Can Buying Perplexity Save Apple on AI?
26:00 Share

Today on the AI Daily Brief, is Apple buying perplexity even a big enough move? Speaking of moves, apparently Mark Zuckerberg is trying to just buy everyone. The AI Daily Brief is a daily podcast and video about the most important news and discussions in AI.

Quick announcements before we dive in. First, thank you to today's sponsors, Blitzy, Vanta, Agency.org, and Superintelligent. To get an ad-free version of the show, go to patreon.com slash ai-dailybrief. And if you are interested in sponsoring the show, shoot me a note at nlw at breakdown.network. We are currently selling for the fall and even into next spring. But with that, let's get into the headlines. Welcome back to the AI Daily Brief Headlines Edition, all the daily AI news you need in around five minutes.

There apparently is no one good and technical in AI that Mark Zuckerberg has not tried to buy. The latest news is a rumored acquisition of Nat Friedman and Daniel Gross's investment fund. The information reports that talks are underway to partially buy out the fund, which

which has just an absolute murderer's row of investments. These guys are some of the absolute most prolific investors in all of AI and were early in Perplexity 11 Labs, as well as Ilya Sutsgever's Safe Superintelligence. It's not exactly clear how much of the fund would be bought out versus just them coming, but whatever combination it is, it looks like the price tag would be over a billion dollars. Now, just like in previous cases, there is clearly a big focus on the people here themselves.

If the deal goes through, both partners would join Meta to work under Alexander Wang in this new star-studded superintelligence division. Prior to forming the fund, Gross led machine learning efforts at Apple, including as part of the early development of Siri. The product was based on technology that Apple acquired from his startup, Q. As well as investing in safe superintelligence, Gross is also listed as a co-founder alongside Sutzkever and serves as the company's CEO.

Nat Friedman is a serial founder who made his way to Microsoft in 2016 via acquisition. In 2018, when he was a vice president at the company, he was appointed CEO of the newly acquired GitHub. He parted ways with Microsoft in 2021 to focus on venture investing. And apparently Friedman has been involved with Meta's AI efforts for a little while now.

The information writes, in May 2024, Friedman joined an advisory group to consult with Meta's leaders about the company's AI technology and products. Earlier this year, Zuckerberg asked Friedman to lead Meta's AI efforts altogether. Friedman declined, but helped brainstorm other candidates, including Wang. And apparently Zuckerberg had been skeptical that Alexander Wang would leave scale, but it was Friedman who convinced him that a deal was possible. Now, frankly, the numbers are not adding up here for me.

Freeman and Gross's investments are incredibly valuable, and it seems like sources say that Zuckerberg doesn't really care about the assets and is just looking to acquire the talent. But in addition to that, Save Superintelligence just raised $2 billion at a $32 billion valuation just a few months ago. Now, you have to think that that is largely based on Ilya's reputation, but still you also have to think that Gross leaving for Meta would materially impact those investments.

Interestingly, and this gets back to Zuckerberg trying to buy everyone, apparently he also tried to buy safe superintelligence earlier in the year as well. The sourcing is very light on that particular deal, just saying that Ilya rebuffed Zuckerberg's attempt, but we don't even really have the exact timing.

Omar Kanji of Dragonfly Capital wrote that the deal would be a big one. He writes,

a range of seed bets, runway to a vertical stack, and info rights across the landscape. Mixes strategic investment and brand building and business development and talent acquisition and strategy all into one deal. Now, one more piece of news on Zuckerberg's acquisition efforts. Over the weekend, The Verge reported that he had also made offers to buy Perplexity and Miramarati's thinking machines as well. Writing in the Command Line newsletter, Alex Heath commented, At this point, it's becoming easier to say which AI startups Mark Zuckerberg hasn't looked at acquiring.

Speaking of Miramarati, the former CTO of OpenAI has apparently closed $2 billion in funding at a $10 billion valuation. This is one of the largest early-stage funding deals in venture history. The six-month-old startup is still pre-product and hasn't even made it clear what they're working on. Buko Capital was one of about a billion tweets like this one. I need $2 billion. No, you may not know about the product. No, I will not share anything about the financials. No, your voting rights won't matter. You have 24 hours.

He shared a pull quote from that Financial Times piece that reads, Because of its highly clandestine nature, a number of funds that Marotti pitched to passed on the deal. One of those people added Marotti's pitch offered no information about her product or financial plans. So what does she have? Well, of course, the answer is talent.

