Why Google Gemini seems poised to win the AI ​​race over OpenAI

If you want to win in AI — and I mean win in the biggest, most rewarding, most lucrative ways that shape the world in your image — you have to do a bunch of hard things simultaneously. You must have a model that is undoubtedly one of the best available in the market. You need the almost unlimited resources needed to continue improving this mod and spread it widely. You need at least one AI-based product that a lot of people use, and ideally more than one. It needs access to as much of users’ other data — their personal information, their online activity, even the files on their computers — as possible.

Each of these elements is complex and competitive; There’s a reason why OpenAI CEO Sam Altman keeps ranting about his needs Trillions of dollars In the account alone. But Google is the only company that seems to have all the elements sorted already. Over the past year, and even in the past few days, the company has taken steps that indicate it is poised to be the largest and most influential force in artificial intelligence.

Much of the necessary infrastructure work has been carried out in the past year. Google in November Gemini chest 3, Which is widely considered to be the best overall large language model on the market. Wins in most (somewhat doubtful) Standardized testsMost experts agree that it is either At or near the top of the list For most tasks. Its reign won’t last forever, of course — we’re still in the “there’s a new best model every six weeks” phase of AI — but Google has proven that its best work is always the best work in the industry.

An important factor in the Gemini 3 was the way it was trained: using Google’s TPUs, a highly specialized chip manufactured by the company. Building for years Exactly for this kind of purpose. Sure, Google is vulnerable to some of the manufacturing issues and high RAM prices that everyone else is, but unlike almost all of its competitors, it doesn’t rely on Nvidia’s supply chain. Google can improve its entire system to make it better, faster and cheaper. No one has this kind of complete control over the fate of artificial intelligence.

So what do you do when you have the technology in place? Put it in front of people and turn it on. on monday, Google and Apple announced That Gemini will power the next generation of Siri coming later this year. This is a big win for Apple, which is… He reportedly pays $1 billion annually Hoping to turn Siri into an AI assistant that’s actually useful for a change.

For Google, it’s no less important. Obviously, Apple saying “this is the best technology out there” is a strong signal for the market, but more than that, Siri instantly became one of the most popular ways people interact with Gemini. Apple Craig Federighi He said in 2024 Siri processes “nearly 1.5 billion requests per day,” and while we don’t know the exact details of the new deal, it’s assumed that a large percentage of those requests will soon be run through Gemini. (Here’s hoping “setting a timer,” the one thing Siri continues to do well, doesn’t get a new, more complex back end.) Compare that with ChatGPT, which Altman said last year was getting 2.5 billion claims daily. Gemini is growing quickly but is still lagging behind ChatGPT, but adding Siri to the mix will help Google catch up more quickly.

The technology deal isn’t the same as Gemini usurping Siri entirely, of course, and Google would certainly also like to ask Siri questions to Gemini the way it currently does with ChatGPT. But the deal is still important because every user matters: the more user activity and data these companies can collect, the better their models and products will become. The recent research experiment was in part about… This particular flywheelThis absolutely applies to artificial intelligence.

Another Google announcement this week promises more flexibility. Announced a subscription feature It’s called “interpersonal intelligence.”“, which connects Gemini to the vast ocean of information Google has about you in order to give you better answers. Every time you ask a question, Gemini can now answer it by looking at your recent searches, the videos you’re watching on YouTube, your emails, photos, files, your Chrome history, and more. You can’t really overstate how big of a deal this is: Google no longer has to ask you to give it a lot of context, and it hopefully makes excellent, detailed prompts in every Once, or create complex custom instructions Google already knows a scary amount about you, and now Gemini does too.

Currently, Personal Intelligence is in beta for a subset of paying AI customers. Eventually, Google plans to make it available to everyone, everywhere. It plans to bring it to Google’s most important product of all, the most popular web page on the planet: its search engine. So far, AI in search is still just a tab next to general search results, but Google sees it very clearly as the future of search. And you want to turn Gemini into a gateway to all Google data about you, the Internet, and the world.

In 2022, when ChatGPT was launched, it was clear that Google had faltered. But credit is due: For a company not exactly known for its ability to focus on a cohesive product strategy, Google has managed to marshal its significant resources in one direction. Now, if chatbots are in fact the future — and most of the AI ​​industry is still betting that they are — then there is simply no other company currently set up to truly compete with Google. Google has forms. It has the resources to improve it. It now has the distribution needed to get people to use its bots, and the data required to make them uniquely personal and useful. At least for now, ChatGPT has brand strength and daily active users. But Google has almost everything else. Even iPhone.