Amazon has reportedly joined the growing list of suitors angling to acquire TikTok as a new ban deadline approaches. According to The New York Times, the retailer has made a “last-minute bid” to buy the app, and has approached Vice President JD Vance and Commerce Secretary Howard Lutnick about the offer.

It’s not clear why Amazon is making an offer this late in the process. TikTok is currently facing an April 5 deadline to strike a deal, though President Donald Trump has said he would likely extend it if necessary. The Times reports that White House officials don’t seem to be “seriously” considering the longshot bid. Amazon declined to comment.

Amazon joins several other companies and investors that have proposed bids. The group includes YouTuber MrBeast (with backing from a group that includes Roblox CEO David Baszucki), Reddit cofounder Alexis Ohanian (who joined a group of investors known as “Project Liberty”) and Perplexity AI, which has proposed integrating the shortform video into its search engine.

Even with the deadline just days away, we still don't know exactly when a decision will be made about TikTok’s future. Trump has said he wants to announce a deal before the Saturday deadline.

This article originally appeared on Engadget at https://www.engadget.com/social-media/amazon-reportedly-wants-to-buy-tiktok-now-too-173957102.html?src=rss

©

© picture alliance via Getty Images

Roblox is again updating its parental control features in an effort to give parents more visibility into (and control over) their children’s activities on the platform. The company is updating its in-app “Safety Center” to counter concerns that Roblox, which unlike social media apps allows children under 13, puts its youngest users at risk.

The company added the centralized hub for parental controls last year, in an update that also added new restrictions on younger kids’ ability to exchange private messages with other users. But while that update gave parents the ability to control some of their children’s settings from their own devices, it was missing some key features like blocking.

With the latest update, parents can now view their child’s friend list and block users on their behalf. Once another user has been blocked, they’ll no longer be able to exchange direct messages with the child and kids will need to get parental approval to unblock the person.

Similarly, parents will now be able to block specific experiences within Roblox. Parents have already had the ability to limit in-app content based on its age rating, but parents have still at times complained about inappropriate content surfacing in certain games. Now, parents can block their young kids from participating in certain experiences. (Importantly, Roblox notes that kids will be able to remove apps from the blocked experiences list once they turn 13.)

Finally, Roblox is adding more “granular insights” to the in-app screen time metrics that allow parents to see how their kids are spending their time on the platform. The Safety Center will now list “the 20 experiences in which their child has spent the‬ most time over the last week, sorted by total time.” Parents can then opt to block specific games or experiences directly from the screen time menu if something seems off.

Roblox’s recent push to beef up safety features for younger users come amid a broader reckoning about the effect that online platforms can have on teens. Platforms have tried to address these concerns by enhancing parental controls and dedicated safety features for teens. But lawmakers are also pushing for change, with multiple states passing measures requiring some form of age verification (some of which have since been struck down). There’s also a bipartisan bill in the Senate to ban all preteens from social media entirely.

This article originally appeared on Engadget at https://www.engadget.com/gaming/roblox-parental-controls-now-include-a-block-button-120005897.html?src=rss

©

© Roblox

An Arkansas law requiring social media companies to verify the ages of their users has been struck down by a federal judge who ruled that it was unconstitutional. The decision is a significant victory for the social media companies and digital rights groups that have opposed the law and others like it.

Arkansas became the second state (after Utah) to pass an age verification law for social media in 2023. The Social Media Safety Act required companies to verify the games of users under 18 and get permission from their parents. The law was challenged by NetChoice, a lobbying group representing the tech industry whose membership includes Meta, Snap, X, Reddit and YouTube. NetChoice has also challenged laws restricting social media access in Utah, Texas and California.

In a ruling, Judge Timothy Brooks said that the law, known as Act 689, was overly broad. “Act 689 is a content-based restriction on speech, and it is not targeted to address the harms the State has identified,” Brooks wrote in his decision. “Arkansas takes a hatchet to adults’ and minors’ protected speech alike though the Constitution demands it use a scalpel.” Brooks also highlighted the “unconstitutionally vague” applicability of the law, which seemingly created obligations for some online services, but may have exempted services which had the "predominant or exclusive function [of]... direct messaging" like Snapchat.

“The court confirms what we have been arguing from the start: laws restricting access to protected speech violate the First Amendment,” NetChoice’s Chris Marchese said in a statement. “This ruling protects Americans from having to hand over their IDs or biometric data just to access constitutionally protected speech online.”

It’s not clear if state officials in Arkansas will appeal the ruling. “I respect the court’s decision, and we are evaluating our options,” Arkansas Attorney general Tim Griffin said in a statement.

Even with NetChoice’s latest victory, it seems that age verification laws are unlikely to go away anytime soon. Utah recently passed an age verification requirement for app stores. And a Texas law requiring porn sites to conduct age verification is currently before the Supreme Court.

This article originally appeared on Engadget at https://www.engadget.com/social-media/arkansas-social-media-age-verification-law-blocked-by-federal-judge-194614568.html?src=rss

©

© NurPhoto via Getty Images

TikTok's deadline to sell off or cede its US operations is once again approaching. The 75-day extension delaying the enforcement of a nationwide TikTok ban is set to run out Saturday, April 5, unless the company finally reaches an agreement to find a new owner. Who might buy it, and what happens if no deal is reached? Here's the shape of things right now:

Potential buyers

President Donald Trump has suggested that “a deal” is imminent and could be reached before that deadline. What such an agreement will look like is still extremely unclear. A number of interested buyers have come forward, including YouTuber MrBeast, Reddit cofounder Alexis Ohanian and Perplexity AI. Oracle, which almost bought TikTok in 2020 and was a key part of TikTok’s earlier negotiations to address national security concerns, is also in the running.

Depending on which bid wins out, TikTok could end up looking very different. ByteDance, TikTok’s Chinese owner, won’t allow another company to take control of the app’s coveted recommendation algorithm. That means a new owner would have to rebuild TikTok’s central feature. Perplexity AI, for example, has said it would take on this task of reverse-engineering the app's “For You” feed, making it open source and implementing some kind of Community Notes-style fact checking feature. Reddit cofounder Alexis Ohanian, who joined a group of investors known as “Project Liberty,” has hinted at a vision for TikTok that somehow ties in with blockchain technology.

Option 2

There’s also another, potentially less disruptive option on the table. The White House is also considering a deal that would allow TikTok’s existing US investors to “roll over their stakes into a new independent global TikTok company,” according to The New York Times. Under this arrangement, Oracle could also end up overseeing TikTok’s US data while ByteDance hangs onto its algorithm, as the Financial Times reported in March.

