Tag Archives: author_name|Karissa Bell

Another former Facebook employee will testify at Congress about safety issues at Instagram

Posted on by

Another former Facebook employee is going public with allegations that the company failed to act on its own research showing that young Instagram users were having harmful experiences on the platform. Arturo Bejar, a former Facebook employee and consultant for Instagram, is scheduled to testify at a Senate Judiciary Committee hearing Tuesday, November, 7.

Bejar, who detailed his efforts to raise the alarm internally about Instagram safety issues to The Wall Street Journal, was a Facebook employee between 2009 and 2015 and returned to the company in 2019 to advise Instagram’s well-being team. He told The Journal that internal research showed that more than 20 percent of users younger than 16 “felt worse about themselves after viewing others’ posts” and that 13 percent “experienced unwanted sexual advances in the past seven days.”

Bejar’s disclosures come two years after another former Facebook employee, Frances Huagen, came forward with internal research about Instagram’s harmful effects on teen mental health, along with other disclosures about controversial decisions within the company. Soon after, Instagram paused work on a dedicated app for kids, and dozens of states opened an investigation into the company.

Last week, 41 state attorneys general sued Meta for “harmful and psychologically manipulative product features” that harmed the mental health of its youngest users. According to The Wall Street Journal, Bejar consulted with state officials on their case. Now, he’s set to publicly air his experiences in front of Congress,

“From Arturo’s disclosures, we now know that Mark Zuckerberg, Adam Mosseri, and other Meta executives were personally warned that millions of teens face bullying, eating disorder material, illicit drugs, and sexual exploitation, often within minutes of opening the app,” Senators Marsha Blackburn and Richard Blumenthal, both of whom sit on the judiciary committee, said in a statement. “Rather than address these deadly harms, Facebook continued to hide this information from the public and Congressional oversight, ignored recommendations to protect teens, rolled back safety tools, and dismantled teams responsible for kids’ safety.”

Meta didn’t immediately respond to a request for comment. The company told The Wall Street Journal it disagreed with Bejar’s claims that well-being research wasn’t addressed, and said the company had introduced several safety updates as a result of the work of Bejar and his team.

This article originally appeared on Engadget at https://www.engadget.com/another-former-facebook-employee-will-testify-at-congress-about-safety-issues-at-instagram-185339860.html?src=rss

TikTok says it’s removed millions of fake accounts since start of Israel-Hamas war

Posted on by

TikTok is pushing back on critics who claim the video app is falling short in its content moderation duties amid the Israel-Hamas war. In a statement, the company offered new details about the number of accounts and videos it has taken down since the October 7th attacks by Hamas.

According to TikTok, it removed more than 925,000 videos “in the conflict region” and millions more “pieces of content” from around the world. The company also said it’s experienced “spikes in fake engagement” in recent weeks. “Since Oct. 7, we've removed more than 24 million fake accounts globally and more than half a million bot comments on content under hashtags related to the conflict.”

The new details come as TikTok has faced increasing scrutiny over how its app is recommending content related to the ongoing conflict. According to NBC News, some lawmakers have recently stepped up their calls for the app to be banned amid allegations that TikTok’s algorithm is disproportionately promoting pro-Palestinian content. In its update, TikTok said that such claims were based on “unsound analysis” of its data. 

“Unfortunately, some misinformed commentators have mischaracterized our work to prevent the spread of hate speech and misinformation surrounding the crisis in Israel and Gaza, especially as it relates to antisemitism,” the company said. “Over the last few days, there has been unsound analysis of TikTok hashtag data around the conflict, causing some commentators to falsely insinuate TikTok is pushing pro-Palestine content over pro-Israel content to U.S. users.” The company added that, in the United States, the hashtag #standwithisrael had been viewed 46 million times since October 7, while #standwithpalestine had been viewed 29 million times.

TikTok isn’t the only platform to face increasing scrutiny over its moderation policies as tensions surrounding the conflict spill over onto social media platforms. Meta has faced accusations that it “shadowbanned” Instagram accounts that posted about conditions within Gaza, which it attributed to a “bug.” X, formerly known as Twitter, is dealing with a European Union investigation into its handling of misinformation related to the conflict.

