At a time when Openai and its Chatgpt model hogs all the limelight, it is easy to forget that Google AI has OG. The company has been on it for more than a decade, with its AI efforts, has been operated on a large scale deepmind researchers. In fact, the current AI revolution and successes are the direct result of a paper – and which has now achieved a mythical reputation – which is called meditation that you all need. The paper was written by Google researchers. And yet, when we think about AI, we think about Openai and Chatgpt. With Google Gemini 2.5 Pro, the latest model of tech giants, it is changing.

Openai has made all the limelight hog that it has been better in converting AI research into real products. While Google has been a lab leader, Openai has been better in bringing its AI tool in public. Google, in a way, was caught off-guards by Chatgpt 3 in 2022. The company scrambled to respond, and also ended by making some mistakes in a beloved. Its initial challenges for GPT 3 were not good at all.

In 2025, to use a very-clich phrase, the empire is striking back. And it is doing this with Gemini 2.5 Pro (experimental).

Since it came to the scene a few weeks ago, Gemini 2.5 Pro (Experiment) has worn its users. This is Google’s first AI system that is better in many examples compared to the best tools available from Openai. And whether in real use, or in benchmarks, Gemini 2.5 Pro has excelled. Take a look at the benchmark:

Advertisement

But more new Gemini than benchmark has affected users. Social media – mostly X aka Twitter because this is the place where action is – nowadays people are filled with thumbs, stating how Gemini 2.5 Pro coding, or writing, or analyzing documents. Unlike the recent Chatgpt 4o, which turned their users into a parrot and to flatter their users, Gemini has a measured tone and a more official personality. It does not unexpeate the sound in any way. Instead, it will be a professional assistant in a way.

When I try an AI tool such as Gemini, Deepsek or Chatgate, I usually ask it to write a paragraph or two in the style of John Fosa. Now, Fosse is a writer – and here I am talking about their work translated into English – which uses a minimal language and a rare voice, but the way the Norwegian writer attains his strength makes the rhythm with repetition. For AI, nowadays coding is easy. It excels on logic. But writing a few lines of imagination is still a difficult task for it. And on that, it is very difficult to write a paragraph for the AI ​​tool that reflects the style used by Fosse. They can write in Hemingway style – Ok, like – but not foster. In my experience, the only AI tool that has managed any kind of consistent in it has been the Gemini 2.5 Pro.

Another reason is that Gemini 2.5 Pro is such an important AI device. Unlike the best tools from the choice of Openai and Anthropic, Gemini 2.5 Pro is free to use. In this aspect it is somewhat like Deepsek R1. But unlike Dipsek R1, which was often getting a hammer on its servers due to load, Gemini 2.5 Pro may probably handle millions of users simultaneously. Power powering google services is the best if any technology company does not have the best. In many ways, this Gemini 2.5 Pro first creates the top-class AI model that most people in the world can use.

Now Gemini is as good as 2.5 Pro, there is another aspect for it, and it is how Google vs Openai is going to look in future. Till now it seemed that openi was getting into all blows. But I suspect that Gemini 2.5 Pro is just the beginning of a counter strike. Google has some best AI researchers. It also has the kind of data that other companies can only dream. And finally it is a huge scale – in everything – which is not currently in Openai.

Google still needs people to get better in showing their AI tools. For example, its notebook, which can turn a paragraph into a long podcast, is luxurious. Similarly, there are many other Gemini and Google AI devices that they do that are the top classes. But they are scattered in Google services and products, and in many cases they need to be accessed – such as literally – an engineering degree because Google has placed them in virtual silos that are accessible to developers only.

But I think these are the swallows that a company like Google will solve in the coming months. The new Gemini interface in itself is a major improvement in terms of access and user experience compared to whatever was a year ago. A more pressure case for Google was to prove that it could match chat and cloud. This is done with Gemini 2.5 Pro.