Meta’s Developing its Own LLM to Compete with OpenAI

Meta’s trying to make an even bigger push into generative AI, by way of a brand new LLM system that it hopes will make it the chief within the AI race.

In response to The Wall Street Journal, Meta’s presently constructing a brand new AI system, which it hopes will finally rival OpenAI’s GPT techniques in compute energy and complexity. The system will construct on Meta’s present Llama models, and convey Meta up to the mark with others within the generative AI race.

As per WSJ:

“The planned system, details of which could still change, would help other companies to build services that produce sophisticated text, analysis and other output. It is the work of a group formed early this year by Meta Chief Executive Mark Zuckerberg to accelerate development of so-called generative AI tools that can produce humanlike expressions.”

Certainly, numerous studies have prompt that Zuckerberg turned “obsessed” with AI after seeing the early hype round ChatGPT, which has modified the path of lots of Meta’s tasks, and decreased its emphasis on the metaverse, its longer-term VR wager.

Although that additionally stays a key factor in Meta’s planning. However generative AI has opened up new alternatives, and Meta is uniquely positioned to faucet into such, based mostly on its unmatched databanks, and systematic assets.

On the latter, WSJ additionally studies that Meta’s presently within the strategy of buying extra Nvidia H100 chips to energy its AI ambitions. Nvidia chips are in high demand, and limited supply at present, which, at a base stage, signifies that the corporate with essentially the most chips stands the most effective probability to win out within the broader AI improvement race.

Earlier this yr, Elon Musk reportedly purchased 10,000 GPUs for his generative AI venture.

It’ll be fascinating to see how Meta plans to deploy its personal AI and LLM instruments, and the way they’ll be built-in into its in-app experiences. Meta’s already experimenting with a spread of AI-powered components, together with visual creation tools and an Instagram chatbot.

Perhaps, by a extra superior AI system, it’ll look to boost and enhance these experiences, whereas additionally using higher-end AI inside its advert focusing on and creation course of.

Total, it’s little shock to see Meta trying to make an even bigger transfer within the AI area, given its long-term work on this space, and the potential it has to compete, based mostly on its trove of conversational knowledge and interplay. With these issues factored in, it does seem to be Meta ought to be capable to create a greater AI mannequin, and it’ll be fascinating to see how, precisely, it seems to deploy and make the most of such techniques.

In response to WSJ, Meta expects to start out coaching its new AI system in early 2024. No phrase but on a public launch timeline.