They now have a roster of more than 30 leading researchers drawn from the big AI labs, which at Zuckerberg's $100 million a head offer is at least $3 billion worth right there, right? Reporting states that Andreessen Horowitz led the round with participation from Cerro-Guo's conviction partners. It confirmed that Murati will hold majority voting rights, ensuring she retains complete control over decision-making at the company.

Lastly, speaking of OpenAI, there was quite a bit of hullabaloo over the weekend when people started to notice that all of the information around their deal with Johnny Ive had suddenly been scrubbed from the website.

People ran to speculate that somehow the bromance between Johnny and Sam Altman had come apart, but it turns out it has to do with a lawsuit. OpenAI later clarified, writing that this page is temporarily down due to a court order following a trademark complaint about our use of the name IO. We don't agree with the complaint and are reviewing our options. Bloomberg's Mark Gurman writes, "...the Johnny Ivan OpenAI deal is on track and has not dissolved or anything of the sort."

They were sued over the name I.O., and there was a restraining order issued by the judge. They had to pull all materials with the name. The legal threat has some echoes of what Johnny Ive went through over a decade ago in bringing the phone to market. At the time of launch, Cisco owned the trademarks to iPhone and iOS. And apocryphally, Steve Jobs steamrolled them by claiming that they had no legitimate intention to use the iPhone trademark. And yet, Apple still pays a licensing fee to use the iOS brand.

Menlo Ventures' Didi Das, though, thinks that this is a lot bigger than just a trademark. He tweeted portions of the lawsuit and wrote, "...breaking. Google X spin out IYO, which makes smart earbuds from 2018, alleges Sam Altman and OpenAI heard their pitch, got Johnny Ive to try it before copying it, buying his company for $6.5 billion, and calling it IYO. Most dramatic must-read tech lawsuit this year."

I'm sure that story will continue to evolve and we will keep an eye on it. But for now, that is going to do it for today's AIDB Brief Headlines edition. Next up, the main episode. This episode is brought to you by Blitzy. If you're a technology leader, here's something that probably sounds familiar. Your organization's competitive edge is buried in legacy code that desperately needs modernization, but the resources required feel out of reach.

That was the case for a global investment analysis firm. They needed to migrate 70,000 lines of complex MATLAB financial algorithms to Python. Algorithms that drive investment decisions for trillions in assets. Their estimate? Months of high-cost specialized engineering work. Instead, they partnered with Blitzee.

Blitzy's autonomous AI preserved mathematical precision and generated over 80% of the new codebase, completing the migration with just five days of engineering time. They cut the timeline by 95% and saved 880 engineering hours. If your organization is facing similar modernization challenges, visit blitzy.com to schedule a consultation and discover how AI-powered development can transform your technical capabilities.

Today's episode is brought to you by Vanta. In today's business landscape, businesses can't just claim security, they have to prove it. Achieving compliance with a framework like SOC 2, ISO 27001, HIPAA, GDPR, and more is how businesses can demonstrate strong security practices.

The problem is that navigating security and compliance is time-consuming and complicated. It can take months of work and use up valuable time and resources. Vanta makes it easy and faster by automating compliance across 35+ frameworks. It gets you audit-ready in weeks instead of months and saves you up to 85% of associated costs. In fact, a recent IDC whitepaper found that Vanta customers achieved $535,000 per year in benefits, and the platform pays for itself in just three months.

The proof is in the numbers. More than 10,000 global companies trust Vanta. For a limited time, listeners get $1,000 off at vanta.com slash nlw. That's v-a-n-t-a dot com slash nlw for $1,000 off. Today's episode is brought to you by Agency. Agency.

an open source collective for interagent collaboration. Agents are, of course, the most important theme of the moment right now, not only on this show, but I think for businesses everywhere. And part of that is the expanded scope of what agents are starting to be able to do. While single agents can handle specific tasks, the real power comes when specialized agents collaborate to solve complex problems. However,

Right now, there is no standardized infrastructure for these agents to discover, communicate with, and work alongside one another. That's where Agency, spelled A-G-N-T-C-Y, comes in. Agency is an open-source collective building the Internet of Agents, a global collaboration layer where AI agents can work together. It will connect systems across vendors and frameworks, solving the biggest problems of discovery, interoperability, and scalability for enterprises.

With contributors like Cisco, Crew.ai, Langchain, and MongoDB, Agency is breaking down silos and building the future of interoperable AI. Shape the future of enterprise innovation. Visit agency.org to explore use cases now. That's A-G-N-T-C-Y dot org.

