Category: Technology

It looks like we’re entering the era of crazy-thin phones. This week, Samsung finally announced the Galaxy S25 Edge, its slimmest smartphone yet. Is there actually a point to it, or is Samsung just trying to beat Apple to its rumored super-thin iPhone? Engadget’s Sam Rutherford joins us to dive into the S25 Edge, as well as some pre-Google I/O news from the Android Show.

Subscribe!

iTunes

Spotify

Pocket Casts

Stitcher

Google Podcasts

Topics

Surface Pro 12-inch review: less weight, less power? – 2:04

ASUS’ Xbox handheld photos leak in FCC filing – 21:39

OpenAI restructures business, announces plan for hostile takeover-proof public benefit corp – 26:14

The EPA announces plans to shut down the Energy Star Program – 31:16

Telemessage, a Signal clone favored by Trump administration officials has been hacked – 34:44

Samsung subsidiary buys Masimo and now it owns all the fancy speakers – 36:35

Half-Life 3 is fully formed and playable?! – 40:59

Around Engadget – 49:53

Pop culture picks – 51:42

Credits 

Hosts: Devindra Hardawar and Cherlynn Low
Guest: Sam Rutherford
Producer: Ben Ellman
Music: Dale North and Terrence O’Brien

This article originally appeared on Engadget at https://www.engadget.com/mobile/smartphones/engadget-podcast-who-needs-samsungs-super-thin-galaxy-s25-edge-113049744.html?src=rss 

TikTok recently began experimenting with an in-app meditation feature that encouraged teens to “wind down” after 10PM. Now, the company is making the feature official for all users and turning it on by default for all teens under the age of 18.

With the change, teens will hit a full-screen “guided meditation exercise” when attempting to scroll after 10PM. The prompt is apparently something you can opt to ignore, but teens who do will encounter a second “harder to dismiss” prompt. TikTok’s adult users will also be able to access the in-app meditations via the app’s screen time controls (the feature will not be on by default for adults).

The company notes that its initial tests of “Sleep Hours” were successful, with 98 percent of teens opting to keep the late-night meditation settings on. Previous attempts by TikTok to limit screen time have a somewhat different track record. Documents that surfaced as part of a lawsuit against the company showed that teens were spending about 107 minutes a day in the app even when screen time was set to a 60-minute limit.

Since then, TikTok has beefed up some of its safety features, including its parental controls, amid increasing scrutiny of the company. TikTok’s fate in the US is still, officially, in limbo as President Donald Trump signed off on another extension of a deadline to ban the app last month. Terms of a final deal that will allow it to remain in the country permanently have yet to be announced, though there are a number of interested buyers.

This article originally appeared on Engadget at https://www.engadget.com/social-media/tiktok-will-try-to-force-teens-to-meditate-after-10pm-231118942.html?src=rss 

Last month, Meta hosted LlamaCon, its first ever generative AI conference. But while the event delivered some notable improvements for developers, it also felt a bit underwhelming considering how important AI is to the company. Now, we know a bit more about why, thanks to a new report in The Wall Street Journal.

According to the report, Meta had originally intended to release its “Behemoth” Llama 4 model at the April developer event, but later delayed its release to June. Now, it’s apparently been pushed back again, potentially until “fall or later.” Meta engineers are reportedly “struggling to significantly improve the capabilities” of the model that Mark Zuckerberg has called “the highest performing base model in the world.”

Meta has already released two smaller Llama 4 models, Scout and Maverick, and has also teased a fourth lightweight model that’s apparently nicknamed “Little Llama.” Meanwhile, the “Behemoth” model will have 288 billion active parameters and “outperforms GPT-4.5, Claude Sonnet 3.7, and Gemini 2.0 Pro on several STEM benchmarks,” the company said last month.

Meta has never given a firm timeline of when to expect the model. The company said last month that it was “still training.” And while Behemoth got a few nods during the LlamaCon keynote, there were no updates on when it might actually be ready. That’s probably because it could still be several months. Inside Meta there are apparently questions “about whether improvements over prior versions are significant enough to justify public release.”

Meta didn’t immediately respond to a request for comment. As the report notes, it wouldn’t be the first company to run into snags as it races to release new models and outpace competitors. But the delay is still notable given the Meta’s lofty ambitions when it comes to AI. Zuckerberg has made AI a top priority with Meta planning to spend as much as $72 billion on its AI infrastructure this year.

This article originally appeared on Engadget at https://www.engadget.com/ai/metas-behemoth-llama-4-model-might-still-be-months-away-221240585.html?src=rss 

X has once again been accepting payments from people associated with terrorist groups and other entities subject to US sanctions, according to a new report from the Tech Transparency Project (TTP). According to the report, X has not only accepted payments in exchange for its premium service, but in some cases has provided an “ID verified” badge.

The report once again questions whether X is complying with US sanctions that restrict companies’ ability to do business with individuals and entities that have been deemed a security threat. Last year, the TTP published a similar report that identified more than two dozen verified accounts that were affiliated with sanctioned groups, including leaders of Hezbollah and accounts associated with Houthis in Yemen. Many of those checkmarks were subsequently revoked, with X promising to “maintain a safe, secure and compliant platform.”

But some of those accounts simply “resubscribed” to X’s premium service or created fresh accounts, according to the report, which is based on research between November 2024 and April 2025. “TTP’s new investigation found an array of blue checkmark accounts for U.S.-sanctioned individuals and organizations, including several that appeared to simply re-subscribe to premium service or create new accounts after their old ones were restricted or removed by X,” the report says. “Moreover, some of the accounts were ‘ID verified,’ meaning X conducted an additional review to confirm their identity.”

