Bluesky has rolled out an update that fixes one important issue that could lead to impersonation on the decentralized social network. Now, when you verify your identity on the platform with your own domain, Bluesky will no longer free up your old .bsky.social username. In the past, going through the authentication process will make your original .bsky.social name available again, which means you'll have to sign up again to secure your old handle if you don't want impersonators to scoop it up. 

The social network has been trying to fix its verification issues, which became a significant problem after the service welcomed an influx of new users in recent months. A third-party entity from Cornell Tech who analyzed the app's userbase previously found that 44 percent of Bluesky's 100 most-followed accounts have a doppelganger. As a response, Bluesky adopted a more aggressive impersonation policy and required parody, satire or fan accounts to label themselves as such in both their handles and their bio in late November. It also explicitly prohibited identity churning, in which users would start off as impersonators to gain followers and then switch their identity later to avoid enforcement action.

In addition to reserving your old handle, the updated Bluesky app adds a "Mentions" tab in notifications, making it much easier to find replies to your posts. As The Verge notes, it also has a new button that lets you choose how replies show up, so you can choose to see them in a linear order or in threaded discussions. 

This article originally appeared on Engadget at https://www.engadget.com/social-media/blueskys-latest-update-addresses-an-important-verification-problem-140055367.html?src=rss

©

© NurPhoto via Getty Images

Clickbait videos have always been annoying, but there are times when they can be downright harmful. YouTube has vowed to strengthen its enforcement efforts when it comes to dealing with "egregious clickbait" on its website, particularly those that cover — or pretend to cover — breaking news and current events. The website describes egregious clickbait as "videos where the title or thumbnail promises viewers something that the video doesn't deliver."

YouTube says these videos leave viewers "feeling tricked, frustrated, or even misled" if they come to the website looking for truthful and timely information on important issues. If you've ever watched a clickbait video, you'd know that's definitely true. You may have trained yourself on being able to spot and skip them over the years, but some people might still not know the difference between clickbait and legitimate content. 

One example of egregious clickbait, according to YouTube, is if a video says "the president resigned!" without actually addressing the president's resignation. Misleading thumbails are considered egregious clickbait, as well. If a thumbnail reads "top political news" and the video doesn't contain any political news, then it will also be subjected to YouTube's enforcement action. 

The website will start cracking down on clickbait videos in India — it didn't say how it will expand from there, but we've asked it for more information. For now, it will delete any video that violates this policy without issuing strikes. After it goes through old videos, it will then prioritize new uploads, presumably so that they don't reach more people that they should. 

This article originally appeared on Engadget at https://www.engadget.com/entertainment/youtube/youtube-will-crack-down-on-egregious-clickbait-starting-in-india-130010064.html?src=rss

©

© FreshSplash via Getty Images

Another day, another new Threads feature. Meta’s app will “soon” be rolling out a feature that allows users to re-share photos and videos to their timelines with credit to the original poster but without the original post attached.

Once available, users will need to long press on a photo or video from their feed and select “use media.” They’ll then be able to draft a new post with the image of the clip attached, with a watermark from the original poster showing in the top left corner. “This is a quick, easy way to add your creative takes to trending images and clips without quote posting,” Threads chief Adam Mosseri wrote in a post.

The feature is presumably meant to make it easier for people to reshare other users’ work with some credit for its original creator. Mosseri has fielded a number of complaints on Threads from frustrated users who see their work ripped off without credit. In a post last month, Mosseri said that Meta was “trying to shift more distribution from aggregators to creators on both Instagram and Threads,” though he acknowledged it can be “very difficult to do so” when the original post was lifted from a non Meta-owned platform. On Instagram, the company recently tweaked its algorithm in an attempt to boost creators over aggregators.

Judging by some of the reactions to Threads’ latest update, however, a number of creators are still unhappy with Meta’s approach. “It just seems like blatant content theft,” one photographer replied. “Quoting the post is a far more effective way of sharing someone's content with your own comments while crediting the creator,” another user said.

In a separate post, Meta noted that individual creators are able to disable media reuse in their app’s settings, so it is possible to block your posts from being shared this way. (There are also settings to prevent others from quote posting entirely.) However, for those worried about accounts straight up ripping off their posts in the interest of engagement farming, there’s still little any user can do to prevent less scrupulous accounts from copying their content with screenshots or other means.

This article originally appeared on Engadget at https://www.engadget.com/social-media/threads-new-use-media-feature-is-its-latest-attempt-to-fight-stolen-memes-and-engagement-farming-224119483.html?src=rss

©

© Meta

Grubhub has agreed to pay $25 million to settle charges from the Federal Trade Commission (FTC) and the Illinois Attorney General. The company was accused of a laundry list of sketchy behavior, including misleading customers about delivery charges, deceiving delivery drivers about income and listing restaurants on the platform without consent. Last month, the food delivery startup Wonder bought Grubhub for one-tenth of what it was worth during the pandemic.