Today's episode is brought to you by Superintelligent, specifically agent readiness audits. Everyone is trying to figure out what agent use cases are going to be most impactful for their business, and the agent readiness audit is the fastest and best way to do that.

We use voice agents to interview your leadership and team and process all of that information to provide an agent readiness score, a set of insights around that score, and a set of highly actionable recommendations on both organizational gaps and high-value agent use cases that you should pursue. Once you've figured out the right use cases, you can use our marketplace to find the right vendors and partners. And what it all adds up to is a faster, better agent strategy.

Check it out at bsuper.ai or email agents at bsuper.ai to learn more. Welcome back to the AI Daily Brief. Today we are once again talking about Apple and its AI strategy, and for the first time in some time, some interesting indications that they may be looking to make some big moves.

By way of background, Apple had been conspicuously absent from the AI conversation since ChatGPT launched. While Microsoft was investing tens of billions into OpenAI and Google and Amazon were investing in rival Anthropic as well as investing in their own models, Apple was sort of just sitting back on its heels, not quite sure what it wanted to do. Now, for a little while, this wasn't all that surprising.

Apple historically has not been the first mover. They always try to be the best mover. They try to innovate by skating to where the puck is going, but will often let people bumble around with nascent half-formed experiences before they bring it all together and really get it right for consumers. And so when in 2024, at their Worldwide Developer Conference event, Apple announced their Apple Intelligence Platform, the way that they were describing it made a lot of sense. This was basically going to be AI for normies.

Rather than selling a whole bunch of advanced features, it was really simple, clear integrations into daily life that Apple was betting would actually be useful for the average consumer who didn't care what AI stood for. The problem was that none of it was really ready. The features that were released were underwhelming, and many of the features that were promised were just never released. Most notably, Siri has not gotten anything resembling an upgrade, making it just embarrassingly far behind at this point.

And while one might be tempted to say, does this even matter to Apple? Or is this just the type of thing that all of us who are deep inside the AI space think about? It definitely matters to Apple. Apple intelligence was a big part of the logic for getting people to upgrade their iPhones. And getting people to upgrade their iPhones is pretty much the way that Apple makes its money. iPhone sales are by far its biggest source of revenue. And so if people decide that each incremental update is not enough to get a new phone or a new device, that actually hits their bottom line.

Now, recently there have been indications that Apple is taking this seriously. They've been moving executives around and it seems like trying to potentially rethink their strategy here. But another WWC came and went, and this time they basically self-consciously did not engage with AI almost at all.

There were no major AI announcements, nothing about a better Siri. The only thing that AI folks were tracking that was somewhat exciting was Apple potentially opening some of their models to developers in a bid to get them to build more AI-related things for iOS and macOS.

All of this has left the discourse for some time focused on what Apple can do to actually get back in this game. I did a show a few months ago, for example, called Six Hail Marys for Apple's AI comeback, and right at the top of my list was a big M&A move.

Specifically, I thought that they should buy Anthropic, and I am far from the only person to suggest this. On May 22nd, Parker Ortolani tweeted, "Fastest way for Apple to solve its problem is to buy Anthropic and let them rip." But who knows how realistic that is? Probably not. Now, the tweet generated a lot of discussion, most of which was people articulating why Anthropic just wouldn't be interested. Most people pointed out that it doesn't seem like Anthropic has any real interest in that even if theoretically Apple has enough money, but that still hasn't stopped people from speculating.

Couple weeks later, Josh Avant wrote, Apple should just buy Anthropic. Anthropic is doing well in the developer market, their aesthetics are so good, and they have pretty righteous ideals. Apple could buy them in cash at par for their $61 billion valuation and still have $100 billion left.

Again, everyone popped in to say that Anthropic wouldn't be interested, blah, blah, blah, blah, blah. But you get the idea. This is where the state of the discourse has been. Now, interestingly, Apple has had some interaction with Anthropic. Back in May, Bloomberg's Apple whisperer Mark Gurman reported that they were teaming up to build something akin to a vibe coding platform. The summary was this. Apple is partnering with Anthropic to develop a new vibe coding software platform that uses AI to write, edit, and test code on behalf of programmers.

The system is a new version of Xcode, Apple's programming software, that will integrate Anthropic's Claude Sonnet model, and Apple will roll out the software internally before deciding whether to launch it publicly. The partnership reflects Apple's greater willingness to partner with others after struggling to develop homegrown technology, and the company is expected to add Google Gemini as an alternative to ChatGPT later this year.

