on one call out At a meeting with investors last spring, Meta's chief executive Mark Zuckerberg said he believed they had an opportunity to “introduce artificially intelligent assistants to billions of people that will be useful and meaningful.”
After a year, he is implementing his statement.
On Thursday, Meta will begin rolling out new versions of its AI-powered smart assistant software to its apps, which include Instagram, WhatsApp, Messenger and Facebook. The latest technology will be rolled out in more than a dozen countries, including Australia, Canada, Singapore and the United States.
AI software will become practically ubiquitous – inside news feeds, in search bars, and in chats with friends. People will be able to ask the assistant, Meta AI, for help completing tasks and getting information, such as what concerts might be in San Francisco on Saturday night or the best options for vegan enchiladas in New York.
Meta AI is powered by LLaMA 3, the company's newest and most powerful large language model, an AI technology that can generate prose, carry out conversations, and create images.
“With LLAMA 3, Meta AI will now be the most intelligent freely available assistant,” Mr. Zuckerberg said in an interview. “And because we've reached our desired quality level, we're now going to make it more prominent and easier to use across all of our apps.”
The effort marks Meta's largest rollout of products incorporating powerful AI technology. The social networking giant began incorporating generic AI into its apps in a limited capacity last year, launching a series of AI-powered chatbots and characters that could interact with users in September. But this new initiative goes even further in scope and purpose, placing AI products among the most visible and most used parts of Meta's apps.
Other tech giants are also incorporating AI into their products, as Silicon Valley start-ups raise billions of dollars to build AI-powered apps and services that they believe will usher in the next generation of computing. Will define the phase.
Last year, Microsoft incorporated OpenAI's ChatGPT into the software giant's Bing search engine. Google has integrated AI into products like Docs, Gmail, and Google Search. Start-ups like Perplexity and Anthropic also aim to provide more AI-powered products and services to consumers.
Meta's efforts come from the sheer scale of its products, which are used by approximately four billion people globally every month. It is one of the few companies that is “open source” most of the AI technology it is creating, meaning anyone can look at the underlying technology and use it to create products or services for free. Could.
Mr. Zuckerberg said the new AI rollout was part of Meta's historical “playbook” of adding a feature to its apps “when we felt it was ready.” He pointed to products like Stories and Reels, two video and image products that appeared in Instagram, and how they were later merged into Facebook and WhatsApp.
When ChatGPIT arrived in late 2022, wowing people with the way it answered questions, wrote term papers, and generated computer code, the tech industry raced to create similar technology — albeit on devices Sometimes they made mistakes and created lies.
Due to such flaws, OpenAI and other major AI companies said they would not open source the underlying technology that powers these chatbots. (The New York Times has sued OpenAI and Microsoft, claiming copyright infringement of news content related to AI systems.)
Meta took a different approach. It open sourced the first version of LLAMA in February 2023, before releasing a more powerful version less than six months later. Other companies, including Google and a major French start-up, Mistral, have followed suit. The companies have said that by open sourcing the technology, independent researchers and engineers everywhere can help find and improve problems in the technology.
“We have always believed in this principle and are pleased to see the industry embracing the power of open source and the positive possibilities it creates,” Ahmed Al-Dahle, Meta's vice president of generic AI, said in an interview.
Mr Dahle said LLaMA 3 showed vast improvements over Meta's previous large language models and described it as “much better” than what people are used to.
Meta has also fine-tuned the AI model to make the types of questions answered by Meta AI a little less conservative, meaning the assistant will be less likely to refuse to answer certain questions. In the past, Meta, Microsoft and others aimed to limit their chatbots from discussing third-rail topics such as politics, religion and medical advice for fear of repercussions from political or interest groups.
To attract users, Meta will also add a faster image-generation technology to the AI assistant, and plans to incorporate the AI technology into its Ray-Ban Meta smart glasses later.
The challenge will be to convince people that the new assistants can be useful. Mr. Dahle said Meta is working on helping people learn what kinds of questions they can ask Assistants.
“Despite how prevalent these AIs have become, there is still an education factor on how to interact with AI,” he said.
Like most of Meta's products, the new Assistants are free to use – and may be hard to resist if you're a regular user of the company's apps.
Meta executives don't seem concerned about AI saturation. “We're excited to share our next-generation Assistant with more people and can't wait to see how it improves people's lives,” the company said.