Under the proposed settlement, Grubhub has to make changes to remedy the problems. The requirements read like a “stop doing that” list, one per charge. This includes notifying customers of full delivery costs, being honest with drivers about pay and listing restaurants only with their consent.

The FTC says Grubhub, to appear more robust than it was, added as many as 325,000 unaffiliated restaurants to the platform without permission since at least 2019. Customers ordering from those businesses discovered added fees and “numerous ordering problems.” Meanwhile, the agency says the restaurants “bore the brunt of diners’ ire,” leading to damaged reputations and lost money.

The company also allegedly added junk fees after advertising to customers that they’d pay a low-cost, flat rate for deliveries. The FTC says Grubhub labeled them “service fees” or “small order fees,” but they were simply delivery fees under another name. The agency quotes a former Grubhub executive as calling it a “pricing shell game.”

The FTC also accused the company of blocking customers’ accounts with large gift card balances, leaving them no way to regain access. The agency said diners who complained to the company either weren’t told their accounts were blocked or weren’t given any meaningful way to contest the ban.

The false pay allegations include advertising that Grubhub drivers could make up to $40 hourly in the New York area. In reality, the median driver pay in that area was around $10 hourly — and only 0.1 percent of drivers are said to have made the advertised rate. And in Chicago, an ad promised earnings of up to $26 hourly when the median was $11.

Grubhub denies the allegations but says it settled to put the matter behind it. “At Grubhub, we’re committed to transparency so that every single day diners, restaurants and drivers can make well-informed choices to do business with us,” the company wrote in a statement. “While we categorically deny the allegations made by the FTC, many of which are wrong, misleading or no longer applicable to our business, we believe settling this matter is in the best interest of Grubhub and allows us to move forward.”

“Our investigation found that Grubhub tricked its customers, deceived its drivers, and unfairly damaged the reputation and revenues of restaurants that did not partner with Grubhub — all in order to drive scale and accelerate growth,” FTC Chair Lina M. Khan wrote in a statement. “Today’s action holds Grubhub to account, putting an end to these illegal practices and securing nearly $25 million for the people cheated by Grubhub’s tactics. There is no ‘gig platform’ exemption to the laws on the books.”

This article originally appeared on Engadget at https://www.engadget.com/big-tech/grubhub-settles-with-the-ftc-over-adding-restaurants-without-their-consent-221006330.html?src=rss

©

© Grubhub

Instagram is adding an option to schedule DMs. Social media expert Lindsey Gamble unearthed the feature, and Instagram confirmed to TechCrunch that it's rolling out scheduled DMs to all users.

When you type a message, simply hold down the send button and you can select a date and time. It seems messages can be scheduled up to 29 days in advance. Until all timed-up messages are sent, you'll see a banner reading something like "x scheduled messages."

This will be handy for folks who want to schedule birthday messages for a bunch of friends at once or, for instance, to remind someone to pick them up from the airport on a certain day. It'll also be useful for people who tend to take care of correspondence at night and don't want everyone to know how late they're staying awake. That's definitely not something I ever do with emails.

It's worth noting that Instagram is rolling out this DM scheduling feature before all users are able to time up posts and Reels in advance. For now, that feature is limited to folks who have set up a professional account.

Meanwhile, Instagram is rolling out several limited-time, end-of-year features to help you celebrate the holidays and your 2024 memories. For one thing, there's a collage tool for Stories that has an end-of-year theme. Based on images Instagram shared, it appears that you can go with a Happy New Year overlay.

There are multiple Add Yours templates based around New Year's as well, such as one you can use to prompt friends to share photos in the how 2024 started/how 2024 ended format. If you hit the like button on end-of-year Stories, you'll see a custom effect. There's a New Year font and Countdown text effect for Stories, Reels and feed posts as well.

Festive chat themes for the holidays include New Year's, one called "chill" and, of course, another based on Mariah Carey. Last but not least, if you use certain emoji based around celebrations or phrases like "Happy New Year" or "hello 2025" in DMs or notes before the end of the year, you'll see a little Easter egg of some kind.

This article originally appeared on Engadget at https://www.engadget.com/social-media/instagram-is-adding-a-dm-scheduling-feature-before-everyone-can-schedule-posts-203957229.html?src=rss

©

© Instagram

Snap is changing up its program that allows creators to make money from shortform videos. The company announced a new monetization program that will allow the app’s influencers to make money from Spotlight videos that are one minute or longer by earning a share of their content's ad revenue.

The change will streamline Snap’s monetization features across Spotlight, its in-app TikTok competitor, and Stories, where Snap first launched its revenue sharing feature. It also means the company will end its Spotlight Reward Program, the creator fund-like arrangement that paid creators directly. That program will be discontinued January 30, 2025, with the new monetization arrangement taking effect February 1.

Snap announced the update as TikTok moves closer to an outright ban in the United States. The ByteDance-owned service is currently facing a January 19, 2025, deadline to sell or be banned f the Supreme Court doesn’t intervene. In its announcement, Snap notes that Spotlight viewership is “up 25% year-over-year” and that “there is a unique and growing opportunity for creators to monetize this format in the same way they do with Stories.”