This would likely be a smoother transition than a full-on sale that would require a new owner to rebuild a significant part of the app. Aspects of it are also very similar to terms TikTok already agreed to under a previous agreement with Oracle known as Project Texas. However, this sort of deal may also not be entirely legal, as Politico recently pointed out. The Protecting Americans from Foreign Adversary Controlled Applications Act, the law passed by Congress and signed by former President Joe Biden last year, calls for a divestment of TikTok and its algorithm.

It’s also not clear whether Chinese officials would be willing to sign off on such an arrangement, though Trump has suggested he could use the threat of tariffs as negotiating leverage. “We’re dealing with China also on it because they may have something to do with it. And we’ll see how that goes,” Trump told reporters over the weekend.

No deal

So what happens if Trump’s “deal” isn’t finalized by Saturday? The answer, it seems, is… probably nothing. Trump has said he would likely extend the deadline again if the clock runs out on the current extension. (Once again, it's not entirely clear if another extension is legal, but CNN reports that a second extension might be an uphill battle if challenged in court.) In the meantime, TikTok is unlikely to go dark like it (briefly) did in January when the ban law first went into effect.

This article originally appeared on Engadget at https://www.engadget.com/social-media/tiktoks-ban-deadline-is-coming-what-happens-next-162420478.html?src=rss

©

© CFOTO via Getty Images

Meta has spent the last few years remaking Facebook’s main feed into a “discovery engine” that primarily serves up recommended content from pages, groups and accounts users don’t already follow. But while the company has said the change is necessary to compete with TikTok, it’s also frustrated Facebook users who miss seeing posts from people they actually know.

Now, Meta is trying to remedy this with a revamped “Friends” section of the app that will only show posts and content related to your Facebook friends. The company says the update is the first of many meant to bring “OG” features back into the 21-year-old social network.

With the update, the “Friends” tab of the Facebook app that used to only be for friend requests will now be home to a feed exclusively made up of content related to your friends. This includes feed and Story posts, as well as Reels, birthdays, friend requests and “people you may know suggestions.”

While Meta is spinning the change as a part of a larger push to make Facebook feel more “social,” it’s not the first time the company has offered dedicated feeds for “friend only” content. In fact, Mark Zuckerberg showed off dedicated “friends” feed in 2022 amid the company’s push to bring more recommendations to users’ “home” feeds. That feed, which surfaces posts from friends in reverse chronological order, is still available, though it’s somewhat buried in the app.

It’s not clear what else Meta has in store for other “OG” Facebook features that could play on nostalgia for the 21-year-old social network. (Facebook’s once infamous “poke” feature already had a brief resurgence last year.) But it’s apparently a priority for Zuckerberg.

“I actually think that there's this whole opportunity that I think is going to be pretty fun to to go after and build which is just to kind of go one-by-one and build up a a bunch of these things that used to be these joyful experiences that people had as part of Facebook that just kind of don't exist on the internet today,” Zuckerberg said in a podcast appearance promoting the change.

This article originally appeared on Engadget at https://www.engadget.com/social-media/mark-zuckerberg-decides-content-from-your-actual-friends-is-an-og-facebook-feature-180803853.html?src=rss

©

© Meta

Instagram is adding another tool to its arsenal of anti-bullying measures. This time, however, Meta is enlisting school officials directly in an effort to speed up its response to safety issues affecting young teens. The company is introducing a new “Schools Partnership Program” that aims to give teachers and school administrators a way to intervene when students experience bullying on Instagram.

The program offers schools something that many parents have long said they wanted: a way to fast-track in-app reports that directly affect young students. Meta is promising that schools who join the program will have their reports “prioritized” in its moderation queue, and that it will respond to these reports within 48 hours. Schools that sign up will also get a “school partner” badge appended to their Instagram account, along with in-app information about how to use its reporting tools and other resources.

Notably though, it doesn’t sound like Meta is hiring additional content moderators to aid with these new fast-tracked reports. When asked about increasing headcount to support the program, an Instagram spokesperson told Engadget only that reports from schools will be “expedited” via its existing reporting channels, and that schools will be able to add "additional context" to their reports.

Meta says it’s already tested the program with 60 schools and that early results have been “promising.” Now, middle schools and high schools in the United States can sign up to join a waitlist for the program.

The pilot program is far from the first time Instagram has tried to address bullying. The app introduced comment limiting features in 2021, and added a tougher "restrict" feature last year. But while these features can help teens avoid certain types of negative interactions, they still required teens (or their parents) to tweak their settings. By now giving schools more authority to report specific instances of bullying or other bad behavior, Meta is offering some hope that it will be easier for schools to address the underlying behavior.

This article originally appeared on Engadget at https://www.engadget.com/social-media/instagram-has-a-new-program-for-school-officials-to-report-bullying-160027748.html?src=rss

©

© Meta

Earlier this year, with a TikTok ban looming, Perplexity AI threw its hat into the increasingly crowded ring to take over the embattled social media company. Now, the maker of the AI search engine is outlining its vision for TikTok if its bid were to win out.

In a blog post, the company made a lengthy pitch for "rebuilding TikTok in America," an effort that would see the AI company rebuild the app’s signature algorithm, add Community Notes, and integrate the shortform videos into its search engine.

Of the many potential buyers, it’s not clear how seriously Perplexity’s bid is being considered. On Friday, the Financial Times reported that ByteDance’s existing US investors are "exploring a deal alongside software giant Oracle" in an effort to "appease" President Donald Trump. The story also notes that ByteDance "strongly prefers" this arrangement.

Still, Perplexity is making the case that it is best suited to own the platform.

Perplexity is singularly positioned to rebuild the TikTok algorithm without creating a monopoly, combining world-class technical capabilities with Little Tech independence. Any acquisition by a consortium of investors could in effect keep ByteDance in control of the algorithm, while any acquisition by a competitor would likely create a monopoly in the short form video and information space. All of society benefits when content feeds are liberated from the manipulations of foreign governments and globalist monopolists.

The AI company also says it would rebuild TikTok’s "black box" recommendation algorithm from scratch and would make the "For You" feed open source. Perplexity also suggests it would add "Community Notes features" as well as the citations used by its own search engine in an effort "to turn TikTok into the most neutral and trusted platform in the world." That would be a significant departure from TikTok’s current fact-checking program, which relies on international media organizations.

This is the first time Perplexity has publicly discussed its vision for TikTok in detail. It’s also notable that some of its plans — like open-sourcing the "For You" feed and adding Community Notes — feel ripped straight from Elon Musk’s playbook for X. In any case, the White House will likely need to make a decision about TikTok’s future soon. Trump’s executive order that temporarily saved the app from a ban is set to expire April 5, although the president has indicated he would "probably" extend it if necessary.

