Mark Zuckerberg, CEO and founder of Meta, discussed the evolution of artificial intelligence in a lengthy interview with the Kallaway YouTube channel. In the interview, the creator of Facebook was very critical of the approach of OpenAI, the start-up behind ChatGPT, Google, Microsoft and other players who offer proprietary AIs. Developed and owned by private companies, proprietary AIs are protected. Their source code is not accessible to the public. This is the case for Google’s Gemini, ChatGPT or Anthropic’s Claude.
Also read: Meta reorganizes its metaverse activity to boost Ray-Ban connected glasses
Meta bets on open source
Unlike its rivals, Meta focuses exclusively on building open-source AI models. Llama language models are, in fact, open to all researchers and all developers. They can use the models to build their own chatbots and AI services. As he explains in the interview, Meta strongly believes in the future of open source artificial intelligence rather than proprietary AI.
“I don’t think AI technology is something that should be sort of hoarded and a company can use it to build whatever core, single product they build.”says Mark Zuckerberg
Meta founder laments that tech giants “talk about building real AI”. For Zuckerberg, this approach looks like a futile attempt to “create God”as if the future was going to revolve around a single artificial intelligence.
“It’s almost like they think they’re creating God or something and… it’s just – that’s not what we’re doing. […] It’s not like there’s just one app on people’s phones that people use. There is no one creator that people want all their content from. There is not a single company where people want to buy everything”analyzes Mark Zuckerberg, suggesting a future made up of a multitude of different artificial intelligences.
These different AIs will depend on the needs and interests of the users. Very engaged in the race for AI, Meta has unveiled a range of models made available to research in recent years. Examples include Audiocraft, an AI that is capable of generating music, or CM3Leon, an image generator similar to Dall-E. At the same time, Meta has relied on its models to animate a plethora of AI functionalities integrated into its platforms, such as Meta AI. Added to WhatsApp, Instagram, Facebook and Messenger, the bot stands out as an alternative to ChatGPT, but it is not yet available in Europe.
AI, glasses and the smartphone in your pocket
Furthermore, the billionaire outlined the outlines of how AI could be integrated into his strategy regarding the metaverse and augmented reality. For Mark Zuckerberg, generative AI is one of the technologies that will give meaning to connected glasses.
“I think you can create a great experience with cameras, a microphone, speakers and the ability to do multimodal AI, even before you have any type of display on these glasses”affirms the founder of Meta, specifying nevertheless that the group is working on the creation of AR glasses equipped with screens.
Meta has also added its AI assistant to its connected glasses designed in partnership with Ray-Ban. Restricted to North America for the moment, it is capable of translating text, understanding and describing the world seen through the cameras housed in the mount. These smart glasses do not aim to replace the smartphone. According to Zuckerberg, the rise of AR or AI-powered glasses will not not automatically bury the cell phone. On the other hand, it is likely that you will use it less and less in the future… in the same way that you use your computer less in favor of the smartphone:
“It’s not like we’re going to stop having a phone. It’s just going to sit in your pocket, and you’ll pull it out when you really need to do things with it. But increasingly, I think people are just going to start thinking, ‘Hey, I can take this picture with my glasses. I can ask AI this question, or I can send someone a message.’ It’s just a lot easier with glasses.”.
To not miss any news from 01net, follow us on Google News and WhatsApp.