Under the new “unified” program, creators are eligible to earn money from Spotlight videos or Stories if they meet the following requirements:

-Have at least 50,000 followers.

-Post at least 25 times per month to Saved Stories or Spotlight.

-Post to either Spotlight or Public Stories on at least 10 of the last 28 days.

-Achieve one of the following in the last 28 days:

    -10 million Snap views

    -1 million Spotlight views

    -12,000 hours of view time

Some of those metrics are a bit higher than Snap’s previous requirements for Stories, which set the bar at only 10 Story posts a month. But, as TechCrunch notes, the new threshold is much higher for Spotlight creators, who could previously earn money from the company’s creator fund with only 1,000 followers and 10,000 unique views. The change also pushes creators to make longer content for Spotlight as they can no longer be paid for videos shorter than one minute. 

If TikTok does end up being banned, Snap will be one of several platforms trying to lure creators to its product. And while the app is known primarily for its private messaging features, the company says that the number of people posting publicly has “more than tripled” in the last year, and that it will be “evolving and expanding the total rewards available to creators” going forward.

This article originally appeared on Engadget at https://www.engadget.com/social-media/snap-will-expand-ad-revenue-sharing-to-creators-on-spotlight-193029473.html?src=rss

©

© Ramsey Cardy via Getty Images

Threads is rolling out its take on the Starter Packs concept from Bluesky. Instagram CEO Adam Mosseri's post about the update explains that collections of recommended profiles will be suggested when users start following topics and as part of the For You feed.

Interestingly, Threads doesn't seem to have a name for this feature. But it's not the first time that Threads has adopted an idea that originated at rival platform Bluesky. Custom feeds were the last concept that made its way from Bluesky to a brief testing phase to an official release on the Meta-owned social network.

Bluesky is aware of how many of its features have started popping up on Threads. Last month, when Meta's platform first announced that it would have its own take on Starter Packs, the company posted a dig at its competitor on its own Threads account. The pair of platforms have been vying to draw in disaffected former users of X (once known as Twitter). Bluesky reached 20 million users in November, while Threads reported 35 million signups just in that month.

This article originally appeared on Engadget at https://www.engadget.com/social-media/threads-take-on-bluesky-starter-packs-is-live-193106158.html?src=rss

©

© Meta

BeReal, the in the moment social media platform, is far from its 2022 heyday, but that hasn't stopped one organization from going after it. Austrian advocacy group Noyb has filed a complaint surrounding the platform's data consent banner practices. The organization claims that the banner disappears if users accept that their personal data can inform advertising practices, but if they click reject then the banner appears daily. Noyb filed its complaint with the French data protection authority (CNIL) as Voodoo, a French company, bought BeReal in June — the practice in question started in July. 

"BeReal’s daily attempt to pressure its users into accepting the tracking for personalised advertising has a significant impact on user behaviour. Consent given under these circumstances is not freely given, which means it doesn’t meet the requirements established in Article 4(11) GDPR," Noyb argued in its complaint. It asked the CNIL to fine BeReal and force it to be compliant. 

Noyb also pointed to guidelines the European Data Protection Board (EDPB) adopted in early 2023 around such a technique. "The continuous prompting deceptive design pattern occurs when users are pushed to provide more personal data than necessary for the purposes of processing or to agree with another use of their data, by being repeatedly asked to provide additional data or to consent to a purpose of processing," the EDPB stated. "Such repetitive prompts can happen through one or several devices. Users are likely to end up giving in, as they are wearied from having to refuse the request each time they use the platform. "

This article originally appeared on Engadget at https://www.engadget.com/big-tech/bereal-accused-of-annoying-users-into-sharing-their-data-160024570.html?src=rss

©

© picture alliance via Getty Images

Instagram, Facebook, Threads and Messenger are coming back online after widespread "technical issues" took down many of Meta's biggest apps for several hours Wednesday. "We’re 99% of the way there - just doing some last checks," the company wrote in an update on X nearly four hours after first acknowledging the outages. 

The company didn't elaborate on the source of the issue, or when the problems might be resolved entirely. As of 4:30 PM PT Wednesday, Meta's status dashboard still showed some "major disruptions" to its business and transparency tools, though other services were marked as "resolved" or "recovering." Facebook, Instagram, WhatsApp, Threads and Messenger seem to have largely recovered, though.

Earlier in the day, there were over 90,000 reports of issues for Facebook.com alone on Downdetector, as users across X and Bluesky reported problems loading and using Meta’s apps. For a while even Meta’s company site was displaying the text “This page isn’t available right now.” When Engadget first reached Meta for comment, the company pointed to a brief statement on X acknowledging that "a technical issue is impacting some users’ ability to access our apps."

Meta last dealt with a major outage in March 2024 that prevented users from accessing its apps and services for two hours. That outage was attributed to a "technical issue" by Meta communications director Andy Stone, which is the same explanation Meta has offered so far today. 

