Bubbles pop eventually.
When the dot-com bubble popped in 2000, it caused the worst decline in the NASDQ’s history, dropping by more than 70%. Many companies that had so confidently slapped ‘.com’ on the end of their names suddenly had to face a reckoning, and many did not survive.
They did not survive because aesthetics would only get them so far. The shiny new object, which back then came in the form of a website on the world-wide web, would eventually need to prove its worth.
A bubble popping represents a reversion back to reality and fundamentals, and that is when the promises and predictions of technology are really tested.
And one of the more significant players carrying out this testing are investors. Having a cool product with potential is fine on Day 1. But on Day 2, investors are looking at the prospect of their returns. They want to know what they will end up getting out of it all.
Google was no exception to this.
It had a great product, solving the internet’s greatest problem at the time; finding information reliably using a clean and simple user interface. When the company was founded in September 1998, it was serving tens of thousands of search queries per day. A year later in 1999, this had grown to 3.5 million.
Initially, these queries were being used to provide learnings for improving the user experience for a product that was free. But such efforts were not generating much money:
...despite the splendor of Google’s new world of searchable web pages, its growing computer science capabilities, and its glamorous venture backers, there was no reliable way to turn investors’ money into revenue.1
This lack of revenue became a particular problem after the dot-com bubble popped. Concern started to grow among investors about the company’s prospects, and some even considered pulling out.
So Google needed a way to make money. Having a brilliant search engine that made it significantly easier for people to navigate the internet just simply was not enough. Larry Page and Sergey Brin needed to start thinking about what Day 2 was going to look like.
With growing investor pressure, igniting an apparent state of emergency and a sense of urgency, Google’s founders turned to surveillance capitalism. It leveraged the large datasets about its users to generate behavioural insights about them and ads to search queries. Google doubled-down on its advertising operations and came up with rhetoric to justify its new practices, namely that “securing this holy grail of advertising would ensure relevance to users and value to advertisers.”2
When Day 2 arrived for Google Search, it turned to ads. And it generated billions of dollars.
Now that we are in the midst of an AI hype cycle, with increasing worries about the bubble finally popping, the question is what will Day 2 look like for these various companies betting on AI. How will AI companies justify the huge amounts of capital being poured into model development, data centers and talent?
My prediction is that AI companies will eventually be forced to turn to the business model that still powers many of the ‘free’ internet services today. They will do so because it is a model that has proven itself repeatedly. First it was Google, then it was Facebook, then TikTok, and so on. Revenue from targeted advertising remains king.
But this could be interesting.
Meta has had Messenger and WhatsApp for a long time, yet it has never found a way to reliably produce ads on these platforms. Would OpenAI be able to solve this with ChatGPT? Or what about Anthropic with Claude?
If AI companies do manage to solve it, and if this is the pathway that they turn to when the reckoning comes, then the implications for data rights will be significant:
Purpose limitation will be thrown out the window (if it has not already). AI companies will engage in a mission creep that justifies data originally collected for one purpose (model memory for user improvement) used for another (model memory for advertising).
Data maximisation will accelerate, as more data will be needed to feed the surveillance capitalist machine and garner more behavioural insights to generate more relevant ads.
Regulatory capture may very well take place, whereby AI companies reinforce surveillance capitalism as the norm and lobby governments and legislators to embed this into the political consensus and eventually legislation.
There are some who argue that bubbles are good because they can be catalysts for innovation. It can sometimes produce “bursts of progress” fuelled by “extreme commitment” and “mounting enthusiasm.”3
Maybe this means that the current hype around AI will open up routes to the promised lands of superintelligence. Maybe we will eventually be able to harness technologies that cure cancer, discover new physics and provide ever-lasting abundance to humanity at large.
But before we see the fruits of the hype (if at all), AI companies will need to contend with the reckoning of Day 2. And in doing so, they will probably revert to the safety of surveillance capitalism, as the tech giants before them have done.
When the AI bubble pops, maybe we just get more ads.
Great. Can’t wait.
Shoshana Zuboff, The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power (Profile Books 2019), p.71.
Shoshana Zuboff, The Age of Surveillance Capitalism: The Fight for a Human Future at the New Frontier of Power (Profile Books 2019), p.74.
Byrne Hobart and Tobias Huber, Boom: Bubbles and the End of Stagnation (Stripe Press 2024), p.68.