This article originally appeared on Engadget at https://www.engadget.com/social-media/perplexity-ai-says-it-would-rebuild-tiktoks-algorithm-and-add-community-notes-features-200449390.html?src=rss

©

© RICCARDO MILANI via Getty Images

Meta’s Threads is finally allowing users to change their default feed in the app, in a move that addresses what has long been one of the most frustrating parts of the service. With the update, users will be able to make the “following” feed or a custom feed the default view in the app rather than Threads’ sometimes chaotic “for you” feed.

Mark Zuckerberg said that Meta would test the feature way back in November, but the company has said little about it since. Now that it’s official, the change could help address criticism that Threads isn’t as useful for following real-time events because it pushes users to the “for you” algorithmic feed, which often surfaces older posts with many replies rather than newer posts from followed accounts.

It could also help Meta as it tries to lure more users from X and Bluesky, which saw a surge of new users last year. Threads currently has 320 million monthly active users, according to a Meta spokesperson, while Bluesky has about 33 million.

The long overdue change arrives alongside several other upgrades to the service. The app is also finally getting a setting that allows users to limit quotes and replies to followers only. That should address complaints about reply guys and unnecessary dunking (previously, the app’s quote controls had options for “profiles you follow” and mentioned accounts only).

Meta

Threads is also making changes that make it easier to surface topics in various parts of the app. As a refresher, topics on Threads are essentially the site’s version of hashtags (with a few notable distinctions). Now, the app will surface topics in more places in an effort to steer users toward more relevant posts.

While drafting a new post, Threads will allow users to browse topics, including ones that are currently trending in the app. The service will also more prominently highlight trending topics alongside relevant feed posts. Meta is also testing the ability to add topics to Threads profiles, much like how you can add hashtags to Instagram profiles. The company notes that taking advantage of topics could help users’ expand their reach, noting that “internal data” suggests that posts with a tagged topic “generally receive more views than those that without one.” That will likely be useful info for Threads creators trying to decipher the app’s sometimes confusing algorithm.

This article originally appeared on Engadget at https://www.engadget.com/social-media/threads-will-finally-let-everyone-change-their-default-feed-170001140.html?src=rss

©

© NurPhoto via Getty Images

It looks like Meta’s campaign to bury a dishy, tell-all memoir from a former employee isn’t exactly working. In fact, it seems that the company’s legal maneuvers to block the book have had the exact opposite intended effect.

Careless People has debuted at number one on the New York Times bestseller list. It’s also spent much of the week in the number three position on Amazon’s bestseller list. That’s despite the fact that Sarah Wynn-Williams, a former policy director at Facebook, has been barred from promoting or publicly discussing the book following legal action by Meta.

The memoir details much of the inner workings of Facebook during the nearly seven years Wynn-Williams worked there. It reveals new details about the company’s relentless ambition to bring the social network to China, as well as numerous shocking stories about how Mark Zuckerberg, Sheryl Sandberg and other executives behaved behind closed doors. (In addition to her revelations in Careless People, Wynn-Williams has also filed a whistleblower complaint with the Securities and Exchange Committee. That complaint has not been made public, but according to NBC News, she alleges that Facebook “misled” investors.)

Meta has come out forcefully against Wynn-Williams and the book, describing it in a statement as “a mix of out-of-date and previously reported claims about the company and false accusations about our executives.” The company says Wynn-Williams was fired in 2017 “for poor performance and toxic behavior.”

Meta initiated arbitration proceedings against Wynn-Williams last week, resulting in a decision that for now prevents her from publicly discussing or promoting the work. But, as the bestseller lists show, the move has done little to kill interest in the work.

This article originally appeared on Engadget at https://www.engadget.com/social-media/careless-people-hits-the-top-of-the-new-york-times-bestseller-list-235008451.html?src=rss

©

© ROBYN BECK via Getty Images

Sarah Wynn-Williams joined Facebook in 2011 when it was very much in its “move fast and break things” era. A former United Nations diplomat, she had aspirations of helping Facebook boost its standing on the global stage.

Over nearly seven years at the company, she had a front-row seat to some of Facebook’s most consequential (and controversial) moments. Last week, she published Careless People: A Cautionary Tale of Power, Greed, and Lost Idealism, a memoir about her time at the company and how she slowly became disillusioned about Facebook’s role in the world.

Since it was announced, Meta has thrown the full weight of its PR machine against the book. The company initiated arbitration proceedings, which resulted in a ruling that now prevents Wynn-Williams from promoting her own book. Meta spokesperson Andy Stone said the arbitrator’s decision “affirms that Sarah Wynn Williams’ false and defamatory book should never have been published.” He said that “urgent legal action was made necessary by Williams, who more than eight years after being terminated by the company, deliberately concealed the existence of her book project and avoided the industry’s standard fact-checking process in order to rush it to shelves after waiting for eight years.”

Her publisher, Flatiron Books, said in a statement it was “appalled” by Meta’s actions, noting that “the arbitrator's order makes no reference to the claims” in the book, which “went through a thorough editing and vetting process.”

After reading the book, it’s easy to see why Meta PR has made such a fuss. Wynn-Williams worked closely with the company’s top executives, traveling around the world with Mark Zuckerberg and former COO Sheryl Sandberg as they met with foreign leaders and shaped the policies that would define Facebook. She also delivers many firsthand accounts of what Zuckerberg, Sandberg and current Chief of Global Affairs Joel Kaplan did and said behind closed doors. Much of it isn’t pretty.

Here are some of her most shocking claims:

Zuckerberg requested to be “gently mobbed” during a visit to Asia

Wynn-Williams was often tasked with organizing the meetings various Facebook executives took with heads of state and other government officials. One of the more bizarre details is Zuckerberg’s request ahead of a three-week long trip to Asia. According to Wynn-Williams, Zuckerberg requested that she arrange either a “peace rally” or a “riot” during his visit. Facebook’s CEO never explains the request, but Wynn-Williams speculates he wanted "to test out how effective his product is in turning Facebook’s online tools into offline power.”

At first, I think he's joking. Peace rallies are not my area of expertise, and-to be frank-I've never been asked to organize a riot before. Especially one for a tech CEO. I assume something has been lost in his communication-that it's some sort of mix-up. "Riot" and "peace rally" are such completely different things. Then Debbie emails to say that she ran into Mark and he told her that he wants a peace rally or a riot and we need to come up with some ideas that will enable him to be surrounded by people or be "gently mobbed."

