Meta is reorganizing its AI teams as it joins the growing ranks of companies trying to create artificial general intelligence, or AGI. Mark Zuckerberg, who has been increasingly focused on the company’s AI research, said the change would help the company “accelerate” its research and, eventually, improve the metaverse.

Meta currently has two teams pursuing AI research: the Fundamental AI Research (FAIR) team, started in 2013, and a team solely focused on generative AI experiences for users of its apps. With the change, Zuckerberg said, the company would bring the two “closer together” as it looks to expand both groups. Meta’s CEO didn’t say how many workers it might add to its AI efforts, but the expansion is notable considering the company has shed more than 20,000 jobs since 2022.

In a video posted to Threads, Zuckerberg said the changes would “support our long-term goals of building general intelligence, open sourcing it responsibly, and making it available and useful to everyone in all of our daily lives.” The change is also the latest way that Zuckerberg is trying to position Meta as a leading AI company in an increasingly crowded field of buzzy generative AI companies and projects.

Creating AGI, a type of AI often compared to human-level intelligence, has become a particular fascination for many of these companies, including Elon Musk’s x.ai, OpenAI and Google. Now, Zuckerberg is throwing Meta’s vast resources at the effort. “We're building a massive amount of infrastructure,” Zuckerberg wrote. “At the end of this year, we'll have ~350k Nvidia H100s — and overall ~600k H100s H100 equivalents of compute if you include other GPUs.”

At the same time, Zuckerberg made it clear he has no plans on giving up on the metaverse, which he says will also benefit from AI advancements. “The two major parts of our vision — AI and the metaverse — are connected,” he wrote.

Linking the company’s AI work to the metaverse isn’t a new strategy for Zuckerberg, who spent much of last year’s Connect event hyping generative AI. Still, it could be a risky one. Zuckerberg’s metaverse is still not widely understood, or especially popular. But Zuckerberg seems to think that may change as the company improves its smart glasses and the AI embedded in them. “By the end of the decade, I think lots of people will talk to AIs frequently throughout the day using smart glasses like what we're building with Ray Ban Meta,” he said.

Meta is reorganizing its AI teams as it joins the growing ranks of companies trying to create artificial general intelligence, or AGI. Mark Zuckerberg, who has been increasingly focused on the company’s AI research, said the change would help the company “accelerate” its research and, eventually, improve the metaverse.

Meta currently has two teams pursuing AI research: the Fundamental AI Research (FAIR) team, started in 2013, and a team solely focused on generative AI experiences for users of its apps. With the change, Zuckerberg said, the company would bring the two “closer together” as it looks to expand both groups. Meta’s CEO didn’t say how many workers it might add to its AI efforts, but the expansion is notable considering the company has shed more than 20,000 jobs since 2022.

In a video posted to Threads, Zuckerberg said the changes would “support our long-term goals of building general intelligence, open sourcing it responsibly, and making it available and useful to everyone in all of our daily lives.” The change is also the latest way that Zuckerberg is trying to position Meta as a leading AI company in an increasingly crowded field of buzzy generative AI companies and projects.

Creating AGI, a type of AI often compared to human-level intelligence, has become a particular fascination for many of these companies, including Elon Musk’s x.ai, OpenAI and Google. Now, Zuckerberg is throwing Meta’s vast resources at the effort. “We're building a massive amount of infrastructure,” Zuckerberg wrote. “At the end of this year, we'll have ~350k Nvidia H100s — and overall ~600k H100s H100 equivalents of compute if you include other GPUs.”

At the same time, Zuckerberg made it clear he has no plans on giving up on the metaverse, which he says will also benefit from AI advancements. “The two major parts of our vision — AI and the metaverse — are connected,” he wrote.

Linking the company’s AI work to the metaverse isn’t a new strategy for Zuckerberg, who spent much of last year’s Connect event hyping generative AI. Still, it could be a risky one. Zuckerberg’s metaverse is still not widely understood, or especially popular. But Zuckerberg seems to think that may change as the company improves its smart glasses and the AI embedded in them. “By the end of the decade, I think lots of people will talk to AIs frequently throughout the day using smart glasses like what we're building with Ray Ban Meta,” he said.

TikTok has shared more about its plans to fight misinformation ahead of the 2024 presidential election. And, like many of its social media peers, the company’s 2024 efforts will closely resemble what it’s done in the past.

