What I learned talking to Google about chatbots

[ad_1]

Those who attended the World Economic Forum earlier this month had a slew of massive issues to discuss: the energy crisis, the war in Ukraine, Sino-American tensions and inflation, to name a few. But the topic causing a stir on the dining tables was something else: ChatGPT, the AI ​​platform recently launched by OpenAI, a small US startup, that seems to talk and create content (almost) like a human.

“It’s all anyone wants to talk about,” one CEO lamented, while attendees in Davos shared stories about their own experiences with ChatGPT in the office, and worried about whether their kids were using it to do their homework.

Magic arises in part because its powerful abilities make it a “game-changer that society and industry need to prepare for,” says the World Economic Forum. But it also points to an intriguing question: Could this so-called artificial intelligence, which answers questions, undermine the dominance of Google’s powerful search platform? Is the Silicon Valley giant, notorious for disruption, about to crash itself?

Until recently, it was hard to imagine anything that would challenge Google’s dominance in search. In the quarter century since the instrument was first created, it has so ingrained itself in most of our lives as well as our language, even becoming a verb in its own right. Google accounts for 85 percent of the market, with 99,000 Google searches performed every second (a mind-boggling 8.5 billion searches every day). Besides, rival platforms like Microsoft Bing or Yahoo are fighting over scraps.

The numbers seem to put Google in an unassailable position. Regulators have periodically worried about its market dominance. They also explain why Alphabet, Google’s parent company, is now worth $1.3 trillion. But, like David and Goliath, sometimes a little boy can surprise a giant by attacking in a new way. And some Davos fans believe that ChatGPT can do just that.

Some at Google think it makes sense for a tinkerer like ChatGPT to serve as a lightning rod for early public concern about generative AI.

It is certainly not hard to imagine that a smart tool powered by AI technology could provide a more convenient way to search for information in the future. Especially since Microsoft just acquired a $10 billion stake in OpenAI, and plans to integrate the system into its products in order to “completely transform it,” says CEO Satya Nadella. What is remarkable about this innovation is that it did not come from a tech giant at all. On the contrary, Google has so far failed to bring anything comparable to the market, despite having historically been proud of its AI capabilities.

In search of answers, I popped this question with some senior Google executives, who insisted that the issue revolved around ethical questions. Alphabet leadership claims the company has developed AI platforms more powerful than ChatGPT but has decided it can’t launch until it finds a way to control potential social and ethical risks. “Our goal is to be bold and responsible,” said James Manica, Google’s head of technology and society. “We think Google is very competitive in this area.”

A different executive added that “being a first mover isn’t always a good thing.” In fact, some at Google think it makes sense for a renderer like ChatGPT to act as a lightning rod for early public concern about generative AI, because it would make it easier for big companies to move into the space later.

Maybe. The history of technology is full of stories of “second mover” companies that imitated an innovation launched by someone else, then outdid their pioneer. MySpace and Facebook are a case in point. However, some executives at Google’s competitors — such as Microsoft and Amazon — have offered another explanation: They believe the company is haunted by an incumbency problem, as described by US business consultant Clayton Christensen in his classic 1997 book The Innovator’s Dilemma. Perhaps they suggest it’s because Google Search is so dominant that employees have no incentive to flock to develop a technology that might undermine its former success.

There’s another topic I’ve written a book about: silos. OpenAI has one brainstorming group developing its innovation. However, Alphabet has one team within Google that works on AI, and another at DeepMind AI, which it acquired.

On paper, this sounds like twice the talent. But history is also full of companies that failed to innovate because different departments competed instead of cooperating, ultimately undermining the field of innovation. Sony has had this problem before. So did Microsoft.

Alphabet is formidable enough to overcome all of these issues. But the excitement around ChatGPT shows why innovators, especially incumbents, can’t relax.

Follow Gillian on Twitter @gilliantett and email her at [email protected]

Follow @FTMag on Twitter to catch our latest stories first

[ad_2]