Update, December 11, 2024, 4:50PM ET: Added more details on outage length and impact on Meta enterprise tools.

Update, December11, 2024, 7:50 PM ET: Additional comments from Meta about the status of the outage were added.

This article originally appeared on Engadget at https://www.engadget.com/social-media/instagram-threads-whatsapp-and-more-down-as-part-of-meta-outage-184608535.html?src=rss

©

© Reuters / Reuters

Meta is expanding its feature that helps users avoid “Facebook jail.” Earlier this year, the company began allowing Facebook creators to complete in-app “educational training” for first-time rule violations in order to avoid strikes on their accounts. Now, the company is expanding the feature to all Facebook users and opening it up to creators on Instagram.

As Engadget noted back in August, the idea of the feature is a bit like going to traffic school. People who violate one of the company’s rules for the first time will have the option to complete a “a short educational program” in Facebook or Instagram’s app in order to avoid a “strike” on their account and any restrictions that may come with it.

Users will be able to take advantage of the process once in a 12-month period for most first-time offenses. The process to “remove the warning” on an account will be available to anyone on Facebook, as well as creators on Instagram (a broader rollout for Instagram is expected “soon,” according to the company.) Meta won’t allow people to remove strikes for more “serious” offenses, including posts about sexual exploitation or drug sales.

Meta has described the strike-removal feature as part of its effort to reform its penalty system, which has long been frustrating and confusing for users who often get caught up in it unintentionally. In a blog post, Meta said that the in-app education features have already shown some signs of success with Facebook creators. “What we’ve already seen from our initial launch this summer has been promising — those who successfully removed their first strikes for violating a policy were more likely to say they understood Facebook’s policy decisions and became less likely to violate that policy again,” the company said.

This article originally appeared on Engadget at https://www.engadget.com/social-media/meta-expands-its-strike-removal-feature-to-instagram-and-all-facebook-users-170056830.html?src=rss

©

© Meta

It’s that festive time of year again. All of the algorithm elves living inside our favorite music-streaming apps have begun cranking out personalized “best of” lists in the vein of Spotify Wrapped and Apple Music Replay. This year, however, there’s a new kid on the block. Amazon Music just rolled out its own version of a yearly recap.

It’s called Amazon Music Delivered, which I guess is a play on the fact that the company delivers lots of packages in real life? In any event, the yearly recap displays all of your favorite songs, artists and podcasts from throughout the year. The tool will even show off “the top request you’ve made with Alexa.” Uh oh. I asked Alexa to sing that horrible, yet catchy, song about cats like 25 times this year. Will that count?

The feature is available to all Amazon Music customers. This includes Prime members who use the basic version and those who sprang for the monthly Unlimited subscription. Just head to the Library page and tap on the banner that says “2024 Delivered.” You can access a playlist by asking good ole Alexa about your top songs of 2024.

Amazon also says that Alexa might play “an exclusive, special message” from one of your top artists, though the company didn’t note how many artists took part in this promotion. It just says that “eligible customers will see a flashing yellow light on their Alexa devices” if a message is waiting.

Amazon

Just like the other streamers, Amazon Music has also revealed its most popular artists and songs of the year. Taylor Swift took the top artist spot while “Beautiful Things” by Benson Boone snagged the top song prize.

The various Amazon Music tiers can be a bit confusing, as with most of the company’s digital offerings. Prime members automatically get something called Amazon Music Prime, which allows access to a catalog of 100 million songs and podcasts. There are no ads for this tier, but members are tied to playlists and shuffles, which is kind of like how the free Spotify tier works. 

Amazon Music Unlimited removes all restrictions, but costs $10 per month for Prime members and $11 for non-members. Finally, there’s Amazon Music with Ads, which is an ad-filled version of the standard Prime-based tier. It doesn’t cost anything and is available to everyone, even people not interested in free and expedited shipping.

This article originally appeared on Engadget at https://www.engadget.com/entertainment/music/amazon-music-now-has-its-own-version-of-spotify-wrapped-162216481.html?src=rss

©

© Amazon

Elon Musk’s attorneys filed for an injunction against OpenAI and Microsoft on Friday, accusing them of anticompetitive practices. He wants to stop OpenAI’s conversion to a for-profit company. Musk first sued OpenAI earlier this year for allegedly violating its founding mission of building AI “for the benefit of humanity,” but he withdrew the lawsuit a few months later. He filed another lawsuit against OpenAI in a California federal court in August.

The third time’s the charm and all: Musk’s new motion accuses OpenAI and Microsoft of telling investors not to fund OpenAI’s competitors, such as Musk’s xAI, of “benefitting from wrongfully obtained competitively sensitive information or coordination” through its relationship with Microsoft.

— Mat Smith

The biggest tech stories you missed

• What we’re listening to: Celeste Secret Hideaway and The Loop

• The best Cyber Monday 2024 tech deals worth your money

• The best Christmas gifts for everyone on your 2024 holiday list

Get this delivered daily direct to your inbox. Subscribe right here!