This article originally appeared on Engadget at https://www.engadget.com/tiktok-says-its-removed-millions-of-fake-accounts-since-start-of-israel-hamas-war-231851792.html?src=rss

Meta will stop forcing your Threads posts onto Facebook and it can’t come soon enough

Posted on by

It looks like Meta may be pumping the brakes on one of its more aggressive, and unpopular, growth-hacking tactics for Threads. The company appears to be working on a new privacy setting so Threads users can opt-out of having their posts cross-posted to Facebook and Instagram feeds.

The unreleased feature was spotted by reverse engineer Alessandro Paluzzi, who often uncovers early versions of social media features before they officially launch. Paluzzi shared screenshots of a new “suggesting posts on other apps” toggle in Threads’ privacy settings.

The feature comes barely a week after the company acknowledged that it was promoting users’ Threads posts in Facebook feeds in an effort to boost Threads. While Meta has used similar tactics to promote its other apps in the past, the move has been widely unpopular among Threads users, many of whom are not active on Facebook and see the promotions as an intrusive overreach. Meta said last week it was “listening to feedback” in response to user complaints about not being able to opt out.

Notably, it appears as if Meta still intends to automatically enable cross-posting as a default setting. “If your profile is public, your posts may be suggested on other apps so people can discover and follow you,” the opt-out screen states.

The back and forth over the feature comes as Meta has steadily ramped up its efforts to boost Threads growth. The Twitter clone has been growing again in recent weeks, and currently has about 100 million monthly users. Mark Zuckerberg recently said he sees a path for the app to become Meta’s next billion-user service. But in order to reach that many people, the company will need to lean hard on its other apps to attract new sign-ups.

This article originally appeared on Engadget at https://www.engadget.com/meta-will-stop-forcing-your-threads-posts-onto-facebook-and-it-cant-come-soon-enough-174835068.html?src=rss

LinkedIn’s latest premium perk is an AI job coach

Posted on by

LinkedIn is adding a new, AI-powered perk for its premium subscribers: a built-in job coach that uses AI and LinkedIn data to help job seekers find, research and apply for roles. The new feature arrives as the company announced its user base has grown to 1 billion members as it looks to ramp up its investment in AI-driven features.

The Microsoft-owned company has increasingly been experimenting with AI features for its paying members. Earlier this year, it introduced the ability to use generative AI to write better profile descriptions and messages to hiring managers. But the latest AI perks aim to provide an even more personalized experience.

For now, the most prominent feature for job seekers will be AI-generated insights alongside each job posting. The tool can help summarize lengthy job descriptions and weigh in on whether the role is a good fit for a user based on the contents of their LinkedIn profile. For example, it can highlight specific work experiences users’ may want to emphasize in their application and provide tips on how to improve their LinkedIn profile to look more attractive to hiring managers.

LinkedIn's new AI job coach.
LinkedIn

Because LinkedIn is able to draw on its vast trove of career data, the tips it’s able to provide are much more personalized than what you’d likely get if you were to ask other generative AI services for tips, says LinkedIn product manager Rohan Rajiv. “This is made possible by generative AI, but also the datasets that bring all of this together,” Rajiv tells Engadget. “It's your profile, your connections, and all of this that essentially can help you move your job search forward.”

For now, it’s still early days for the feature which is launching in beta to a limited set of LinkedIn Premium subscribers. But the company has signaled it intends to make AI a central part of its service going forward. “Today marks the beginning of a new journey, one where the power of AI is your ally in every career question and decision,” LinkedIn’s Chief Product Officer, Tomer Cohen, wrote in a blog post.

This article originally appeared on Engadget at https://www.engadget.com/linkedins-latest-premium-perk-is-an-ai-job-coach-120044855.html?src=rss

A software company called Threads says Meta tried to buy its domain and kicked it off Facebook

Posted on by

A UK-based software company called Threads Software Limited is threatening legal action against Meta over its use of the name Threads. The company, which says it’s owned the “Threads” trademark since 2012, makes an “intelligent message hub” that uses AI to help businesses keep track of phone calls, emails and other messages.

Threads Software Limited claims that Meta lawyers made four separate attempts, beginning in April 2023, to buy the software company’s threads.app domain, and eventually shut down its Facebook account. “Every offer was declined,” the company said in a statement. “It was made clear to Meta’s Instagram that the domain was not for sale. In July 2023, Meta’s Instagram announced its ‘threads’ social media platform and removed Threads Software Limited from its Facebook platform.”