Zuckerberg refused to take meetings before noon, even with heads of state

Zuckerberg apparently had a “strict” policy about not taking any meetings before noon. This policy applied not only to internal meetings with others at Facebook, but heads of state. Wynn-Williams recounts how Zuckerberg’s “refusal” to take morning meetings almost derailed a planned meeting with the president of Colombia. That same year, she said she was forced to reschedule Zuckerberg’s address at the UN to a later time slot because, according to Wynn-Williams, “the United Nations isn’t important enough for Mark to do an event before noon.”

Meta built Zuckerberg a special room to protect him from Zika

Facebook took “extreme measures” to protect Zuckerberg from Zika during a visit to Peru for the APEC conference. To protect the CEO, who was apparently hoping to soon conceive his second child, the company opted "to build a ‘controlled structure’ on the site of the APEC conference center where ventilation, exposure to others, and bug mitigation can be overseen by Facebook.” Wynn-Williams dubbed it “operation perfect sperm.”

Zuckerberg’s “desperate” attempts to talk to Xi Jinping

Wynn-Williams spends a lot of time delving into Facebook’s plans to bring the social network to China. She also details Zuckerberg’s attempts to get face time with Chinese President Xi Jinping.

In 2015, Zuckerberg flew to Seattle for a “longer than normal” handshake with Xi (who was in the city for a tech summit Zuckerberg was not invited to). He later caused a “diplomatic crisis” when he posted a photo of that handshake, which only shows the back of Xi’s head.

The following year, Zuckerberg and his team hoped to surreptitiously get the CEO in front of Xi at APEC in Peru. Facebook arranged for Zuckerberg to give a speech directly before an appearance by Xi, which meant the two men would be in adjacent dressing rooms. The goal, according to Wynn-Williams, was to engineer a “spontaneous encounter” during which Zuckerberg could make his pitch to get Facebook into China.

But when Xi arrived, he was flanked by “dozens and dozens” of men in military uniforms who created a barrier such that, according to Wynn-Williams, Xi didn't even have to “risk making eye contact with Mark.” She says that Zuckerberg was “hurt” by the demonstration.

Sheryl Sandberg wanted Facebook to promote organ donation

Not long after starting at Facebook, Wynn-Williams says she was told that “Facebook’s first proactive initiative to build relationships with governments around the world will be organ donation.” The idea was apparently Sandberg’s, who ran into a former Harvard friend and transplant surgeon and “offered to help him source donors.”

But when Wynn-Williams brought up the logistical and legal issues that generally make it difficult to transport organs between countries, she says Sandberg was “indignant.”

“Do you mean to tell me that if my four-year-old was dying and the only thing that would save her was a new kidney, that I couldn’t fly to Mexico and get one and put it in my handbag?”

Facebook would later add organ donation to the “life events” users could add to their profiles.

Sheryl Sandberg and the plane crash that wasn't

In 2013, an Asiana flight from Seoul to San Francisco crash-landed on a runway, killing two people and injuring more than 100 passengers. Sandberg, who was promoting her book Lean In at the time, was also traveling from Seoul to San Francisco that day. Following the crash, she posted on Facebook that she and several Facebook colleagues were originally meant to be on that flight but had “switched to United so we could use miles for my family's tickets.”

Sandberg’s supposed near-miss generated multiple headlines, which isn’t surprising, as Wynn-Williams claims Sandberg was personally briefing reporters about the incident. Wynn-Williams was baffled by these reports. She writes that the Facebook COO “always flies United” and “never considered Asiana.” In her recollection, she and another colleague remarked to each other how “weird” it was for Sandberg to do this.

Why Zuckerberg and Sandberg’s Facebook posts get so much engagement

Wynn-Williams heavily implies that there may be some behind-the-scenes manipulation of Zuckerberg and Sandberg’s Facebook posts. She doesn’t offer any proof that this has happened, but says that a meeting attended by “someone from Mark’s personal communications team” raises the possibility.

According to Wynn-Williams, “Sheryl’s assistant describes his job as managing ‘engagement for Mark and Sheryl's social media using all of Facebook's tools.’” When Wynn-Williams asks directly if they are “gaming” the site’s algorithm, she says the assistant replies that “you don’t want to know.”

Joel Kaplan needed a geography lesson

Some of the most telling anecdotes in Careless People involve Kaplan, who joined Facebook’s policy team in 2011 and was promoted to Chief Global Affairs Officer earlier this year. She writes that Kaplan, who was a deputy chief of staff in the George W. Bush White House, was “surprised to learn Taiwan is an island” and that “often when we start to talk about pressing issues in some country in Latin America or Asia, he stops and asks me to explain where the country is.”

Kaplan slow-rolled Facebook’s response to problems in Myanmar

By now, Meta’s failures in Myanmar, where hate speech and misinformation on Facebook helped incite a genocide, are well documented. Wynn-Williams, who early in her tenure flew to Myanmar to try to sell officials there on the company’s connectivity projects, describes her futile attempts to get more resources for content moderation in the country.

She blames Kaplan in particular. She says she “started this long process of trying to hire someone for Myanmar in 2015” and found a human rights expert who fit the bill in May 2016, but Kaplan blocked her from making the hire in February of 2017. He allegedly told her to “move on and get over it.” She later concludes that “when it came to Myanmar, those people just didn’t matter to him.”

(Meta had broadly labeled Wynn-Williams’ claims about Facebook’s actions in Myanmar as “old news,” saying in a statement that “the facts here have been public record since 2018, and we have said publicly we know we were too slow to act on abuse on our services in Myanmar.”)

Kaplan’s alleged harassment

Wynn-Williams also accuses Kaplan of harassment and other inappropriate behavior. After complications while giving birth left her in a literal coma, she writes that Kaplan asked her over one of their “regular” calls during her maternity leave “where are you bleeding from” and got angry when she didn't answer. Later, when she returned to work, he gave her an unofficial performance review on her first day back, saying that she wasn’t “responsive enough.”

Wynn-Williams eventually reports Kaplan’s behavior and an investigation is opened, but that “very quickly it seems to switch from an investigation of Joel, or the facts, to an investigation of me.” She’s fired at her next performance review.

Meta has disputed Wynn-Williams' allegations. “This book is a mix of out-of-date and previously reported claims about the company and false accusations about our executives," a spokesperson said in a statement. "Eight years ago, Sarah Wynn-Williams was fired for poor performance and toxic behavior, and an investigation at the time determined she made misleading and unfounded allegations of harassment."

Update, March 18, 2025, 1:57 PM PT: Wynn-Williams describes her tenure at Facebook as lasting seven years, beginning in 2011. But, according to a Meta spokesperson, she was fired in "late 2017." This story has been updated to reflect that she worked at Facebook for "nearly" seven years.