Intel Arc Battlemage GPU specs may have leaked early

The company teases ‘big graphics news’ this week.

Intel

Intel is gearing up for an announcement on December 3 about its Arc GPUs, and whoops, leaks are spoiling the party. Reports from VideoCardz claim the event will reveal two Battlemage desktop GPUs — the Arc B580 and Arc B570 — that’ll launch on December 12. The B580 will allegedly have 20 Xe2 cores, a 2.8GHz GPU clock and 12GB of VRAM. The B570, on the other hand, will reportedly feature 18 Xe2 cores, a 2.6GHz GPU clock and 10GB of memory. Elsewhere, leaker @momomo_us on X posted listings that suggest the limited edition version of the B580 will be around $250.

Continue reading.

Bluesky implements a ‘more aggressive’ policy against impersonation

Parody and fan accounts will have to be clearly labeled.

Bluesky has updated its impersonation policy to be “more aggressive” after third-party analysis highlighted its verification problem. The Bluesky Safety account said the social media service is removing accounts impersonating other people and those squatting on handles. Bluesky doesn’t have a conventional verification system, so it’s easy for unscrupulous users to pretend to be someone else, either for attention or to scam others. Bluesky now explicitly prohibits identity churning, as well. Accounts that start as impersonators to gain new users, then switch to a different identity to circumvent the ban will still get booted off the app.

Continue reading.

This article originally appeared on Engadget at https://www.engadget.com/general/the-morning-after-engadget-newsletter-121535788.html?src=rss

©

© OpenAI / Will Shanklin for Engadget

Riot Games added some new clauses to its Terms of Service that could put some players in hot water for unbecoming behavior that occurs “across the various places that touch their gaming experience.” Players can face penalties, suspensions and even “Riot-wide bans” if they are caught violating these new rules.

The new clauses first appeared on Riot’s Privacy Notice and TOS on Tuesday. Creators have until January 3, 2025 “to adjust their content to adhere to these rules,” according to Riot Games’ announcement.

The new “Off-Platform conduct” rule gives Riot Games the right to issue game penalties when hateful content is brought to their attention. The new clause says Riot Games won’t proactively search for social media for violations. Instead, players can report cases about an offending player’s conduct on a livestream in which one of its games serves as “the background of the content produced.” If Riot determines that the player violated its TOS, Riot could issue a penalty “as if that behavior occurred in-game,” according to the TOS.

The new TOS rules also focus on “stream sniping” in which streamers and players hijack or interrupt another stream’s online sessions. The penalty system for these infractions are still in the “early testing phases” using a limited number of English-speaking Valorant creators.

Repeat offenders of these and other infractrations could result in complete bans from all Riot Games accounts. The new TOS states that Riot can issue multiple penalties and bans for streamers and players who commit “a seriously egregious violation” or “problematic behavior.”

Riot Games has dedicated a lot of time recently to discussing and exploring ways to rein in offensive and ill behavior in its games. Valorant studio head and Riot Games Senior Vice President Anna Donlon acknowledged in May that Riot “absolutely needs to do better” to protect its players laying out new actions they planned to take such as implementing new penalties and increasing support for manual reviews of reports, according to Eurogamer.

Twitch has also taken steps to curb some of its users’ disturbing and damaging behavior. Last year, Twitch CEO Dan Clancy announced new terms for the streaming platform during the opening ceremony for TwitchCon in Las Vegas. The new rules established suspensions and bans for streamers who dox or swat other players and the ability for streamers and moderators to anonymously send warnings to chatters who commit bad behavior during sessions.

This article originally appeared on Engadget at https://www.engadget.com/gaming/riot-games-is-cracking-down-on-players-off-platform-conduct-173303058.html?src=rss

©

© Riot Games

Australia is set to ban under 16s from social media services after the Senate passed a bill to that effect by 34 votes to 19. The legislation will return to the House of Representatives, which will need to approve amendments before it becomes law. That is all but a formality as the government holds a majority in that chamber. The bill, which has been fast-tracked, sailed through the lower house in a 102-13 vote earlier this week.

The government has said that the likes of Snapchat, TikTok, Instagram and X will be subject to the new rules, which won’t come into force for at least 12 months. However, officials still have to confirm which platforms the ban actually covers as they aren't detailed in the bill. The BBC notes that the country’s communications commissioner, Michelle Rowland, will determine that with help from a so-called eSafety Commissioner. The latter will be responsible for enforcing the law.

The rules will not apply to health and education services, gaming platforms or messaging apps, nor those that don’t require an account. So, the likes of Fortnite, Roblox and YouTube are likely to avoid any ban.

Companies that are subject to the legislation could face fines of up to $49.5 million AUD ($32.1 million) if they fail to comply. They will have to employ age-verification tech, though the specifics of that have yet to be determined. The government plans to assess various options in the coming months, but Rowland confirmed this week that platforms won't be able to compel users to submit a personal document (such as passport or driver license) to verify their age.

Researchers have claimed that mooted age-verification systems may not work in practice. Critics, meanwhile, have raised concerns over privacy protections.

