Gerstner invested in Priceline at $1B valuation; it became Google's largest advertiser and built a $100B business.
“We started investing in it when it was a billion dollar business and Priceline became the single largest global advertiser on Google and build a $100,000,000,000 vertical search business by providing better access to hotels and hotel inventory but that business would not have existed without Google.”
“We started investing in it when it was a billion dollar business and Priceline became the single largest global advertiser on Google and build a $100,000,000,000 vertical search business”
Gerstner notes Google compounded at similar rates to other platforms despite his early worries about search's future.
“And I was talking of I was worried about the future of search and the future of vertical search starting in 2010, But look at Google's compounded at about the rate the other big platforms have in The US.”
Gerstner says Priceline built a $100 billion vertical search business by becoming Google's largest global advertiser.
“We started investing in it when it was a billion dollar business and Priceline became the single largest global advertiser on Google and build a $100,000,000,000 vertical search business by providing better access to hotels and hotel inventory”
Gerstner says Priceline built a $100 billion vertical search business by becoming Google's largest global advertiser.
“We started investing in it when it was a billion dollar business and Priceline became the single largest global advertiser on Google and build a $100,000,000,000 vertical search business by providing better access to hotels”
Gerstner's core Google insight was search volume and revenue per search would both increase over time.
“The high quality insight was simply that Google was going to do more searches in the future than they did in the past and that the revenue per search was going to go up over time.”
Gerstner warns that companies relying on Google's search ecosystem face potential value destruction from AI disruption.
“You're not gonna have to search through 10 blue links. And so, yes, a lot of the companies that were beneficiaries of the Google ecosystem are now suspect.”
Gerstner says top tech leaders now believe the $20 trillion web search market is up for grabs.
“And I will tell you that the conversation around that table, there's increasing confidence that whatever comes next, right, the top of the funnel is up for grabs,”
Gerstner says there's growing confidence that search dollars will redistribute over 5-7 years, even Google must cannibalize search.
“And make no mistake, Google is well positioned to make its claim for that, but its claim is going to come at the expense of search.”
Gerstner argues Google's culture evolved from product-led growth to a research institution.
“the culture at Google had evolved in such a way that it had become more of a research institution than a product led growth institution.”
Gerstner characterizes Google as a monopoly that has been a tax collector on the Internet for twenty years.
“Google has been a monopoly, a tax collector on everything on the Internet for twenty years.”
Gerstner says GPT has become the default verb for AI, not Google's Bard.
“Do you bard or do you chat GPT? GPT has become the default verb for all things AI.”
Gerstner contends GPT has become the default verb for AI, not Google's Bard.
“So what is the verb in the age of AI? Do you bard or do you chat GPT? GPT has become the default verb for all things AI.”
Gerstner characterizes Google as a twenty-year monopoly tax collector on the Internet with monopoly search pricing.
“Google has been a monopoly, a tax collector on everything on the Internet for twenty years. Okay? And they have monopoly pricing on search.”
Gerstner says Google made a mistake letting OpenAI go first despite having the technology and capabilities.
“And to relinquish that spot by letting OpenAI go first, when you had the technology, you had the supercompute, you had the capabilities to release that product, I think was a mistake.”
Gerstner argues waiting until Google's 2004 IPO still captured 95% of all search value ever created.
“you didn't have to invest in search until Google's IPO in 2004 to capture 95% of all ever generated by Internet search.”
“I think there's plenty of time. If this is going to be as big and as transformational as humanity as we all think it will be, remember, you didn't have to invest in search until Google's IPO in 2004 to capture 95% of all ever generated by Internet search.”
Gerstner argues Google 2004 IPO investors captured 90 percent of internet search profits, illustrating patience in venture.
“You could have waited to invest in Google in 2004 and their IPO and you would have captured 90 plus percent of all the profits ever generated in internet search.”
Gerstner conducted search experiment with Paul Reeder showing ChatGPT delivered answer instantly versus three to four minutes on Google.
“I said to Paul, you use Google and I'll use ChatGPT. I got the answer instantaneously. It took him three or four minutes of hunting and pecking around.”
Gerstner notes Google barely reduced headcount from 187,000 to 183,000 over the same period.
“You know, if you look at Google over that same period of time, I pulled the numbers this morning.”
Gerstner contrasts Meta's 87,000 to 67,000 headcount reduction with Google staying flat at 187,000 to 183,000.