To start, the company is introducing its in-app election guide, which will point users to information about voting, including how to register and how to vote by mail. While the election guide will look similar to years past, the company is introducing it earlier than it has in past elections (its election resources for 2020 and 2022 debuted much closer to their respective contests than this year’s, which comes just as the presidential primaries are getting underway).

TikTok’s approach to misinformation in 2024 hasn’t changed much since 2022. The company will continue to bar political ads, including those that come via creators’ branded content. It will also keep working with fact-checking organizations, which help determine whether content is eligible for recommendations or warrants an “unverified” label. The company says it will “expand media literacy resources to these labels” later this year, which could make the notices more useful to users who encounter them, though it’s unclear what that will look like just yet.

TikTok’s plan for 2024 comes as experts warn that the rise of generative AI could fuel a new wave of viral election misinformation. On its part, TikTok acknowledges that generative AI has created “new challenges” for the social media industry, but stopped short of offering any new policies to address it. Instead, the company says it will keep enforcing its existing rules, which prohibit “misleading” AI-generated content and require creators to disclose when videos use “realistic” AI-generated content.

“As the technology evolves in 2024, we'll continue to improve our policies and detection while partnering with experts on media literacy content that helps our community navigate AI responsibly,” the company wrote in a blog post.

When it comes to disinformation and coordinated efforts to manipulate its platform, TikTok is making a notable change. The company says it plans to release detailed reports on covert influence campaigns ahead of the 2024 contest It already discloses some information about influence campaigns it uncovers in its quarterly transparency reports, but the disclosures are typically short on details. That should change this year, according to TikTok, which says it will begin releasing “dedicated covert influence operations reports” later this year.

TikTok recently pulled a tool that allowed researchers and others to study the popularity of hashtags on its app. The change, first reported by The New York Times, came shortly after researchers published a report using data from the tool that criticized the company.

As The New York Times points out, the tool was one of the few publicly-accessible methods of tracking details about the popularity of specific hashtags. TikTok, like other social media companies, has made it difficult for outsiders to track how content spreads in its app.

The tool in question is a feature called Creative Center, which provides data about the popularity of hashtags to would-be advertisers and others. Researchers at Rutgers’ Network Contagion Institute had used Creative Center’s search function to track hashtags deemed “sensitive” to Chinese government interests. The researchers compared the prevalence of the hashtags between TikTok and Instagram and concluded that many "sensitive" topics were "dramatically underrepresented on TikTok" compared with Instagram.

Soon after the report was published, the researchers said the search feature in Creative Center disappeared without an explanation. “Search capacity for Hashtags has itself now been removed from the user interface entirely, which NCRI discovered to have occurred on Christmas day, days after this report’s initial release,” they wrote in an addendum to the report. They added that TikTok had also disabled direct access to a number of “sensitive” topics they had previously tracked, including hashtags related to US politics and other geopolitical issues.

In a statement to The New York Times, TikTok confirmed the change. “Unfortunately, some individuals and organizations have misused the Center’s search function to draw inaccurate conclusions, so we are changing some of the features to ensure it is used for its intended purpose,” a company spokesperson said.

The dust-up is the latest example of mounting tensions between social media companies and researchers trying to study thorny topics like misinformation. Meta has also found itself at odds with researchers, and reportedly plans to deprecate CrowdTangle, a tool widely used by researchers and journalists to study how content spreads on Facebook. X has also greatly restricted researchers’ access to data since Elon Musk took control of the company, making its once open APIs prohibitively expensive to most groups.

In TikTok’s case, the company may be particularly sensitive to what it considers improper use of its tools. The company has for years denied that it aligns its content policies with the interests of the Chinese government as numerous government officials have called for the app to be banned. More recently, the company faced increased scrutiny over its handling of content related to the Israel-Hamas war — criticism that was also fueled by what the company said was an inaccurate portrayal of hashtag data.

That said, the company has made some concessions to researchers. TikTok began offering an official Research API to some academic institutions last year, and reportedly plans to make the tools available to some civil society groups that have questioned the company’s content moderation practices.

But for researchers, the move to abruptly cut off a tool will likely fuel more questions about just how willing the company is to work with them. “This lack of transparency is of deep concern to researchers,” the NCRI researchers wrote.