
Imagine getting an email from your former intern offering to buy the most valuable part of your company for $34.5B. That's basically what happened to Google's CEO this week. And while the bid from AI-search startup Perplexity seems ludicrous, it’s actually a brilliant move in a much bigger war for the future of search.
Recommendations
My favorite finds
Deals
Ford’s Answer to Cheap Chinese EVs Starts With a $30,000 Electric Pickup
Industry News
Luxury brand Fabergé sold to tech investor for $50mn
World
Trump tells Goldman Sachs CEO David Solomon to replace bank’s economist over tariff predictions
Companies
The Ex-Googler Who Threw a $34.5B Wrench in Google’s Monopoly
It was the kind of move that makes Silicon Valley spit out its oat milk latte.
An AI startup, Perplexity, made an unsolicited, all-cash offer of $34.5B to buy Google Chrome.
Yes, that Chrome—the world’s most popular browser, the front door to the internet for billions, and a crown jewel of the Google empire.
This wasn’t just a bold acquisition attempt. It was a public declaration of war, a shot fired directly at the fortress in Mountain View. But the real genius wasn’t the offer itself; it was the timing.
The bid landed with surgical precision right in the middle of a high-stakes legal battle, where the Department of Justice (DOJ) was asking a federal judge to consider breaking up Google.
This, my friends, is a masterclass in what we’re calling “Remedies Chess”: using a massive, headline-grabbing market offer to influence a trillion-dollar legal conversation.
The Insider Turned Rebel
At the center of this corporate drama is Aravind Srinivas, Perplexity’s co-founder and CEO.
His resume is pure Silicon Valley royalty: an engineering degree from the prestigious IIT Madras, a PhD from UC Berkeley, and research stints at the holy trinity of AI labs: OpenAI, DeepMind, and, yes, Google itself.
He’s the prodigy who saw how the sausage was made and decided he could build a better factory. Now, he dismisses his old stomping grounds as a “giant bureaucratic organization” caught hedging its bets.
It’s the classic David vs. Goliath story, but with a delicious twist: this David once sorted mail in Goliath’s R&D department.
The PR Stunt That Wasn’t (Entirely)
On the surface, analysts screamed “PR stunt.” After all, Perplexity knew Google wouldn't sell.
But dig a little deeper, and you’ll see the cunning strategy. The move was designed to give the judge in the DOJ’s antitrust case a little nudge.
Before Perplexity’s offer, forcing Google to sell Chrome was a radical, almost theoretical, punishment. But by stepping up as a “credible buyer” with a fat check, Perplexity made the unthinkable seem plausible.
It was a low-cost action—making an offer they knew would be rejected—designed to generate enormous media value and reshape the boundaries of a landmark antitrust case.
Google’s Gilded Cage
To understand why a startup like Perplexity even has a shot, you have to understand the trap Google has set for itself.
It’s a textbook case of the Innovator’s Dilemma: when a company becomes so successful with its current business model that it can’t pursue the very innovations that could disrupt it.
Google’s search and ad business pulls in ~$200B+ a year. It’s built on a simple, 2-decade-old premise:
You type a query.
Google gives you ten blue links.
You click one (and see the ads next to it).
AI “answer engines” like Perplexity are designed to blow this up. They give you a direct, synthesized answer, eliminating the need to click any links at all.
This presents Google with a terrifying choice:
A) Cannibalize its own cash cow by building a true answer engine and killing its ad-click revenue.
B) Risk being eaten alive by a hungrier competitor.
One analysis projected that fully integrating ChatGPT-like tech could slash Google’s operating income by a staggering $36B.
This exposes the fundamental conflict at the heart of Google's business: is the customer the user or the advertiser?
For 20 years, Google masterfully served both. But answer engines force a choice. They are built for the user. Google’s current model is financially optimized for the advertiser.
This is the crack in the armor that startups like Perplexity are aiming to exploit.
The New Ad Playbook: Selling Questions, Not Clicks
As Perplexity attacks Google's core product, it's also trying to reinvent the business model. Forget bidding on keywords. Perplexity is rolling out “sponsored follow-up questions.”
Here’s how it works:
You ask about the best running shoes.
Perplexity gives you a detailed, AI-generated answer.
