Meta has begun dismantling its $2 billion acquisition of Manus, completing an operational separation from the Chinese-founded AI startup and halting data sharing between the two companies. This is the most concrete step yet toward complying with a divestiture order Beijing issued roughly two months ago on national security grounds.

Meta has cut Manus off from its internal systems, Bloomberg reported, preventing employees from using Manus tools for internal projects as the two companies move toward a full separation.

Meanwhile, according to May reports, the co-founders of Manus have held preliminary discussions about raising approximately $1 billion from outside investors to reclaim the startup from Meta, a move that could pave the way for a Chinese joint venture structure and an eventual listing in Hong Kong, a venue that has seen a surge in AI listings this year for Chinese AI startups like MiniMax and Zhipu.

What was supposed to be a landmark exit for Chinese AI is quickly unraveling. The move underscores Beijing’s determination to retain control over strategically sensitive technology, regardless of a company’s offshore incorporation.

In addition to the forced divestiture, Chinese authorities have since expanded travel restrictions to researchers and executives at private firms, requiring government approval before heading abroad. China is also tightening its grip on foreign capital, with reports indicating that top AI firms, including Moonshot AI, StepFun, and ByteDance, will need government sign-off before accepting U.S. investment, adding another layer to Beijing’s sweeping effort to control its AI sector.

Even as Meta moves to sever ties with Manus, the agentic AI startup has continued to ship new features, rolling out integrations with Similarweb and Shopify.

Manus drew widespread attention with a viral agent demo relocated its staff to Singapore in mid-2025 before announcing a $2 billion acquisition by Meta in December. Chinese regulators moved to scrutinize the transaction earlier this year, citing potential violations of technology export controls and foreign investment rules.

Manus investors, including California-based venture firm Benchmark, have already received their proceeds from the acquisition, while Asian backers, including Tencent, HSG, and ZhenFund, have indicated they will cooperate with the unwinding process, according to the WSJ.

Manus’ Chinese origins with parent company Butterfly Effect drew scrutiny on both sides of the Pacific, with Senator John Cornyn questioning whether American capital should flow to a Chinese-linked firm.

Meta and Manus did not immediately respond to a request for comment outside regular business hours.

Meta is adding an AI assistant and a desktop version to Edits, its video-editing app built to compete with ByteDance’s CapCut. The company previewed the features at an invite-only creator event in Los Angeles on Wednesday. The AI assistant is currently in testing with event attendees. The desktop version is “coming soon.”

The AI assistant uses a creator’s Instagram data, including views and video-retention metrics, to analyse what is working and suggest new content ideas. It can recommend topics based on performance trends and flag trending audio. The goal is to keep creators inside Meta’s ecosystem instead of turning to external tools like ChatGPT for brainstorming or analytics.

Meta launched a similar AI assistant for Facebook creators last week. YouTube Studio already offers an AI-powered Inspiration tab, and TikTok has its own creative assistant. The feature is table stakes in the platform war for creators, but integrating it directly into the editing app, rather than the social platform, is a deliberate move to make Edits stickier.

The desktop version addresses one of the biggest gaps between Edits and CapCut. CapCut has offered desktop editing since launch, giving creators precise control on a larger screen for advanced workflows. Meta says creators will be able to sync projects seamlessly between mobile and desktop.

Several features launched today alongside the announcements. A Beta tab gives creators early access to experimental features still in development. Expanded audience insights now include demographic breakdowns and peak engagement times. Creators can also search specific topics in the app’s Inspiration feed to discover trending Reels and templates, and create multiple versions of a single video to A/B test performance before publishing.

Meta did not share download numbers for Edits. But it said content made with the app sees a 10% higher save rate and 2% higher reshare rate than content not made on Edits. More than half of people watching Reels on Instagram see Edits-created content every day.

Edits launched last year as a direct response to the TikTok and CapCut threat. The timing of its arrival coincided with the US ban scare that briefly took both TikTok and CapCut offline. Meta moved fast to fill the gap. The AI assistant and desktop version are the next steps in turning Edits from an opportunistic launch into a permanent platform tool that keeps creators producing content for Instagram rather than its competitors.

Meta is testing a new “Series” feature for Reels that’s designed to make it easier to keep up with serialized content on Instagram and Facebook, the company told TechCrunch.

Select creators will now be able to create a series using their Reels, both new and old. Each Reel will become an episode in a bigger story with a dedicated hub on their profile, mirroring an experience that’s already available on TikTok.

The launch of the new feature indicates that Meta is embracing longer-term viewing, rather than quick scrolling sessions, as is normally the case with short-form videos. With organized series, the social media giant is looking to bring audiences back repeatedly and build stronger viewing habits. As for creators, the new feature offers a more organized way to release episodic videos, like tutorials and challenges.

Meta told TechCrunch that it’s considering ways to monetize the new feature, but didn’t share specifics on what that could look like.