While there are certainly valid concerns about the harms of social media, such platforms can be a lifeline for younger people when they’re used responsibly. They can help vulnerable kids find resources and peers they can turn to for advice. Social media can also help those in rural areas forge authentic social connections with others who live elsewhere.

Under 16s who continue to access banned platforms won’t be punished. Resourceful teens may find it very easy to bypass restrictions using a VPN, which could make the law largely toothless. The online world also extends far beyond the reach of a small number of centralized social media platforms. There are other pockets of the internet that teens can turn to instead. For instance, there are still a large number of active forums for various interests.

When the legislation becomes law, Australia will set the highest minimum age for social media of any jurisdiction. France has tabled legislation to block users under 15 from social media without parental consent and it’s now pushing for the European Union to move forward with a similar undertaking across the entire bloc. Norway plans to bring in legislation along those lines, while the UK's technology secretary recently indicated that it was an option for that country.

Utah last year passed laws to limit minors' social media use. The state's governor repealed and replaced those earlier this year following legal challenges. However, in September, a judge blocked the most recent legislation just days before it was set to take effect. Other states have considered similar laws.

This article originally appeared on Engadget at https://www.engadget.com/big-tech/australia-is-one-step-away-from-banning-social-media-for-under-16s-160454882.html?src=rss

©

© REUTERS / Reuters

Bluesky is bigger than ever. But as the upstart social media service surges, the platform is facing some growing pains. Among them: The influx of new users has opened up new opportunities for scammers and impersonators hoping to capitalize on the attention — and Bluesky’s lack of a conventional verification system.

A recent analysis by Alexios Mantzarlis, director of the Security Trust and Safety Initiative at Cornell Tech found that 44 percent of the top 100 most-followed accounts on Bluesky had at least one “doppelganger,” with most looking like “cheap knock-offs of the bigger account, down to the same bio and profile picture,” Mantzarlis wrote in his newsletter Faked Up.

Unlike many of its counterparts, which offer checkmarks and official badges to government officials, celebrities and other high profile accounts, Bluesky has a more hands-off approach to verification. Instead of proactively verifying notable accounts itself, the company encourages users to use a custom domain name as their handle in order to “self-verify.”

For example, my employer Engadget currently has the Bluesky handle engadget.bsky.social. But if we wanted to “verify” our account, we could opt to change it to Engadget.com. Some media organizations, like The New York Times, Bloomberg and The Onion have done this for their official accounts. Individuals are also able to verify by using a personal website.

But, the process is more complicated than simply changing your handle. It also requires entities to add a string of text to the DNS record associated with their domain. While in some ways it’s a clever solution to verification — only the actual owner of a website would be able to access the DNS record for a domain — it also has a number of drawbacks. It’s a manual process that’s not readily accessible to everyone who might wish to be verified. (Bluesky does sell custom domains for users who don’t already have one.)

Verification is even more complex for those wishing to verify multiple accounts associated with the same domain, which may explain why some outlets, like The New York Times and NPR have custom handles, but don’t extend that verification to their reporters on Bluesky. Even Bluesky’s own tutorial suggests organizations seek assistance from their IT departments.

There are other issues. Once you change your handle to match a domain you own, your old alias (engadget.bksy.social, for instance) becomes available again. So you’ll either need to set up a new account to “squat” on your old handle, or risk an impersonator scooping it up. And even if you add a custom domain, it doesn’t offer foolproof protection against impersonation. A dedicated scammer could use a lookalike domain and “verify” an imposter account.

Screenshot via Bluesky

To make things more confusing, Bluesky itself gives no indication, other than the handle name, that an account has been “verified.” Verified accounts don’t have a visual indicator — like a check or a badge — that differentiates them from unverified ones,

To combat this, some Bluesky users are coming up with their own makeshift workarounds. Hunter Walker, an investigative reporter for Talking Points Memo and early Bluesky user, has been proactively verifying journalists, celebrities and other high-profile accounts himself. So far, he’s verified more than 330 people, including New York Representative Alexandria Ocasio-Cortez, Flavor Flav, Mark Cuban and Barbra Streisand.

“I have a pretty high standard for journalism and reporting, and everything I say, I like to triple check the sources,” Walker tells Engadget. “I like to make sure it's confirmed. And it became apparent to me, participating in Bluesky, that on a basic level, nothing was confirmed.”

Walker estimates he’s spent about 16 hours over the last couple weeks verifying accounts. He has different methods depending on the user, but it often involves communicating with someone from another account officially linked to them, like a company email address. For celebrities, their representatives are often able to confirm their official Bluesky handles.

“I’ve caught so many scammers and imposters, and it's not always who you would expect,” Walker says. “Regular journalists sometimes have three or four imposters.” He says he’s been inundated with requests for his unofficial verification, and notes that a number of people he’s verified also use a custom domain. “They want something else … because a domain is not verification of identity.”

