Search, Artificial Intelligence, and dancing with Microsoft is the topic that was covered in an exclusive interview

GEM, Google, and Sundar: Messengers of Chaos and Untrusted AI in Search and AI in Google & Alphabet

It was a panic at Microsoft when they added the technology to their search engine Bing just a few months later, along with the reappearance of the ingenious and flawedchatchat from Openai. ChatGPT proved wildly popular with users, demonstrating new ways to serve up information that threatened Google’s vice grip on the search business and its reputation as the leader in AI.

And yet, unfettered AI language models are also silver-tongued agents of chaos. They will gladly fabricate facts, express unpleasant biases, and say unpleasant or disturbing things with the right prompting. Microsoft was forced to limit the capabilities of Bing chat shortly after launch to avoid such embarrassing misbehavior, in part because its bot divulged its secret codename—Sydney—and accused a New York Times columnist of not loving his spouse.

Google worked hard to tone down the chaotic streak of text-generation technology as it prepared the experimental search feature announced yesterday that responds to search queries with chat-style answers synthesizing information from across the web.

Google’s smarter version of search is impressively narrow-minded, refusing to use the first person or talk about its thoughts or feelings. It completely avoids topics that might be considered risky, refusing to dispense medical advice or offer answers on potentially controversial topics such as US politics.

In March of this year, some big names in research signed a letter calling for a six-month pause on machine learning systems creating more power than GPT-4. Pichai was not a signatory and said in his keynote speech yesterday that the company is currently training a new, more powerful language model called Gemini.

A source at Google tells me this new system will incorporate a range of recent advances from different large language models and may eclipse GPT-4. But don’t expect to get to experience the full power or charisma Gemini can offer. It might seem like another clever autocomplete if the same chaos-taming methods seen in its experiment are applied by Google.

Sundar is already doing that. He just reorganized Google and Alphabet’s AI teams, moving a company called DeepMind inside Google and merging it with the Google Brain AI group to form a new unit called Google DeepMind. Y’all know I can’t resist an org chart question, so we talked about why he made that decision and how he executed it.

It can be pretty hit or miss when it comes to web search right now. There is a lot of strange content on the internet, and the idea of using artificial intelligence to answer questions in a more natural way would remaking the web.

This is a lot of fun. We talked a lot, but we didn’t get any information regarding the plans of Artificial Intelligence or what’s happening with the rest of the world. Maybe next time.

At the time, I was at the launch of Bing. I saw Satya Nadella there. And I’m sure you know this, but

he said

, “I have a lot of respect for Sundar and his team, but I want Google to dance.” He said that he wanted people to know that Microsoft made them dance. I want to know how you felt when he said that. And two, do you think you danced? Are you dancing?

For me, maybe I’ll say it this way. We started working on this new Search Generative Experience last year. To me, it’s important in these moments to separate the signal from the noise. There is a new way to make search better and a way for us to make our user experience better, but we have to get it right. And to me, that’s the North Star. That is the signal. The rest is noise to me. I think it was important to get it right and we have been focused on doing that.

There’s another challenge inside of this, right? If you believe it’s a platform shift, this might be the first platform shift that regulators understand because it’s very obvious what kind of labor will be displaced. Lawyers, mostly, is what I gather, right? They can see that a bunch of white-collar labor will go away, like an email regarding a transaction, every floor of that group can be reduced. They seem focused on the risk. And then there’s the general AI risk that we all talk about.

When Google first did Search, it was an underdog, right? It was involved in a lot of court cases that built the internet, such as the Google Books case and Perfect 10 case. As an underdog, it was delivering a ton of value. Now, you’re at the White House having an AI summit. I’m confident you’re going to end up in government capitals around the world talking about AI. Do you think you are in a different position than the one that started the internet? You’re the incumbent. Are you playing a different role?

Two parts to the question. I believe we can on the first part. For 20 years of tech automation, people have predicted all kinds of jobs would go away. Movie theaters were supposed to end and–

Unemployment over the last 20 years of tech automation hasn’t fully… Twenty years ago, when people exactly predicted what tech automation would do, there are very specific pronouncements of entire job categories which would go away. That didn’t fully play out. So I think there’s a chance that AI may actually… Because I think the legal profession is a lot more than… There’s a chance you know more about being a lawyer. Which is why I can’t opine on it because I don’t know a lot about it. But something tells me more people may become lawyers because the underlying reasons why law exists and legal systems exist aren’t going to go away because those are humanity’s problems. And so, AI will make the profession better in certain ways, might have some unintended consequences, but I’m willing to almost bet 10 years from now, maybe there are more lawyers.

I don’t think the answers to each of these are obvious to me. I think it’s not clear to me you hold back AI in a straightforward way, that’s not the right answer. It has geopolitical implications. So it’s, again, a complex thing we will grapple with over time. I think, from our standpoint, we are a bigger company, so I do think we will come to it in a more responsible way. We will try to find the right answers in some places. It’s possible our approach will be different as we go through it.