This article originally appeared on Engadget at https://www.engadget.com/social-media/the-wildest-details-in-the-facebook-memoir-meta-is-trying-to-bury-183310491.html?src=rss

©

© Macmillan

Twitter may be dead, but the 12-foot tall bird logo from its San Francisco headquarters will live on. The sign — one of two birds that formerly adorned Twitter’s office — has been sold at auction for $34,375.

That's a bit less than $40,000 auctioneers RR House estimated it would fetch, but is nonetheless a pricey piece of social media history. The massive logo that marked the company’s headquarters on San Francisco's Market Street is undoubtedly one of the most iconic symbols associated with the social media company. (The sign was unceremoniously removed after Musk’s rebranding of the company to X.) 

The unknown buyer will also responsible for costs to move the sign from the San Francisco storage facility where it’s currently located. That alone will be no small feat. The sign is 145˝ x 105˝ (roughly 12 feet by 8 feet) and weighs in at 560 pounds, according to the listing. A YouTube video accompanying the listing shows that it took a crane and a team of several workers to remove the sign from the building.

RR Auction

Exactly where "Larry" will end up next isn't clear as the winning bidder's identity hasn't been disclosed, but the 12-foot tall sign is bound to make quite a statement, wherever it lands. 

Notably, this isn’t the first time “Larry” (the nickname former Tweeps gave to the site’s iconic bird logo) hit the auction block. Following Elon Musk’s takeover of Twitter in 2022, much of the contents of Twitter’s former offices were auctioned off in 2023. Other, slightly smaller versions of the logo proved to be popular at the time, with one statue that decorated the office selling for $100,000. The building’s other large bird logo was auctioned off in that sale for an undisclosed price.

Update, March 21, 2025, 1:47 PM ET: Now that the auction has closed, this story has been updated to add information about the winning bid.

This article originally appeared on Engadget at https://www.engadget.com/social-media/the-560-pound-twitter-logo-has-sold-for-34000-180326846.html?src=rss

©

© RR Auction

Meta has notched an early victory in its attempt to halt a surprise tell-all memoir from a former policy executive turned whistleblower. An arbitrator has sided with the social media company, saying that the book’s author should stop selling and publicizing the book, which went on sale earlier this week.

The drama stems from Careless People, a new book by Sarah Wynn-Williams, a former policy official at Facebook who Meta says was fired in 2017. Described by its publisher as an “explosive insider account,” Wynn-Williams reveals some new details about Mark Zuckerberg’s push to bring Facebook to China a decade ago. She also alleges that Meta's current policy chief, Joel Kaplan, acted inappropriately, and reveals embarrassing details about Zuckerberg’s awkward encounters with world leaders.

The book was only announced last week, and Meta has waged a forceful PR campaign against it, calling it a “new book of old news.” Numerous former employees have publicly disputed Wynn-Williams’ account of events that transpired while she worked at Facebook.

Meta also filed an emergency motion with an arbitrator in an attempt to block the book, arguing that Wynn-Williams violated a non-disparagement agreement with the company. In a decision, the arbitrator said that she should immediately stop making disparaging comments about Meta and stop “further publishing or distributing” the book. It’s unclear what exactly this means for the memoir, which is already on sale. In a statement, the publisher Flatiron Books, which was also named in Meta’s complaint, said that "the arbitration order has no impact on Macmillan," which owns Flatiron.

Meta claimed the decision as a victory. “This ruling affirms that Sarah Wynn Williams’ false and defamatory book should never have been published,” Meta spokesperson Andy Stone said in a statement. “This urgent legal action was made necessary by Williams, who more than eight years after being terminated by the company, deliberately concealed the existence of her book project and avoided the industry’s standard fact-checking process in order to rush it to shelves after waiting for eight years.”

In a statement, Flatiron Books said it was "appalled" by Meta's actions and had no plans to stop promoting the book. "We are appalled by Meta's tactics to silence our author through the use of a non-disparagement clause in a severance agreement," a spokesperson said. "To be clear, the arbitrator's order makes no reference to the claims within Careless People. The book went through a thorough editing and vetting process, and we remain committed to publishing important books such as this. We will absolutely continue to support and promote it."

Update March 13, 2025, 8:55 AM PT: This story has been updated to add a statement from Flatiron Books.

This article originally appeared on Engadget at https://www.engadget.com/social-media/meta-is-trying-to-stop-a-former-employee-from-promoting-her-book-about-facebook-004938899.html?src=rss

©

© Chesnot via Getty Images

Google is pushing back on a bill that would make Utah the first state in the US to have a law requiring app stores to conduct age verification of their users. The company has formally requested the state's governor to veto the bill, passed by the state legislature last week, as it urges states to consider a different approach to app safety.

In a blog post from Google public policy director Kareem Ghanem, the company said the Utah law poses “real privacy and safety risks” and that states shouldn’t rush to pass laws “pushed by Meta” and other social media companies. Instead, Ghanethe says that Google has proposed an “alternative legislative framework” that would allow the developers of potentially “risky” apps to request “age signals” from app store owners like Google.

The statement is the first time Google has publicly opposed the bill, which would make Apple and Google responsible for age verification and parental permission features for children under 18. A spokesperson also confirmed that the company requested that Utah Governor Spencer Cox veto the bill. Cox has previously signed off on laws that imposed age verification and parental permission requirements on social media companies, though the measures were revised and later blocked by a judge. A spokesperson for Cox didn’t immediately respond to a request for comment.

While it’s not surprising that Google is opposing the law and others like it (there are at least eight other states considering similar measures), it’s notable that the company is proposing its own parameters for a law that would take into account what kind of content is available in a given app, which would place more of a burden on social media companies.

Ghanem argues that not all apps should be subject to age verification. “This level of data sharing isn’t necessary — a weather app doesn’t need to know if a user is a kid,” he writes. “By contrast, a social media app does need to make significant decisions about age-appropriate content and features.” He also argues that app store safety proposals should come with a ban on all personalized ads to anyone under 18, pointedly noting that "other companies" should follow Google's lead on the issue.

Google obviously has a vested interest in not being responsible for age verification of its users, which would impose significant logistical and legal risks for the company. But many other privacy and digital rights groups have also opposed age verification laws.

This article originally appeared on Engadget at https://www.engadget.com/big-tech/google-asks-utahs-governor-to-veto-the-states-app-store-age-verification-bill-233733280.html?src=rss

©

© NurPhoto via Getty Images

One day after X went down for hours, security researchers are throwing cold water on Elon Musk’s public comments about who might be behind the DDoS attack. On Monday, as X was still struggling to remain online, Musk said in a post that the site had been brought down by a “massive cyberattack” executed by “a large, coordinated group and/or a country.” Later that day, in an interview with Fox News, he said the attack involved "IP addresses originating in the Ukraine area."