“You know? And like I said, they've tripled their earnings per per employee while going from 87,000 to 67,000 employees.”
Gerstner calls Google the greatest monopoly with greatest monopoly profits in capitalism history.
“But we also saw in the case of Google, as I've said, it's the greatest monopoly with the greatest monopoly profits in the history of capitalism.”
Gerstner contrasts Meta's headcount reduction and earnings explosion with Google's minimal changes and weaker stock performance.
“Facebook is actually reduced headcount significantly and their earnings per employee has exploded and Google kind of, you know, they're fine, but they really just haven't made many changes.”
Gerstner says Google's query growth had already slowed to 4% annually before AI, with 10 billion queries daily.
“And so the first thing that we just say is that the basing engagement, even before AI had slowed down a lot, because you're already at 10,000,000,000 queries a day.”
Gerstner calculates AI answers cost 10x more than traditional search: 4 cents versus 0.3 cents per query.
“Now, what does it cost to do that for $7.50 tokens today? And of course, this will go down over time, but it's 10x more, right? It's 4¢ per query.”
Gerstner states Google processes 10 billion queries daily, more than one per human globally.
“So, you know, a couple queries are more than a query for every human on the planet, and it has the most efficient system in the world for doing that.”
Gerstner says Google offers AI engineers $500k cash plus $1.5M in monthly-vesting RSUs.
“Google's offering 500,000 in cash plus a million and $0.5 in RSUs vesting monthly, right, to these engineers. So every single month, those RSUs are immediately available for sale.”
Gerstner says Google offers AI engineers $500K cash plus $1.5M in monthly-vesting RSUs, totaling millions annually.
“Okay? But instead of cash, for example, Google's offering 500,000 in cash plus a million and $0.5 in RSUs vesting monthly, right, to these engineers.”
Gerstner reports using ChatGPT exclusively instead of Google for a 45-minute tutoring session with his son.
“I found myself we never opened Google a single time, Bill. And it really unlocked a lot of learning and led to this forty five minute conversation back and forth that we had.”
Gerstner argues users are shifting from Google's ten blue links to ChatGPT, Claude, and Apple Intelligence.
“Because people are going to turn to Apple Intelligence instead of 10 blue links to find answers to the questions they have.”
Gerstner states Google was dominant in 10 blue links but is not dominant in the new AI era.
“They were dominant in 10 blue links. They are not dominant in this. There are a lot of other players in the age of What”
Gerstner's survey work shows people are turning to ChatGPT, Claude, and Apple Intelligence over traditional search.
“We do lots of survey work on this, Scott. People are turning to chat GPT, they're turning to Claude, they're turning to, why are we so excited about Apple Intelligence?”
Gerstner asserts Google lost its dominance in the AI era despite dominating traditional search.
“They were dominant in 10 blue links. They are not dominant in this. There are a lot of other players in the age of”
Gerstner explains Google released Kubernetes for free as defensive move against Amazon's AWS dominance.
“Google did it again with Kubernetes, which is a piece of technology that lets you move your workloads very easily between clouds. They were worried Amazon was running away with AWS.”
Gerstner says big four tech companies will spend $220 billion on capex this year.
“if you look at the big four, they're gonna spend $220,000,000,000 this year on CapEx.”
Gerstner reports Google confirmed AI-enhanced search results monetize at the same rate as traditional search results.
“one of the key questions there was whether or not the AI enhanced search results were gonna monetize at the same level as non AI enhanced search results.”
Gerstner reports organic search traffic down 20-40% as companies push AI-embedded answers above traditional results.
“They're saying organic search is down 20 to 40%. Why? Because they're putting AI on top of it. They got to push it down the page.”
Gerstner compares Biden's AI diffusion rule to hypothetically blocking Google exports in 2000.
“Can you imagine if in the year 2,000, Washington DC had said, oh my god, Google is so powerful that we're not gonna allow any other country in the world to have access to this Google machine that might give them answers to questions unless Google comes to Washington and gets federal approval”
Gerstner says OpenAI reached 400 billion annual searches eight years faster than Google did.
“I did I saw some analysis from my team this week will post that OpenAI reached, you know, 400,000,000,000 annual searches, eight years faster than Google.”
Gerstner says OpenAI reached 400 billion annual searches eight years faster than Google.
“I saw some analysis from my team this week will post that OpenAI reached, you know, 400,000,000,000 annual searches, eight years faster than Google. Right? So they're doing over a billion searches a day now.”