Still, it is a far cry from collaborating with someone on a next-generation coding platform to actually acquiring them. And outside of the speculation of podcasters and Twitter denizens, there's been nothing to suggest that Apple was actually looking in that direction. And yet, we did get an interesting little nugget at the beginning of this month. This report is from Deirdre Bosa.

in a very rare disclosure during the Google antitrust trial, one of them, Apple's Eddie Q revealed that the company had begun talks with perplexity, saying they were impressed with the product and exploring integration. And I spoke to Dimitri Shevelenko, that's the startup's business lead, and he told me that they were just as surprised to see Q go public because if there's one rule when dealing with Apple, it's this, you don't

talk about doing business with Apple. Yet here was Q breaking protocol and signaling serious interest. That's a big deal.

There was clearly some interaction happening between Apple and Perplexity. Although at the time, it seemed like it was just about some sort of integration. Although based on the reporting, it was a pretty serious integration. Some even suggested that they were looking to adopt Perplexity as an alternative to Google on the iPhone. So lots of people took notice, but mostly just went on with their lives. Until the very end of last week, when Bloomberg published this piece. Apple executives have held internal talks about buying AI startup Perplexity.

Once again, this comes from Mark Gurman, who is the best sourced person inside Apple at this point, of all the major media outlets. Gurman writes,

What would be included in this deal? Well, part of it looks to be, indeed, as we discussed, a way to fill the gap of a potential breakup with Google. Now, at this stage, Apple's partnership with Google, which makes them the default browser on their devices, makes Apple about $20 billion a year, but could be cut off because of these antitrust cases. Another reason to potentially bring in perplexity is, of course, to get a big dose of AI talent.

As you're well aware at this point, if you are a regular listener, the extreme length that big tech is going to to get premium AI talent has reached a new level with Mark Zuckerberg running around with $100 million offers, completely resetting the compensation base for top engineers.

Now, what's clear at this stage is that Apple is serious about a potential deal of some kind with Perplexity. Apple's AI team has been actively evaluating Perplexity's tech. But at this stage, to the extent that the Apple executives have held internal conversations about an acquisition, they haven't apparently discussed it with Perplexity.

Said Perplexity, we have no knowledge of current or future M&A discussions involving Perplexity. In fact, on Perplexity's side, it just raised another round at a $14 billion valuation, and it also appears to be very close to a large deal with Samsung as well. Given that Samsung is basically Apple's biggest rival in the phone market, this could be something of a challenge for any sort of M&A.

So what do people think about this deal? Well, after the reporting came out, Gurman took some time to get editorial and wrote a piece called Apple Will Need to Leave Its M&A Comfort Zone to Succeed in AI.

Gurman points out that whereas big acquisitions have long been a part of Google and Meta's playbook, think the massive WhatsApp acquisition, or maybe the best acquisition of all time, Instagram in the case of Meta, and of course, Google buying assets like YouTube, Apple's biggest acquisition ever was only $3 billion, and that was for Beats back in 2014. In fact, Apple has only made three M&A deals over a billion dollars in its entire history. And of the deals they've done, they haven't exactly gone well, or at least they haven't been easy.

Because of this, Apple's team is default skeptical of big M&A. And yet, as Gurman points out, Apple is facing a very different set of circumstances these days and may have no choice but to do a deal. The stakes of the AI race are just too high. He goes on, "...developing the next generation of products from glasses to robotics to new wearable devices will hinge on AI. The technology will be as foundational to those products as the multi-touch interface was to the iPhone, and that notably came via Apple's 2005 acquisition of Fingerworks."

Apple's in-house AI work has floundered. Sure, the company has its talking points. Apple intelligence is smoothly integrated and rivals don't protect privacy the way it does. But the truth is clear. Apple has missed the AI moment, and it should accept that instead of making excuses. If Apple had built something as groundbreaking as ChatGPT, it would be celebrating it from every rooftop. Instead, the company is downplaying the competition and branding its offering as AI for the rest of us because that's really all it can say. Apple is doing its best with what it has and will continue to shrug off any concerns until it can offer something better.

However, as Gerben points out, the good thing is that Apple has more than $130 billion in cash. So then, with that huge stock, who could be the targets? OpenAI is definitely out, given that it's at a $300 billion valuation, to say nothing of its weird corporate structure that is a boondoggle with its existing relationship with Microsoft.

Anthropic is valued at a range that theoretically they could afford, but given its relationships with Amazon and Google, and just its general size, there's a concern that an Anthropic deal would bring serious regulatory scrutiny. And so that brings Gurman back to perplexity. He gives five reasons that he likes the deal.