He never provided evidence for either claim. But, in a new report from Wired, security researchers offered a very different view on the attack. Security experts interviewed by the publication said that they had seen little evidence that Ukrainian IP addresses played a significant role in the DDoS attack, with one researcher saying the country wasn’t even in the top 20 countries of origin involved.

The report also suggests that, despite Musk’s assertion there were “a lot of resources” involved, X may have inadvertently left its systems susceptible to a DDoS attack like the one that happened Monday. “X origin servers, which respond to web requests, weren't properly secured behind the company's Cloudflare DDoS protection and were publicly visible,” Wired writes. “As a result, attackers could target them directly. X has since secured the servers.”

Notably, this wouldn’t be the first time Musk has blamed an unspecified “cyberattack” when faced with an embarrassing failure of X’s systems. Last year, Musk blamed a “massive DDoS attack” for crashing a planned livestream with Donald Trump, who was running for president at the time. Musk never explained how a DDoS attack could bring down only one feature on the site. The Verge later reported that there had been no such attack.

X didn’t respond to a request for comment.

This article originally appeared on Engadget at https://www.engadget.com/social-media/security-researchers-arent-buying-musks-spin-on-the-cyberattack-that-took-down-x-203402687.html?src=rss

©

© Andrew Harnik via Getty Images

Bluesky continues to beef up its video capabilities. The app’s latest update adds support for uploads of up to 3 minutes, a notable expansion from its previous limit of one minute. The change should give video creators on Bluesky much more flexibility in what they can post. It will also likely be appreciated by fans of Flashes, the new Bluesky-based photo and video app.

The update also brings some much needed inbox-management tools to Bluesky DMs, with the addition of a new “requests” section for incoming messages from users you don’t know. You’ll also now be able to reject incoming chats, which should help reduce the number of spammy DMs that have been flooding many users’ inboxes in recent weeks.

The latest Bluesky update rolled out as CEO Jay Graber appeared onstage at SXSW for a keynote about the growing decentralized social media platform, which has just under 33 million users. Graber, who was wearing a t-shirt emblazoned with a Latin slogan “mundus sine caesaribus” (a world without Caesars) — an apparent play on Mark Zuckerberg’s affinity for Roman Empire-inspired attire — spoke about why Bluesky is “billionaire-proof” compared with other apps.

Screenshot via X

“If a billionaire came in and bought Bluesky or took it over, or I decided tomorrow to change things in a way that people really didn't like, then they could fork off and go on to another application,” she said. “That openness guarantees that there's always the ability to move to a new alternative, still using the same network. There's much less tolerance in the ecosystem for a very heavy-handed change.”

She also opened up about how Bluesky is thinking about its business model and generative AI. On the business side, she once again said the company is “exploring subscriptions” as well as paid “developer services” for those who build experiences on top of Bluesky. In a similar vein, she also said the company was considering how it could eventually turn some of its features, like moderation services or custom feeds, into a “marketplace” where the company may get a cut of potential transactions.

When it comes to large language models, she said the company was “working with some partners on developing a framework for user consent, for how they want their data to be used for generative AI.”

This article originally appeared on Engadget at https://www.engadget.com/social-media/bluesky-adds-3-minute-video-uploads-and-inbox-management-features-220317950.html?src=rss

©

© SOPA Images via Getty Images

It seems that Instagram is working on a “community chat” feature that allows people to organize groups of up to 250 people in the app. The so-far unreleased feature was spotted by developer Alessandro Paluzzi, who has a solid track record of uncovering new features within Meta’s apps.

According to screenshots shared by Paluzzi, it seems that community chats will function similarly to Discord. Individual users can form the chats around specific topics and control who can join, though there’s apparently a limit of 250 people per community.

Unlike Instagram’s broadcast channels, which allow creators to blast out messages to their followers, anyone who is in the community chat can participate in the conversation. There are also built-in moderation features. “Admins can remove messages and members to keep the channel safe,” the screenshot says. “We also review Community Chat against our Community Standards.”

It’s not clear when, or if, the feature may launch. An Instagram spokesperson described it as an internal prototype that’s not being tested outside the company. But Meta has previously released similar features in its other apps. WhatsApp began experimenting with a “Communities” feature in 2022, and brought “Community Chats” to Facebook and Messenger later that same year. Mark Zuckerberg said at the time it was meant to help people find “a new way to connect with people who share your interests.”

This article originally appeared on Engadget at https://www.engadget.com/social-media/instagram-is-experimenting-with-a-discord-like-community-chat-feature-234832236.html?src=rss

©

© Anadolu via Getty Images

Utah has once again passed a bill that attempts to place new limits on children’s ability to access social media and other apps. The state legislature passed a bill that requires app store makers to verify the ages of users and get parental consent for the downloading of apps. Should the governor sign the bill into law, it would be (as The Verge points out) the first such measure in the country.

The Beehive State has previously attempted to restrict how children can use social media. The state passed two laws in 2023 that imposed age verification and parental permission requirements onto social media companies. Those laws were later revised amid legal challenges, and a Utah judge blocked them from taking effect last fall.

The “App Store Accountability Act” takes a different approach. It puts the burden on app stores (namely, Apple and Google) to conduct age verification and parental permission rather than individual social media platforms. Unsurprisingly, Meta and other social media companies have long preferred the app store-centric approach to age verification.

On Wednesday, Meta, Snap and X issued a rare joint statement praising the Utah measure. “Parents want a one-stop shop to verify their child’s age and grant permission for them to download apps in a privacy-preserving way,” the companies said. “The app store is the best place for it, and more than a quarter of states have introduced bills recognizing the central role app stores play. We applaud Utah for putting parents in charge with its landmark legislation and urge Congress to follow suit.”

Apple and Google so far haven’t publicly weighed in on Utah’s bill, though we’ve reached out to both for comment.

Given the controversy surrounding previous attempts to impose age verification requirements on platforms, the Utah bill will likely face legal challenges if and when it’s signed into law. Even so, other states are likely to follow Utah’s lead on this issue. At least eight more states are considering similar measures.

This article originally appeared on Engadget at https://www.engadget.com/big-tech/utah-is-poised-to-pass-an-age-verification-law-for-app-stores-000542476.html?src=rss

©

© NurPhoto via Getty Images

Elon Musk discovered a simple 30-second “fridge trick” that can reverse diabetes, but the discovery has spooked pharmaceutical companies so much they put a $78 million bounty on his head, forcing the Tesla CEO to flee the country. At least, that’s what a collection of AI-generated Facebook ads claim.

Facebook ads depicting deepfakes of Elon Musk and Fox News personalities claiming that the Tesla CEO has discovered the cure for diabetes have been circulating on the platform for weeks. The ads seem to be part of a wider scam that uses the deepfakes to sell unproven supplements.