Walker maintains “starter packs” of journalists and other prominent accounts he’s verified. Recently, he took it a step further, working with another user to create a custom labeling service that will append different emojis to accounts he’s verified to make his “verification” more prominent. Users who subscribe to the service will see a 😎 next to celebrities and public figures, and a 🌐 next to journalists.

Screenshot via Bluesky

While these kinds of efforts can act as a stopgap, Walker won’t be able to verify every notable account on Bluesky himself. He’s suggested that other communities, like university researchers, could undertake a similar ad hoc verification effort. But, without help from Bluesky or a third-party identity service, he expects impersonation to remain an issue.

And widespread impersonation can often lead to bigger problems for a platform like Bluesky. “Sloppy verification is an early signal of broader deception and catnip for organized disinformation actors,” Cornell Tech’s Mantzarlis wrote, noting that Vice President Kamala Harris “at one point had 20 impersonator accounts” on Bluesky even though she’s never had an official presence on the platform.

On its part, Bluesky has acknowledged that impersonation is an issue. In an update this week, the company said it had seen “a predictable uptick in harmful content” that coincided with its recent growth. In a statement to Engadget, Bluesky spokesperson Emily Liu said the company had “quadrupled” its moderation team, which would help ensure reports of impersonation are handled more quickly. Liu also said that Bluesky was working on “easier visual signals we could use for verification so it's a better user experience,” though it’s not yet clear what form that might take.

But Bluesky, which currently has just 20 full-time employees, seems reluctant to consider other approaches to verification outside of custom domains. “We've been working behind the scenes with official organizations and high-profile individuals like celebrities and elected officials to get their accounts verified on Bluesky with their website,” Liu said. “With domains as verification, we want to put the tools of verification in each org's hands, instead of making Bluesky the company the sole arbiter of who deserves to be verified on the network.”

Bluesky’s hesitation to play the role of verifier is in many ways understandable. Verification has a long and messy history on other platforms. On Twitter, a symbol that was originally created to fight impersonators quickly morphed into a sometimes divisive status symbol. On Instagram, verification has often been exploited by scammers. Now, both companies allow users to buy blue checkmarks, though both platforms also proactively verify certain types of accounts, like those belonging to government officials.

Bluesky CEO Jay Graber, however, has signaled that she’s potentially open to alternate approaches to verification. In a livestream on Twitch this week, she said the company “might at some point” become a “verification provider.” TechCrunch, which reported the remarks, said that her comments suggested a future system in which there are multiple “providers” of verification. Graber added that she’s “not sure when” such a scenario would play out.

Walker, who repeated several times his firm belief that “Bluesky has the juice,” hopes that his verification project might be able to nudge Bluesky to take a different approach. “I'm really hoping that people pay attention to the question of trust and the question of identity. The cool thing about the open source nature of it all, is we have a chance to build things on this and make it how we want it.”

This article originally appeared on Engadget at https://www.engadget.com/social-media/bluesky-has-a-verification-problem-190047733.html?src=rss

©

© Carl Court via Getty Images

Earlier this year OpenAI unveiled Sora, a text-to-video AI model, showing off detailed scenes and complex camera motion from relatively simple prompts. It's been radio silence since then, but the company recently granted artists free early access to the tool for testing. However, a group of around 20 of those just leaked access to Sora in protest, saying they were acting as "PR puppets," prompting OpenAI to suspend access, The Washington Post reported. 

"We received access to Sora with the promise to be early testers, red teamers and creative partners. However, we believe instead we are being lured into 'art washing' to tell the world that Sora is a useful tool for artists," the group wrote on the AI art repository site, Hugging Face.

Pointing out that OpenAI recently hit a $150 billion evaluation, they noted that artists have been providing unpaid testing and feedback. The group also noted that all Sora-generated content needed to be approved by OpenAI, making it "less about creative expression... and more about PR and advertisement." The group then said it had released the tool to let anyone play with it, saying it hopes that OpenAI will "support the arts beyond PR stunts." 

In response, OpenAI shut down early Sora access after just three hours while it looks into the situation. "Hundreds of artists in our alpha have shaped Sora’s development, helping prioritize new features and safeguards,” OpenAI spokesperson Niko Felix wrote in a message to The Post. “Participation is voluntary, with no obligation to provide feedback or use the tool.” Another artist in the program, André Allen Anjos, chimed in as well saying that the protest artists' stance didn't reflect the views of most artists in the program. 

Though Sora isn't yet widely available, the tool has been scrutinized over its training materials. In March, OpenAI CTO Mira Murati said she wasn't sure if Sora took training data from YouTube or other video platforms. The next month, YouTube's CEO specifically warned OpenAI that training models on its videos was against its terms of service. 

This article originally appeared on Engadget at https://www.engadget.com/ai/openai-suspends-access-to-sora-video-generation-tool-after-artists-protest-133015289.html?src=rss

©

© CFOTO via Getty Images

Instagram spruced up its DM features on Monday. You can now share your live location with friends, handy for meetups at concerts or other crowded places. The social platform also now lets you add nicknames in one-on-one or group chats, and there are over 300 new stickers to share in DMs.

