Meta recognized that he joined the world of AI late

Meta acknowledges that it was late in developing its own artificial intelligence (AI) infrastructure, but ensures that it has not only caught up but will launch new products with this technology by the end of the year, as well as the next generation of its viewer. of virtual reality, the base of the metaverse in which he continues to invest and work.

Meta increased funding to develop generative artificial intelligence initiatives, as he confirmed in the presentation of the financial results for the first quarter of 2023.

CEO Mark Zuckerberg discussed AI on the investor call, explaining that they are currently focused on two areas: using AI for recommendation systems and ratings for services like Facebook and Instagram. , and the development of new models of generative foundations.

These new models are at the base of “new classes of products and experiences”, one of them being AI agents, which they hope to present to billions of people “in ways that are useful and meaningful”, for example, in the messaging with business and customer service.

Meta has its own large language model, which it has called LLaMa, with 65 billion parameters, which is already available to researchers. They have also provided three open source visual models and, along with their training data, the DinoV2 visual computing model and a tool to animate drawings.

The main investment, however, has been in AI infrastructure and capital expenditures. In this regard, the manager acknowledged that they started late with the development of their own structure, but pointed out that they are no longer behind schedule and that, even now, they have “the ability to do leading work in this space at scale“.

Source: dpa

(Reference image source: file)

Visit our news channel on Google News and follow us to get accurate, interesting information and stay up to date with everything. You can also see our daily content on Twitter and Instagram