Engadget has identified scores of pages running versions of these ads since early February. Though many of the ads don’t stay up for very long, the same pages have repeatedly run dozens or even hundreds of versions of the videos, all of which follow a similar format. The vast majority feature Elon Musk and Fox News hosts, but there are also versions with other notable figures, including Oprah Winfrey, former Utah Senator Mitt Romney and Health and Human Services Secretary Robert F. Kennedy Jr.

Using social media to hawk sketchy supplements isn’t exactly new. AFP Fact Check detailed a similar ring of scammy Facebook ads promoting supplements that purported to cure high blood pressure last year. But the addition of AI-manipulated video and deepfaked audio of Musk adds a new layer to such scams.

While this particular scam has apparently gone largely undetected by Meta, the company is well aware that fraudsters often use celebrities and other public figures to lure in unsuspecting users. A Polish billionaire recently won a legal battle with the company over deepfake ads that used his likeness. Last fall, the company announced that it would bring back facial recognition technology in an effort to fight “celeb bait” scams.

Musk, of course, is no stranger to impersonation scams. Crypto scammers have impersonated him across social media for years. His new position as a close lieutenant of President Donald Trump and head of the “Department of Government Efficiency,” however, makes him an even more attractive target to fraudsters hoping to capitalize on his newfound authority. Several of the ads identified by Engadget feature video clips of Musk’s recent public appearances and speeches. These include ads with AI-generated audio overlaid onto video from an onstage interview this month at the Conservative Political Action Conference, known as CPAC, and a photo opp in the Oval Office.

“There are a lot of different ‘blood sugar support,’ type 2 diabetes, reverse diabetes, sort of scams out there these days,” says Jordan Liles, a senior reporter at Snopes who has debunked some of these videos on his YouTube channel. “It's the same sort of process you might see with other scams in that they start with an ad, they go to a long video on a website, and then they want you to order the product. And they want you to believe everything that you just saw is real.”

These ads seem to violate multiple policies of Meta, including its rules around medical misinformation and requirements for advertisers to disclose the use of AI-manipulated media. And many of the ads identified by Engadget appear to have been removed within a day or two. But many of the Facebook pages behind these ads, which have names like “Mindful Medicine,” “Your health compass” and “The Wellbeing Spot” remain active. Several pages repeatedly ran dozens of new ads with similar content shortly after the old ones were removed. 

Because ads that are taken down are removed from Meta’s Ad Library along with their associated analytics, it’s hard to know just how many accounts they are reaching. But the few that remain viewable offer some clues. One such ad, from a page called “Healthy Style,” was active on Facebook, Instagram and Messenger for three days, according to the company’s Ad Library. The advertiser paid Meta between $300 and $399 for the ad, which generated 5,000 - 6,000 impressions. At one point in February, the page was running 177 versions of a similar ad. Another, from a page called “Hydromeduoside” also ran for two days, generating between 10,000 and 15,000 impressions across Facebook, Instagram and Messenger. It cost between $400 and $499, according to the Ad Library. Both ads were removed by Meta because they “ran without the required disclaimer” for ads about social and political issues, according to a notice in the Ad Library, despite likely breaking other rules and having no particular relation to politics.

Meta removed the "Healthy Style" page after an inquiry from Engadget. The company said in a statement it was investigating the ads and taking action on other pages. "The ad review system reviews ads for violations of our policies," a spokesperson said. "This review process may include the specific components of an ad, such as images, video, text and targeting information, as well as an ad’s associated landing page or other destinations, among other information."

The ads follow a similar script. They often open with deepfakes of Fox News personalities discussing how Musk recently “blew up the internet” with his discovery that “big pharma” doesn’t want you to see. The ads then cut to an AI-manipulated video clip of him from a podcast or other public appearance. Many of the ads refer to the “MAHA” website, an apparent reference to Kennedy’s “Make America Healthy Again” movement.

The ads never disclose the actual “trick” for reversing diabetes. In one version, Musk says all you need is a “common fruit.” In another, Kennedy says it requires a “frozen lemon.” Another suggests that all you need is “a lemon and a pair of socks.” The ads urge viewers to visit a linked website to find a guide “before it’s wiped from the internet.” Bizarrely, the videos end with several minutes of silence that show a still image of a random object like a clock, calculator or blood pressure cuff.

Most of these ads link to websites that attempt to spoof mainstream news sites like Fox News, CNN or Yahoo (Engadget’s parent company). These websites typically don’t have obvious links to supplements, but feature additional lengthy videos with AI-manipulated clips of cable news personalities. These videos, which can run upward of 20 minutes, often feature old video clips with AI-generated audio of Barbra O’Neill, an Australian naturopath who was permanently banned from offering health-related services in her home country after claiming that cancer could be cured with baking soda.

(There’s no evidence that O’Neill, who has emerged as a popular wellness influencer in the US in recent years, is actually associated with these websites. As The Guardian reported last year, O'Neill is often impersonated by scammers using AI-generated content to sell unproven remedies on Facebook and TikTok. Her official Facebook page often warns followers about scammers who impersonate her on and off Facebook. A representative for O’Neill didn’t respond to a request for comment.)

Once you make it to the end of the video, a link appears offering a time-limited “deal” on a supplement called “Gluco Revive.” The supplement, which is also available from third-party sellers on Amazon and walmart.com, doesn’t actually claim to treat any condition related to diabetes in its marketing. Instead, it suggests it offers “blood support” that “boosts energy levels” and "supports healthy weight.” According to its label, it has a “proprietary blend” of ingredients like juniper berry powder and bitter melon extract.

The FDA has previously warned consumers about companies selling over the counter diabetes treatments and supplements. “There are no dietary supplements that treat or cure diabetes,” the agency wrote in a warning published in December. “Some ‘all-natural’ diabetes products contain hidden active ingredients found in approved prescription drugs used to treat diabetes … these products may interact in dangerous ways with other medications.”

While it’s impossible to know just how many people have been duped by these ads, there are dozens of listings for Gluco Revive on Amazon, some of which indicate there have been thousands of sales in the last month. The reviews on Amazon, where it’s sold for anywhere from $22 - $70, suggest that some people may have purchased the supplements after seeing one of the ads described above. “It’s not working. I thought someone from the company was going to call me on what to do. Dr. O’Neil [sic] said that we might feel a change in about 17 hours. It’s been 7 days and I feel no difference.”

“This does not work and in fact raised my blood sugar,” another reviewer reports, noting they found the product via Facebook. “When contacting the suppler [sic] they did send me a Return authorization number and address to send the unused bottles …I boxed them up [and] sent priority shipping and it's being returned as no such address. I called USPS and they let me know that it was a scam and they have had 100s of returns to that address [and] that they do not exist.”