Gerstner says ChatGPT reached a billion searches daily eight years faster than Google did.
“ChatGPT got to a billion searches a day. Eight years faster than Google did. Is not internet time. Is AI hyper speed. It is way faster.”
Gerstner states OpenAI is four times the size Google was at IPO in revenue and searches.
“OpenAI today is four times the size that Google was when it went public in terms of revenue, in terms of searches.”
Gerstner reports Google page views drop 8-11% peak to trough after users subscribe to ChatGPT.
“And you can see peak to trough is down 8% year over year. Peak to trough is, let's say, down 11.”
Gerstner's data shows Google page views grow 4% annually for users without ChatGPT subscriptions, contrasting with ChatGPT subscribers.
“absent ChatGDP, ChatGPT, maybe Google page views for a particular user are growing 4% per year. So we are consuming more Google.”
Gerstner's analysis shows Google page views decline 8-11% year-over-year for users who subscribe to ChatGPT.
“And you can see peak to trough is down 8% year over year. Peak to trough is, let's say, down 11. So clearly, page views are going down,”
Gerstner says Google page views grow 4% annually per user before ChatGPT subscription adoption.
“what you see on that chart is that absent ChatGDP, ChatGPT, maybe Google page views for a particular user are growing 4% per year.”
Gerstner notes the world changed dramatically in 24 months, switching from Google to ChatGPT.
“Sam's like the world has already changed dramatically over the course of last twenty four months, but people aren't, you know, it's not changed that much about our daily lives.”
Gerstner argues consumer ownership, not hardware, determines value; Google dominated for twenty years this way.
“The battle for the consumer is ultimately where the value accrues, Sonny, not who runs, you know, what hardware.”
Gerstner says Google's monthly token generation exploded 100x in a year, from 9 trillion to 980 trillion tokens.
“Today, it's 980,000,000,000,000 tokens. So from 9,000,000,000,000 to nine eighty, it's a 100 x increase in a year.”
Gerstner reports Google inference generation increased 100x in a year, from 9 trillion to 980 trillion tokens monthly.
“A year ago, Google per month was doing about 9,000,000,000,000 tokens a month in terms of inference generation, right, compute generation. Today, it's 980,000,000,000,000 tokens.”
Gerstner contrasts Google's unrestricted global expansion with current AI restrictions, wants same for AI companies.
“They just went and competed around the world, and now the world runs on the democratic principles that are embedded within Google. I wanna see the same thing happen in the age of AI.”
Gerstner warns that requiring AI companies to get Washington approval for foreign launches would prevent American technology dominance.
“Imagine in 2001, 2002, if we had said to Google, You can't launch in any other country unless you come to Washington and beg for a license to launch in any other country.”
Gerstner recalls Gurley warning Zillow that relying on Google's free links was dangerous and they built brand independently.
“Eventually Google will pull the rug on, you know, these free links. And so they really, unlike a lot of other websites, focus on building the brand outside of Google.”
Gerstner states Google consumed 98% of all upside in search despite 20 competitors with good technology.
“Google had a lot of competitors. 20 competitors in search. They all have pretty good technologies, but Google consumed 98% of all the upside ever associated with search.”
Gerstner argues Google captured 98% of all upside in search despite having 20 competitors with good technology.
“20 competitors in search. They all have pretty good technologies, but Google consumed 98% of all the upside ever associated with search.”
Gerstner predicts Google's TPU will not match NVIDIA Rubin's token efficiency despite advantages.
“First, Google is a massive consumer of NVIDIA GPUs, Right? Not only for its own workloads, but also the workloads, you know, in Google Cloud.”
Gerstner says Anthropic and OpenAI revenue scaling is 10x faster than Google and Meta at comparable scale
“@patrick_oshag Yes! Size of TAM only explanation for Anthropic & OAI revenue scaling - we have never seen anything like it. 10x faster than Google & Meta at comparable scale. 🎯”
Gerstner calls $450B Anthropic 2027 exit run rate too aggressive, sees $250B as achievable base case
“Too aggressive re Anthropic. Base case of $250 B ‘27 exit run rate (same size as MSFT & Meta) seems wild but achievable. $450 B (size of Apple & Google) seems too aggressive given compute & regulatory constraints. Margins also likely lumpy as big compute comes online. 🧐📈 https://t.co/UVbtDz2CSl”