The software company said that it’s giving Meta 30 days to “stop using the Threads name” and that it will “seek an injunction from the UK courts” if the social media company declines to do so. In a statement, Threads Software’s CEO John Yardley said it was “not an easy decision” to take on Meta, but that the “business now faces a serious threat from one of the largest technology companies in the world.”

Meta didn’t immediately respond to a request for comment. It’s not clear how much money Meta may have offered for threads.app, but Yardley’s statements offer a rare look at the kind of backroom negotiations that can happen in order to secure a sought-after domain or username.

It’s also worth noting that the software maker wasn’t the only company using the Threads name at the time Meta launched its Twitter competitor. Fashion retailer American Threads controlled the @Threads handle on Instagram at the time of the service’s launch. The company jokingly responded to commenters at the time, and posted on the new Threads service about people mixing up the clothing brand with the Meta-owned service. Meta used @threadsapp on Instagram and threadsapp.net on Threads, at the time of the service’s launch.

A retailer called American Threads owned the Threads handle on Instagram when Threads first launched in July.
Screenshot by Karissa Bell via Instagram

A month later, the retailer’s Instagram account changed handles to @americanthreads (and americanthreads.net on Threads) without explanation, while Meta took control of the @Threads handle. Representatives for the clothing brand didn’t immediately respond to a request for comment, but the circumstances are strikingly similar to how Meta quietly gained control of the @Meta handle on Instagram more than a year ago. That username was also controlled by a separate entity — an independent motorcycle publication called META — but the account was later subsumed by the social network without explanation.

Representatives of Meta, the magazine, never commented directly on how their account changed hands, but wrote about their dismay in learning of Facebook's name change. “With the flip of a switch our identity was suddenly watered down, and we watched our name circle the drain and wash away with something we had no control over,” the magazine’s cofounder wrote in a blog post that’s since been deleted. The magazine now uses the name Vahna.

For now, it appears Threads Software Limited is hoping for a different outcome. “Over the last 10 years, we have made a large investment in the Threads name and we did not want to potentially have to write-off this investment simply because Meta happened to like the name we had already coined for a messaging service,” it wrote in a blog post. “For us to change the service name simply to avoid confusion with Meta’s product could well set back the service enough for us to lose that technological lead.”

If you have been offered money in exchange for your domain name or handle from Meta or another social media company, reach out to me at karissa.bell [at] engadget.com or on Signal at +1.628.231.0063.

This article originally appeared on Engadget at https://www.engadget.com/a-software-company-called-threads-says-meta-tried-to-buy-its-domain-and-kicked-it-off-facebook-221928864.html?src=rss

X won’t pay creators for tweets that get fact checked with community notes

Posted on by

X will no longer pay creators for tweets promoting misinformation. Elon Musk said the company is making a “slight change” to its monetization program and that tweets that are fact-checked via community notes will no longer be eligible for payouts as part of X’s revenue-sharing program.

The update appears to be an attempt to remove incentives for high-profile accounts to spread viral misinformation. “The idea is to maximize the incentive for accuracy over sensationalism,” Musk said. X also recently started to require community notes contributors to cite their sources in fact checks.

The latest change comes as researchers, fact checkers and journalists have raised the alarm about the amount of viral misinformation spreading on X amid the ongoing conflict in Israel and Gaza. European Union officials have opened an investigation into the company’s handling of misinformation related to the war.

Following Musk’s takeover of Twitter a year ago, the company laid off teams responsible for curating and promoting reputable tweets about breaking news events and removed tools for reporting misinformation in the app. Instead, the company has relied on its crowd-sourced fact checking tool, community notes.

But critics have said that community notes are subject to manipulation and that the user-contributed fact checks are often unable to keep up with the sheer amount of viral falsehoods, particularly those promoted by verified accounts. A recent analysis from NewsGuard, a nonprofit that tracks the spread of misinformation, found that 74 percent of “the most viral posts on X advancing misinformation about the Israel-Hamas War are being pushed by ‘verified’ X accounts.”

As BBC researcher Shayan Sardarizadeh pointed out, the change to make tweets with community notes ineligible for payments has already been criticized by a number of high-profile accounts whose tweets are often “community noted.” Musk added that “any attempts to weaponize @CommunityNotes to demonetize people will be immediately obvious,” but didn’t say how the company would deal with attempts at manipulation. X didn’t respond to a request for comment.

This article originally appeared on Engadget at https://www.engadget.com/x-wont-pay-creators-for-tweets-that-get-fact-checked-with-community-notes-174206477.html?src=rss