Though Gluco Revive appears to be the most prevalent supplement sold via this scam, some linked pages from the Facebook ads sell similar products, like “Gluco Control," which purports to be a “a 100% natural, safe, and effective blood sugar management product.” The website says it’s made up of gymnema sylvestre, grape seed extract, green tea extract and “African mango extract.” The website features a list of scientific references and media endorsements that appear to be entirely made up. Another page promotes something called “Gluco Freedom,” which has a similar ingredient list, and offers a 180-day supply for $294.

Liles, the Snopes reporter who identified the Gluco Revive ads on Facebook in early February, and has researched many similar scams on social media, suspects the ads are having some success. “If the same product gets weeks and weeks, or months or longer of advertising, that means that it's something that they're seeing results with,” he says.

Have a tip? Get in touch with the author at karissa.bell [at] engadget.com or message securely on Signal: karissabe.51

This article originally appeared on Engadget at https://www.engadget.com/social-media/facebook-scammers-want-you-to-think-elon-musk-can-cure-diabetes-153042842.html?src=rss

©

© Screenshot via Facebook

Less than two weeks before Donald Trump’s inauguration, Meta announced sweeping changes to its content moderation procedures, reportedly at the behest of Mark Zuckerberg and a small group of advisors. Among those caught off guard was the company’s own Oversight Board, the independent organization created by Meta to help shape its most sensitive policy decisions. The group is now looking to examine those changes, and in doing so, will test the enforceability of its own powers.

The changes Meta has enacted drastically reshape how the company polices content across Facebook, Instagram and Threads. It ended its fact-checking program in the US, and rolled back hate speech rules that protected immigrants and LGBTQ users on its apps. Unlike its previously proactive approach, it also reoriented its content moderation procedures so that many types of rule-breaking posts will only be removed if other users reported them.

The changes have raised questions about the role of the Oversight Board, which was created, Zuckerberg once said, because “Facebook should not make so many important decisions about free expression and safety on our own.” If that’s what Meta is now doing, critics have asked, what exactly is the point of an ostensibly independent Oversight Board?

But the Oversight Board is already working to address Meta’s rewritten “hateful conduct” policy, according to board member Paolo Carozza, who spoke to Engadget. When Zuckerberg announced the changes in early January, the board already had four open cases involving Meta’s hate speech rules. The board now plans to use those cases to examine the new policies, which were rewritten to allow people to use dehumanizing language to describe immigrants and accuse LGBTQ people of being mentally ill.

“We deliberately delayed the decision of those cases after January 7, precisely so that we could go back to Meta again and ask a new round of questions,” Carozza, a law professor at Notre Dame who joined the Oversight Board in 2022, told Engadget. “We're trying as much as possible to use the tools that we have to find out more information, bring more transparency and more certainty to how it's going to play out in practice.”

The board, according to Carozza, has already had briefings with Meta as it pushes for more details about the new hate speech policies. But it could still be some time before its findings are made public. The open cases deal with several aspects of Meta’s hate speech rules, including immigration, gender identity, hateful symbols and incitement of violence.

In addition to the questions surrounding each case, Carozza said that the board is also grappling with how to prioritize the case decisions given the renewed importance of the underlying policies. “There are competing concerns about being quick and efficient versus being more thoughtful and deliberative,” he said.

But while the board may hope to provide more transparency about Meta’s decision-making, it’s unclear how much influence the board will ultimately end up having. Under its rules, Meta is only required to comply with the group’s decisions surrounding individual posts. The board’s policy recommendations are non-binding and Meta has a mixed track record at implementing its suggestions.

It’s also unclear how the board might be able to weigh in on Meta’s other changes, like the shuttering of fact-checking programs or shift away from proactive content moderation. “We were quite critical of the fact checking program in general, but our ordinary cases make it a little bit hard to get at that problem because it doesn't come up through an appeals process within the scope of the kinds of cases that we get,” Carozza says. The board, he notes, could write a policy advisory opinion as it has with rules around COVID-19 misinformation and Meta’s cross-check rules for celebrities. But the board is only empowered to make those kinds of non-binding recommendations at Meta's request.

That gets at one of the fundamental tensions of the Oversight Board: it may operate independently, but Meta ultimately dictates how much influence it can wield. “It would be unrealistic to expect that the standard for value and success of the board is that Meta, 100% of the time, does everything we ever tell them to do,” Carozza says, “We’re one piece of a complicated jigsaw puzzle of accountability and oversight.”

Still, the fact that the group wasn’t consulted on such major policy moves has raised some uncomfortable questions for the board. Dozens of civil society groups recently signed an open letter urging board members to resign in protest. In a letter to Zuckerberg, some members of Congress said the board “is rendered toothless” when Meta refuses to follow its own principles.

Carozzo acknowledges the Oversight Board’s limitations, but says that the billions of people on Meta’s apps are ultimately better off with the board intervening where it can. “If everybody were to resign en masse … the only people who would lose are Meta’s end users, especially those who are in especially vulnerable situations [and] communities around the world.”

This article originally appeared on Engadget at https://www.engadget.com/social-media/the-oversight-board-will-weigh-in-on-metas-new-hate-speech-policies-174044682.html?src=rss

©

© ASSOCIATED PRESS

A bizarre bug is causing iPhones to automatically change the word “racist” to “Trump” when using the built-in dictation feature in iOS. The issue, which seems to have been discovered by TikTok users, crops up when using the voice-to-text feature in Apple apps like Messages.

When speaking the word “racist,” iOS briefly transcribes the text as “Trump” before changing it back to the intended word. It’s not clear what could cause this behavior. Engadget was able to replicate the issue, as you can see in the GIF below.

Apple told The New York Times it was due to “phonetic overlap between the two words,” despite the fact that the two words do not sound similar. That also doesn't explain why "racist," with a lowercase "r" would transcribe as "Trump" — seemingly a reference to our current president — and not "trump," the noun. An AI expert who once worked on Siri told the paper it could be a “serious prank” on the part of an Apple employee. 

Apple didn’t immediately respond to a request for comment.

Whether it’s a bug or prank, it surfaced at a particularly embarrassing time for the iPhone maker, which announced on Monday that it planned to invest $500 billion in manufacturing facilities for AI servers. The investment, most of which was already planned, came after Apple CEO Tim Cook met with President Donald Trump at the White House last week. 

This article originally appeared on Engadget at https://www.engadget.com/mobile/iphones-are-briefly-changing-racist-to-trump-due-to-an-ios-dictation-issue-230712021.html?src=rss

©

© NIC COURY via Getty Images