You can turn on the new location-sharing feature for up to one hour. It only displays to people in your private DMs (one-on-one or group), and locations can’t be forwarded to anyone outside the designated chat. The feature is off by default, and you’ll see a “You are sharing your location” indicator at the top of every message thread where you enable it. You can also stop sharing it manually anytime.

Instagram

Nicknames let you add aliases for yourself or friends inside DMs. “Share an inside joke with a nickname, or simplify lengthy usernames so your friends are easier to recognize,” Instagram’s announcement blog post suggests.

The nicknames only appear in your DMs, not anywhere else on the platform. You can pick or swap out your nickname anytime and choose which friends in a chat have permission to change yours. To create a moniker, tap on the chat name at the top of your conversation, select “Nicknames” and choose the username you want to change.

Finally, Instagram added 17 new sticker packs for DMs, making over 300 new ones available. You can favorite the stickers you like, including those your friends share.

This article originally appeared on Engadget at https://www.engadget.com/social-media/instagram-dms-have-a-new-location-sharing-feature-185959274.html?src=rss

©

© Instagram

Threads is making more changes to address long-running complaints from users. This time, the company is testing out improvements to its search and trending topics feature in updates that Adam Mosseri described as “long-overdue improvements.”

On search, Threads is testing the ability to search for posts within specific date ranges and account-specific searches. The changes are similar to some of X’s advanced search capabilities and could make it easier for users to look for a specific post they want to revisit. The lack of advanced search on Threads has long been frustrating and up to now, the most reliable way to search for a specific Threads post was to use Google.

Threads is also experimenting with AI-powered summaries for its “trending now” topics, which shows US users a handful of popular topics on the platform. The app will also show an “expanded set” of up to 15 trends (currently, Threads only displays five trends at a time). The changes sound a little more like how X handles trends, which are sometimes summarized by its “Grok” AI feature. (Grok has somewhat of a mixed track record when it comes to accuracy, however.) It’s unclear how Meta plans to handle these summaries, though the company already uses Meta AI to summarize Facebook comments and discussions in groups.

The new tests are the latest in a flurry of updates from Threads. In the last week, Meta has also announced a test of custom feeds, which it made official just five days later, and improvements to users’ algorithmic feeds. Those changes also just happen to coincide with a huge surge in growth for Threads competitor Bluesky. The decentralized service has grown to more than 20 million users and has had several days where its growth has rivaled Threads daily sign-up numbers.

This article originally appeared on Engadget at https://www.engadget.com/social-media/threads-is-testing-out-advanced-search-features-and-ai-summaries-for-trending-topics-182250201.html?src=rss

©

© SOPA Images via Getty Images

Australia’s majority party has introduced a bill in Parliament that would ban children under 16 from social media. The legislation, which would put the onus on social platforms rather than children or parents, could fine infringing companies up to AUD$49.5 million ($32.2 million).

The Labor Party’s bill would apply to (among others) Snapchat, TikTok, Instagram and X. It would require platforms to cordon off and destroy any underage user data collected. However, the legislation would include exceptions for health and education services, like Headspace, Google Classroom and YouTube.

“For too many young Australians, social media can be harmful. Almost two-thirds of 14- to 17-year-old Australians have viewed extremely harmful content online, including drug abuse, suicide or self-harm, as well as violent material,” Australia Communications Minister Michelle Rowland told Parliament on Thursday. “A quarter have been exposed to content promoting unsafe eating habits.”

Reuters notes that the law would be one of the most aggressive globally in tackling the problems related to children’s social media use. It wouldn’t include exemptions for parental consent or pre-existing accounts. Essentially, social platforms would have to police their platforms to ensure no child under 16 can use their services.

The bill is supported by the majority (center-left) Labor Party and opposition (right) Liberal Party. “This is a landmark reform,” Australian Prime Minister Anthony Albanese said. “We know some kids will find workarounds, but we’re sending a message to social media companies to clean up their act.”

The (left) Australian Greens have criticized the legislation, saying it ignores expert evidence in “ramming” the law through Parliament without proper scrutiny. “The recent Parliamentary Inquiry into Social Media heard time and time again that an age-ban will not make social media safer for anyone,” Senator Sarah Hanson-Young said in a statement. “[The bill] is complicated to implement and will have unintended consequences for young people.”

Last year, US Surgeon General Vivek Murthy sounded the alarm about the risks of underage social media use. “Children and adolescents who spend more than 3 hours a day on social media face double the risk of mental health problems including experiencing symptoms of depression and anxiety,” the 2023 advisory from the Surgeon General’s office read.

The US requires tech companies to seek parental consent to access the data of children under 13, but it doesn’t have any age restrictions. Reuters notes that France enacted a social media ban for children under 15 last year, but it allows children to still access the services with parental consent.

This article originally appeared on Engadget at https://www.engadget.com/social-media/australia-introduces-a-bill-that-would-ban-children-under-16-from-social-media-174547712.html?src=rss

©

© SOPA Images via Getty Images