Microsoft has already made its passionate interest in artificial intelligence very clear recently. Their repeated investments in OpenAI were already a clear sign of this, and it was not necessary to be the brightest to understand even then that the plans of those from Redmond aimed very directly at the use of the technology. the listed company in its wide catalog of products and services. It was so logical that otherwise it would be incomprehensible.

Since the last big investment, it didn’t take long for us to see its first results. The most famous was undoubtedly the arrival of artificial intelligence on Bing, a very quick and intelligent move that made Microsoft’s search engine a real threat to Google for the first time in its history. Just yesterday we told you that Samsung was considering adopting Bing as the default search engine for its devices, something that, for obvious reasons, didn’t work out with Alphabet.

We’ve also learned about Redmond’s plans to bring AI to Microsoft 365 with Copilot, though the rollout is much slower in this case. And this is just the beginning, because we can bet that within the next few months we will see many more movements along these lineseven more so given the great response that the first implementations of this technology received in his catalog.

Artificial intelligence is, this is an important point, quite demanding in terms of the infrastructure responsible for providing such services. NVIDIA has been able to position itself as the reference hardware manufacturer for this market, but we can expect new alternatives to emerge in this regard at some point. What we didn’t expect, although it actually makes a lot of sense, is that Microsoft was one of them, and that their plans also come from afar.

As we can read in The Information (behind the paywall), Microsoft is working on its own hardware for artificial intelligence. These plans have the internal name Project Athena and, as I indicated earlier, they did not respond to this year’s AI boom, since they have been working on it since 2019. According to two sources cited by the mentioned media, the project currently has 300 workers.

The current team is working with OpenAI people, which may mean they already have some kind of working prototype, although it is also possible that they are working in the opposite direction, i.e. optimizing the silicon design for solutions proposed by OpenAI and Microsoft. engineers. Be that as it may, resources they list 2024 as a possible year of their arrivalwhich we are clear will occur, at least primarily, in the company’s data centers.