Google's Dipping a Big Toe. Should It Just Jump In?

As hype escalated around the debut of ChatGPT more than a year ago, I predicted that OpenAI and Microsoft would rapidly develop consumer subscription service models for their nascent businesses. Later that year I wrote a piece speculating that Google would inevitably follow suit. If Google was smart, and careful, it had a chance to become “the world’s largest subscription service.” From that piece:

Google can’t afford to fall behind as its closest competitors throw massive resources at AI-driven products and services. But beyond keeping up, Google finds itself in an even higher-stakes transition: Its core business, search, may be shifting into an entirely new consumer model that threatens the very foundation of the company’s cash flow spigot: Advertising. 

Search is currently the world’s largest distributor of human attention, but that will change if generative AI becomes our presumptive approach to navigate information, services, and commerce. We’re nowhere close to that happening, but it’s not hard to imagine that empty search box being replaced by conversations with presumptively intelligent chatbots. And if Google decides to proactively drive that shift, it’ll come that much quicker.

What I find interesting is how similar generative AI’s inputs are to the search firmament we know so well. Search works by maintaining a fresh, comprehensive index of everything on the Web – exactly the approach taken to train and maintain today’s successful chatbots. Generative AI is the direct descendant of search. In my 2005 book, I imagined search evolving into what at that point was confined to science fiction:

Search is a catalyst in promising attempts at cracking one mankind’s most intractable problems: the creation of artificial intelligence. By its nature search in one of the most challenging and interesting problems in all of computer science, and many experts claim that continued research into its mysteries will provide the commercial and academic mojo to allow us to create computers capable of acting, by all measures, like a human being…search will more likely become intelligent via the clever application of algorithms which harness and leverage the intelligence already extant on the web – the millions and millions of daily transactions, utterances, behaviors, and links that form the web’s foundation – the Database of Intentions. After all, that’s how Google got its start, and if any company can claim to have created an “intelligent” search engine, it’s Google. “The goal of Google and other search companies is to provide people with information and make it useful to them,” [early Google employee Craig] Silverstein told me. “The open question is whether human-level understanding is necessary to fulfill that goal. I would argue that it is.”

Search built the modern Internet as we know it – warts and all. Every moment of every day, tens of millions of businesses compete to win search’s near-infinite auctions of human attention. The substrate of this trillion-dollar economy is advertising – but ever since ChatGPT’s launch, the world has wondered … what if it’s not? Google’s success begat OpenAI’s success. As I wrote in the aforementioned post, the race is on to create a new class of products that offer more value than the once-magical list of blue links. And so far, the model underpinning that race isn’t ads, it’s subscriptions.

Google cannot afford to lose that race.

Hence this week’s news that Google will consolidate all of its AI efforts behind one brand: Gemini. Central to that offering is a subscription service priced identically to OpenAI: $20 a month. Google has been characteristically cautious in rolling out this new approach to consumer monetization – rolling it into something called “Google One,” an upgraded photo storage option which apparently already existed (who knew?!). I live in Google’s world, but I have a “Workspace” account – so I can’t get Gemini with Workspace accounts (note to Google, please please fix this!). Gemini’s new consumer app – which is free, and supposedly a replacement for Google Assistant – is still “coming to select users and countries.” It has a paltry 100K downloads and a lukewarm 3.4 star rating. Not exactly a hit.

I’ve still got serious questions about whether generative AI will ever create a product as universally useful and addictive as search.  But the ball is in Google’s hands now, and the company will be judged by how well it runs with it.

—

You can follow whatever I’m doing next by signing up for my site newsletter here. Thanks for reading.