Below it, you see suggested follow-ups. One, labeled “Sponsored,” might say: “What are the key features of the latest Nike Air Max?”
If you click it, you don’t go to Nike’s website. Perplexity generates another AI answer, this one approved by Nike, detailing the shoe's features.
The real kicker is the pricing. Perplexity is charging on a Cost-Per-Mille (CPM) basis, meaning brands pay for every 1,000 times their question is shown, not clicked. This is how TV and billboard ads work, but it’s heresy in the world of search, where Cost-Per-Click (CPC) is king.
Here’s a look at the philosophical divide:
Metric | Google Ads | Perplexity Ads |
Model | Cost-Per-Click (CPC) | Cost-Per-Mille (CPM) |
Pricing Basis | User clicks an ad link | Ad is shown 1,000 times |
User Action | Drives traffic to an external site | Deepens engagement on Perplexity |
Advertiser Goal | Performance / Direct Conversion | Brand Awareness / Influence |
Reported Cost | Varies (e.g., ~$1-$2 avg. CPC) | High (>$50 CPM) |
The ad world is skeptical. The high CPM rates for an unproven platform are a tough sell for marketers who need to prove direct ROI. But Perplexity is betting that in the age of AI, influencing the conversation is more valuable than winning the click.
The Antitrust Elephant in the Room
Looming over this entire battle is the U.S. government. The DOJ has already convinced judges that Google acted as an illegal monopolist in both the search and ad-tech markets.
An internal Google doc even compared its ad-tech strategy to Goldman Sachs owning the New York Stock Exchange. Yikes.
Now, the battle is over the punishment. The DOJ has asked for the corporate death penalty: a forced breakup, which could include selling off Chrome.
This legal firestorm created the perfect opening for Perplexity’s bid. Court testimony has since revealed that OpenAI would also be interested in buying Chrome if Google were forced to sell.
The lawsuit has effectively hung a “For Sale (Maybe)” sign on one of Google’s prized assets, emboldening rivals and accelerating the very market disruption Google was already facing.
The AI Search Royal Rumble

While the Perplexity vs. Google narrative is juicy, it’s just one match in a much larger corporate cage match.
Here’s a snapshot of the AI chatbot market as of August 2025:
ChatGPT: 60.4%
Microsoft Copilot: 14.1%
Google Gemini: 13.5%
Perplexity: 6.5%
Claude AI: 3.5%
ChatGPT is the champ, but look at the growth. Perplexity, while small, is growing at nearly twice the rate of the leader (13% quarterly growth vs. ChatGPT’s 7%).
Each player has a different strategy:
OpenAI is turning ChatGPT into an “agent” that can browse the web and run code—a new kind of operating system.
Microsoft is waging a two-pronged war: hitting Google on the consumer front with Copilot and selling the shovels in the AI gold rush with its Azure enterprise tools.
The Bottom Line
This isn't just a turf war. It’s a fundamental shift from the “search engine” era (getting a list of possibilities) to the “answer engine” era (getting a conclusion).
The implications are huge:
For users: The experience will be more direct and conversational. But it raises big questions about algorithmic bias and who checks the AI’s work.
For advertisers: The reliable click-based model that built digital empires is now in question.
For businesses: Any company built on Search Engine Optimization (SEO) is facing a potential apocalypse. If Google and its rivals stop sending referral traffic, entire business models could collapse.
Aravind Srinivas didn’t just throw a wrench into Google’s business; he and his competitors are throwing a wrench into a 20-year-old global habit.
The battle isn't just for market share. It's for who gets to define our relationship with knowledge for the next generation.
Community
Open Forum
This isn't a software company. It's a private intelligence agency trading on the S&P 500. The moat isn't the tech, it's the fact the US Army is a client on its platform.
No lies told…
-Matt
Have a Question? Let's Hear It!
I love building this newsletter with you in mind, and my favorite part is hearing from you directly.
Have a burning question, a great idea, or a thought you'd like to share? Send it my way! I'll pick some of our favorite submissions to answer and feature in next week's edition.
P.S. I read every single submission! Let me know if you'd prefer to remain anonymous.
See you next week,
— Matt
P.S. If you enjoy my emails, move’ them to your primary inbox and let me know by sending a reply or clicking the poll below. I read every response.