While the company doesn't have its own foundation models, what it does have is one, a proven consumer-ready product, two, a clear need it fills in the context of Apple, i.e. a strong search layer and a conversational interface for everyday tasks, number three, a decent-sized team, 250 employees with deep AI talent,

Others wrote about this as well. Alex Kantrowitz came out stronger with a post, Kantrowitz writes,

Tim Cook ought to call Perplexity's CEO, Aravind Srinivas, and offer him $30 billion for his AI search engine. And he should do it right away. Apple buying Perplexity is such an obviously good deal for both companies that I feel silly even writing it down. Apple would get a bona fide AI service it could plug right into Safari and Siri that would revamp its disappointing AI offering. Perplexity would get access to the $2 billion-plus Apple devices in circulation and become an AI force on par with ChatGPT.

When Kantrowitz asked Perplexity Chief Business Officer Dmitry Shevelenko about a potential deal with Apple, he said, not likely, but the Metascale deal is so unlikely that I feel like we aren't living in a world of likelies. Kantrowitz also points out that although the Samsung deal is emerging, that only puts a finer point on Apple's need to move quickly before it gets totally locked in. Parker Ortolani, who we heard from earlier in the show, also wrote about this in a piece called The Curious Case of Apple and Perplexity. Do they need each other?

Parker says that when he read Gurman's piece, his initial reaction was that wouldn't work. He writes,

Perplexity, on the other hand, is nimble, fast-moving, extremely daring, and can be the young startup that can take the kind of risks that Apple can't. To say that you can probably have a lot more fun working at Perplexity right now is likely an understatement.

And yet, after he thought about it a little bit more, he started to see the logic. Now, from Apple's side, like Kantoritz, he finds it pretty obvious how perplexity could slot in, especially in a world where Google Search is forced off the iPhones by antitrust regulators. He also points out that, of course, what Apple brings to perplexity is just this massive set of larger users. However, he also makes the point that I think is a really salient one, that perplexity in Apple, while very different in terms of where they are in their life, might actually be too close.

He writes,

The reality is, though, that Apple already knows how to do that. Of course, only if they can get out of their own way. A company like Anthropic, a foundational AI lab that develops models from scratch, is what Apple could stand to benefit from. That's something that doesn't just put them on equal footing with Google. It's something that also puts them on equal footing with OpenAI, which is arguably the real threat.

He concludes, At the end of the day, I suppose that I've made the case both for acquiring perplexity and against it. Ultimately, either path is likely good for users, and I'll be happy no matter what happens. The whole situation is great for perplexity. It helps further legitimize them. But as a fervent observer, I keep coming back to the same two questions. Is perplexity really what Apple needs? And is perplexity better off without Apple?

To wrap up here, let's answer the second question first. Is perplexity better off without Apple? On the one hand, it is absolutely the case that unless Apple is extraordinarily hands-off, which almost structurally they really can't be, perplexity is going to be able to innovate faster without them.

At the same time, Perplexity has already done honestly majestic work to carve out as much mind and market share as they have, given that their feature set is largely overlapping with core functionality of all the foundation models. Chachapiti, Claude, and Gemini are all going to munch closer and closer to Perplexity's core business, and it is a very real possibility that they just get squeezed out.

An Apple deal which puts them directly on 2 billion devices is obviously very beneficial there. So the trade-off is potentially ability to continue to innovate versus bulwark against increased competition.

When it comes to the question of, is perplexity really what Apple needs? I understand what Parker is arguing about them both being product DNA companies and Apple not having that technology DNA. However, for as much as Apple has historically been a product company, they have not done a good job of that when it comes to AI. Their vision for how to think about AI is very tepidly something that seems compelling, but they haven't executed it at all in any meaningful way

And so I don't know how much we can actually argue that they're great product makers, at least in this new space. I think that the big question is if they did do an acquisition like this, how much they let perplexity actually lead their product strategy versus just be something off to the side. The team that's running the show at Apple at this point has shown themselves to be incapable of competing in the right way. If an acquisition doesn't involve actual leadership integration, I don't know that I think it's going to work.

Still, like Parker said, as a consumer, I don't think we can possibly lose in this. Either perplexity gets the muscle to do even bigger and better things, or they get to stay independent and keep innovating incredibly fast. Either way, all of us win. Still a super interesting moment and a story that is worth following for sure. But for now, that is going to do it for today's AI Daily Brief. Until next time, peace.