It’s worth noting that Instagram rival TikTok launched a Series feature back in 2023, allowing creators to post collections of premium content behind a paywall that viewers can purchase for access. It’s possible that Meta could follow a similar playbook with its own series feature in the future.

Meta says it’s already seeing serialized content take off, so its newest test feature is meant to make it simpler to catch every episode. For example, if a creator is doing a “10 days of healthier baking” series, they can bundle all ten videos into a single series. The series would appear in a new tab on the creator’s profile with its own page so people can find all the episodes, watch them in order, and pick up where they left off.

When viewers discover episodes of a series while scrolling their feed or the Reels tab, they will see the option to tap into the full series to see other episodes. Users also have the option to save a series to watch later or stay up-to-date.

For the testing phase, Meta is working with select creators and content producers who have already shared serialized content on Instagram and Facebook.

Meta is taking its “13+” content settings for Teen Accounts global. The company said today that the setting, which it likens to a movie rating, will now apply by default to teenagers across Instagram, Facebook, and Messenger worldwide, extending a system it first introduced last October in the US, UK, Australia, and Canada. According to Meta, nine in ten teens have stayed in the default setting since that launch rather than seeking to change it.

The idea is to shape what shows up for a teen by default. On Facebook, the new 13+ setting is designed to hide content Meta deems inappropriate for teens in Feed and Reels, and to limit teens’ interactions with Profiles, Pages, Groups, and Events that mostly post such material.

On Messenger, it restricts teens from opening links to that content or chatting with accounts that primarily share it on Facebook. A stricter option called Limited Content, already available on Instagram, is due to arrive on Facebook and Messenger later this year.

Meta has leaned on parents to calibrate what “appropriate” means. The company says hundreds of thousands of parents have rated more than 15 million pieces of content, and that in its most recent survey at the end of April, covering parents in the US, UK, Australia, and Canada reviewing recommended Facebook posts, fewer than 2% were judged inappropriate for teens by most parents.

The framing draws on the Motion Picture Association’s public guidelines, though Meta is careful to note that the MPA is not rating or endorsing any of its content.

Alongside the rollout, Meta said it is testing a way to stop teens being served the same kind of post over and over on Instagram. Content about nutrition, weightlifting, or coping with anxiety can be useful, the company said, but is better balanced than shown repeatedly, so it is trialling limits on how many such posts appear in a row in Explore, Feed, and Reels.

It is a quiet acknowledgement that the problem with a recommendation engine is not only what it surfaces once but what it surfaces relentlessly.

To check its own work, Meta commissioned Alice, a safety firm formerly known as ActiveFence, to run adversarial testing against the settings. By Alice’s account, Instagram Teen Accounts in the default mode saw 68% less mature content than an unnamed competitor’s teen experience, rising to 96% less in the stricter Limited Content mode.

The assessment flagged two gaps, including that “car surfing,” a newer viral trend, was not yet restricted the way “subway surfing” already was. Meta said it updated its policies and that the fixes were retested before publication.

The expansion lands against years of pressure on Meta over how its apps affect young users, from the shelving of an “Instagram Kids” project to a steady drip of teen-safety features. The figures and the favourable audit are Meta’s own, and the company frames the announcement as progress. Whether parents, regulators, and the teens themselves agree is the part no press release decides.

Instagram has resolved a security issue that allowed several users’ accounts to get hacked. The attack appeared to rely on tricking Meta’s own AI-powered support chatbot into granting access to a victim’s account.

Over the weekend, several users on Reddit claimed that their Instagram accounts had been compromised, and a number of users on X warned of similar account hijackings. The compromised accounts include the Instagram handle for the Obama-era White House, which appears to have been inactive since 2017; and the account of the U.S. Space Force’s chief master sergeant John Bentivegna.

Security researcher Jane Wong said her Instagram account was also taken over. 

“The password got changed without my knowledge and I was getting different password reset attempts throughout yesterday,” said Wong. “Quite concerning.” 

A video posted on X showed the step-by-step process to hack someone’s Instagram account. The hacker allegedly used a VPN to spoof the targets’ presumed location to avoid triggering Instagram’s automated account protections. Then, the hacker opened a chat with Meta AI Support Assistant and asked the bot to add a new email address to the target’s account. The chatbot can be seen sending a verification code to the email address provided by the hacker; the hacker then shares the verification code with the chatbot, which prompts the chatbot to show a button to “Reset Password.” The hacker enters a new password and takes over the victim’s account. 

TechCrunch was able to verify that the hacker’s public email mailbox, which was displayed in the video, effectively received the verification code. 

The attack relied on the fact that at no point the hacker had to take over the legitimate email address linked to the victims’ Instagram account. 

On Monday, Instagram spokesperson Andy Stone said in a reply to Wong’s post and others that the issue was now fixed. It’s unclear how many Instagram users had their accounts improperly accessed.

Meta did not immediately respond to TechCrunch’s request for comment.