if Google can't solve that.

Well, if Google can't, can't win them back. Right. This is like why people. I mean, the older generation uses Facebook and it's still successful and drives a ton of revenue and blah, blah, blah, but like the younger generation doesn't.

We've had an earthquake. Right. We've had an AI earthquake.

Yeah.

It's going to shake up a lot of things.

Yeah.

My question is, in a year or two, what's going to happen?

Yeah.

Just because your students are using it today doesn't mean they'll use it tomorrow.

True, true. There could be something else that pops up, but I don't necessarily see a reason why they're going to use Google.

I mean, maybe Facebook. Mark Zuckerberg, if you look at every app, they're integrating AI into everything.

The problem though is that like Facebook and from my perspective has tried to throw a lot of stuff up against the wall from an innovation perspective and has basically failed repeatedly. And the only reason, in my opinion, why their stock is performing as well as it is and their company is doing as well as it is is because Mark Zuckerberg has said, all right, fine. I'm not going to put quite as much time and energy and cash into these side projects that clearly aren't working. Right. Like look, VR is a great example. Right. That's the whole reason they changed to meta. I love VR. I'm on VR like at least a couple times every week. But it doesn't have mass adoption. Right. It doesn't for. I don't know what the reasons. I mean a lot of reasons. But like doesn't have mass adoption and if you take away, if you take away WhatsApp and you take away Instagram, there's nothing basically left from a long term like opportunity for Facebook at this moment. There's a lot of like hopes and dreams with some of their new product that came out recently, but we don't know how that's ultimately gonna play out. And Facebook is definitely dying. Right. The older generations are the only ones that really use it and as they get older, die off, whatever Facebook's gonna, it's probably gonna struggle. Right. But I think Google's in the same spot. I think Google, like is going to be Yahoo.

Is Google.

I don't think Google's Yahoo. I think Google's like Facebook.

Okay.

I think it's like the older technology and it will get displaced. I think part of the reason it'll get displaced is because AI will get embedded into more and more apps and there will be less need to go to a Google for discovery.

But I don't know, I think, I mean recently I'm trying to use Siri more and more.

Yeah.

It's one of the, the conscious things. I'm actually trying to do this.

And how is that working out for you? Is it good? Is it getting better? Because for a long time it was like, you know, hopes and dreams and totally fell on its face.

I mean, I think for me the issue is like I recently got a was at the Apple some different Apple headphones and they can integrate better.

Yeah.

And allowed me to do things before I couldn't do or I wasn't aware I could do.

What's your favorite thing you do with Siri?

I tell it to skip songs. That's my favorite thing to do. But they're integrating more with like Spotify and everything else. So before you couldn't use Siri with third party apps.

Interesting.

You could do it with and.

And this is why Apple sucks. And you should have gone to Android because I've been able to use, hey, Google to do things like that for the last like five years.

Right. I mean, I think they specifically wanted people to switch to Apple Music and I wouldn't switch to Apple. And now they're having.

Yeah, you know, this is the whole like walled garden. It's a beautiful walled garden, but you're stuck in them in the garden. You can't leave. Yeah, that's fine. It's just a strategy. Right. And it works out really well for them.

So at the end of the day, what, in 10 years, what percent of market share would Google have versus ChatGPT?

Of what?

Search.

Well, of search. I think Google will still own all search. I just don't think search will matter as much.

Okay.

I think that's the fundamental difference.

I see it not as search. Now Google has to see themselves as a problem solving engine. Engine.

Yeah. And so the question is whether or not ChatGPT can be the dominant default problem solving engine before Google can figure out how to do that and catch up before it's too late.

I think if anything, a lot of the tools will be fairly similar.

Yeah, could be, could be.

I think it's the moat and the brand is what made what, which is what will win at the end of the day. Not the, hey, we have a superior LLM. I don't think most LLMs, they're, they're generally sixes.

Interesting. I think, yeah, maybe that's the technology. Having used a bunch of them, I would disagree. But you know, who knows over time if they all kind of plateau. And what's your favorite consolidate into one? Well, it depends on what you're using it for. Right. So for general stuff, chatgpt, but for more, I mean, it just kind of depends. Like anthropic is really good in certain areas. Gemini is fine for search stuff. Right. I get good search results and it's helpful in terms of answering like my question. But I also think like, look, there's this, there's this like fundamental shift that I haven't fully made honestly that I think my students have and younger people in general have. Where. Like, like I was talking to a friend of mine who works at a private equity fund and he was like, the first thing I do when we get a new company is I just ask ChatGPT to explain it to me. Like I'm a 5 year old and then like an 8 year old and then a 12 year old and then a 20 year old. Right. And he's like, it's Great. By the time I'm done, within like 30 minutes or so, I now understand that business at a level that would have taken me hours to understand before. And, and so my point there is, like, when he approaches it, he's not like, just like, hey, tell me what this company does, right? It's like, hey, here's my problem, solve it for me. Right? And I'm still in the mindset. And I think a lot of people like me are still in the mindset of, like, I have a question, I'm going to search and get an answer. And I think there is this, like, fundamental, like, mental shift that the younger generation is doing a, broadly speaking, a better job of which is I'm just going to go to ChatGPT and ask it to solve my problem.

They're learning how to be inputters instead of problem solvers.

Well, maybe, but, you know, maybe they are just outsourcing their, their problem solving abilities to ChatGPT. Or maybe they're saying, hey, I'm going to use ChatGPT to solve minor problems so I can solve bigger problems, right? And I can get leverage off of it, which is really what technology should provide us, right? Everyone, like, freaked. Like, are you going to freak out that and say we shouldn't have cars because then people won't learn to walk? No. Right. Like, you're going to use cars as an enabler to do more things and get more things done, but it doesn't mean that, like, you're going to give up walking.

It's walking because my car cannot get us to this basement. Can get us close. Otherwise you might be in the car all the time.

Maybe we'll just be like, in Wall E. We'll be in our own little, like, hover vehicles get fat and lazy. It's the, the goal of all humanity.

All right, well, let's end up end this episode and for the audience, we're curious, are you bearish or bullish on OpenAI chat GPT?

Yeah. And what do you, what, what is your, your case? Like, what's your hypothesis?

And we want to learn from you. We're learning as we go and we appreciate the feedback. So go to VC FM and you can also follow us on all of our social channels there and we'll see you next time.