How Google Became The Incumbent

AI shots were fired in the past week, and one company’s volley missed it’s mark entirely.

We’re witnessing an interesting, almost seismic, shift in the tech industry these days. And no, I’m not talking about the layoffs that have swept the tech industry, tragic as they are (but tech isn’t somehow impervious to macro economic developments). I’m talking about Microsoft’s recent rise to the position of cool kid in the class.

For the longest time, Microsoft had sort of taken the role that IBM used to have: the old-school, corporate tech company. Dependable and profitable, sure, but also the sort of tech company where the employees wore suits (*shudder*). It was a long trip from dreary Redmond to the hip, fun, and innovative Silicon Valley.

But over the past few years, Microsoft has made some interesting moves. Not super public ones, but if you knew to look, they were there. And on November 30th 2022, we saw what those move were to come to, when OpenAI, which Microsoft made a huge investment in, launched ChatGPT.

ChatGPT, of course, is a conversational AI based on the company’s already revolutionary (but much less hyped) GPT-3. And it took the world by storm. Even to the point of some people claiming that we had now invented Artificial General Intelligence (we haven’t) and that the Singularity had come (it hasn’t).

One speculation that did catch my eye was about Google. How would Google respond? If we can now access all the info we want by asking a language model a real, human-phrased question, and have it come back with an answer, a real answer, rather than a list of the websites where you were most likely to find your answer (most of the top ones SEO optimized up the wazoo, so really, they could be about anything). Would Google survive this?

Even asking the question shows the seismic shift: “How would Google respond?” It used to be “What Would Google Do”, or, to use the title from Eric Schmidt and Jonathan Rosenberg’s 2014 book; “How Google Works“. Everyone wanted to get inside the Mountain View hallow halls and figure out just how Google worked, how they could emulate, and how they could align themselves with what Google would do next. Now, everyone was asking how Google would respond to someone else’s innovation. Google was no longer in the driver’s seat.

Then of course came the return salvo from Google: they would launch their own chat AI (three months later, and obviously a reaction to ChatGPT, but still) called Bard. But the presentation from their Paris office didn’t impress. It seemed rushed, desperate, nothing like the usual suave and cool Masters of the Universe vide we usually get from Google. They were on their heels.

It probably didn’t help that was presented didn’t impress, either. And it definitely didn’t impress their investors. A single wrong answer (on the Webb telescope) sent their stock price plummeting by as much as 9 per cent. It simply unfinished, which can be OK when you’re a company that has always prided themselves on the fact that most of their products are in near-perpetual beta. But when you launch something like it’s a finished product, and performs worse than your competitor’s “still in sandbox” product, it can’t help but throw you off your game.

But the real shift, as I mention above, is that this has put Google on the defensive. After being arguably the tech leader in more than two decades, and one of the OG disrupters, they’re not playing defence.

As one of the strongest voices in AI, Pete Huang, puts it, they are faced with the problem that chat search is not only fully mature yet, something that’s a much bigger problem for Google, where search is the dominant part of their business, than for Microsoft, where search is a small part of the overall turnover.

Add to that, as Huang notes, that Google has to make the shift to AI and chat-driven interfaces while “maintaining their current product and revenue, and with Wall Street watching”. An ambition to do something new, but not being willing to risk the status quo is the thinking of an incumbent, not a disrupter.

Paul Buchheit, who was the creator of Gmail, out it much more bluntly in a tweet, claiming that Google was “only a year or two away from total disruption”, noting that even if Google makes Bard work as well or better than ChatGPT, “they can’t fully deploy it without destroying the most valuable part of their business!”

Maybe it shouldn’t be a surprise, as it’s hard to consider a company with a turnover in the billions and with more than 100,000 employees as anything but an incumbent. But the magic of Google has been that until now, they kept a decent momentum, as well as a considerable amount of start up vibe to their public brand. But now, they’re very much at risk of becoming a cautionary tale about what happens with innovative companies stop innovating. Is this a new Kodak moment?