The report once again highlights verified accounts associated with members of Hezbollah, including one of its founders, as well as Houthi officials who “are making heavy use of X for messaging and propaganda.” The son of Libyan dictator Muammar Gadhafi, whose account was previously suspended, also currently has a blue check, as does Raghad Saddam Hussein al-Tikriti, one of Saddam Hussein’s daughters. Both have been under sanctions for more than a decade.

X didn’t respond to a request for comment on the report. In response to last year’s report, the company said it would “take action if necessary.” However, it’s unclear if the company changed any of its practices regarding who can pay for premium subscriptions.

“If a small team can use X’s public facing search tools to identify these accounts, it’s unclear why a multi-billion-dollar company cannot do the same,” Michelle Kuppersmith, the executive director for Campaign for Accountability, the watchdog group that runs TTP said in a statement. “It’s one thing to allow terrorists to have a voice on the platform; it’s another entirely to allow them to pay for a more effective megaphone.”

This article originally appeared on Engadget at https://www.engadget.com/social-media/x-is-once-again-selling-checkmarks-to-us-sanctioned-groups-report-says-194352896.html?src=rss 

A team of doctors and scientists have successfully treated a rare genetic condition with the first-ever personalized gene-editing therapy. Results of the groundbreaking treatment have been published in The New England Journal of Medicine, with an accompanying editorial by a doctor who had previously overseen the FDA’s gene-therapy regulation efforts.

The patient in this historic case was KJ, an infant born with CPS1 deficiency, which has about a 50 percent mortality rate within the first week. Patients that do survive can experience severe brain disease, mental and developmental delays, and potential liver transplants. His care team developed a personalized gene-editing treatment based on CRISPR, a technology for modifying human DNA.

The successful gene repair for KJ combined years’ worth of previous federally-funded medical research, including the discovery of CRISPR and human genome sequencing that allowed the mutation to be identified.

This approach to gene editing could potentially be used in the future to aid patients with other genetic disorders, such as sickle cell disease, cystic fibrosis, Huntington’s disease and muscular dystrophy. A pair of CRISPR-based drugs have already received FDA approval for sickle cell disease treatments, but there is still a lot to potentially be explored in this field.

This article originally appeared on Engadget at https://www.engadget.com/science/doctors-successfully-treated-a-baby-with-the-first-ever-personalized-gene-editing-therapy-202307902.html?src=rss 

Coinbase has been betrayed from within. The cryptocurrency exchange said that cyber criminals bribed some of its support agents to share personal information about Coinbase customers. Attackers acquired data such as names, addresses, emails, phone numbers, images of government IDs, masked bank account numbers and masked sections of social security numbers. The perpetrators tricked some Coinbase users into sending them money and also demanded $20 million from the company to not publicly disclose the ill-gotten information.

Coinbase has not paid the ransom and is cooperating with law enforcement to press charges. In the blog post, the company said it would offer a $20 million reward for information that could lead to arresting and convicting the remaining attackers. It also said that users’ login credentials, two-factor authentication codes and private keys are still secure. It will reimburse customers who sent funds to the extortionists and will place additional safeguards on vulnerable accounts. According to an SEC filing, the incident is projected to cost Coinbase $180 million to $400 million.

This article originally appeared on Engadget at https://www.engadget.com/cybersecurity/extortionists-bribed-coinbase-employees-to-give-them-customer-data-174713732.html?src=rss 

Ministers in the UK House of Commons have blocked an amendment to a data bill that would require AI companies to disclose their use of copyrighted materials, according to The Guardian. This transparency amendment was stripped out of the text by invoking something called financial privilege, an arcane parliamentary procedure that suggests that any new regulations would require a new budget.

The official site of the UK parliament says this procedure “may be used by the Commons as grounds for overruling any House of Lords proposal that has cost implications.” It looks like that’s exactly what happened here, with those in favor of removing the amendment bringing up the cost of a potential regulatory body. There were 297 MPs who voted in favor of removing the amendment, with 168 opposing.

Chris Bryant, data protection minister, said that he recognized that this could feel like an “apocalyptic moment” for the creative industries, but that he thinks the transparency amendment requires changes “in the round and not just piecemeal.”

The amendment was passed in the House of Lords earlier this week. Baroness Beeban Kidron of the Lords responded to today’s move by saying that “the government failed to answer its own backbenchers who repeatedly asked ‘if not now then when?’” She also said it was “astonishing that a Labour government would abandon the labor force of an entire section,” referring to the plight of creative workers whose jobs have been or at risk of being replaced by AI. Lady Kidron went on to accuse the government of allowing “theft at scale” and cozying up “to those who are thieving.”

“Across the creative and business community, across parliament, people are gobsmacked that the government is playing parliamentary chess with their livelihoods,” she concluded.

As expected, Kidron will introduce a rephrased amendment before the bill’s return to the Lords next week. This sets up yet another showdown when the bill returns to the Commons for another pass.

Owen Meredith, the chief executive of the News Media Association, told The Guardian that it’s “extremely disappointing that the government has failed to listen to the deep concerns of the creative industries, including news publishers who are so fundamental to uploading our democratic values.” He accused the government of using parliamentary procedure to “dismiss industry concerns, rather than taking this timely opportunity to introduce the transparency that will drive a dynamic licensing market for the UK’s immensely valuable creative content.”

The government’s preferred plan includes the reliance on an opt-out clause. This would give AI companies free rein over any and all content, except in the cases when a creator has explicitly opted out.

Last week, hundreds of artists and organizations banded together to urge the government not to “give our work away at the behest of a handful of powerful overseas tech companies.” The artists involved in this campaign included Paul McCartney, Elton John and Dua Lipa, among others.

America is set to host its own version of the “give everything to AI companies” game show. Republicans have snuck in a provision to the budget bill that would ban regulation on the AI industry for ten years. That’ll end well.

This article originally appeared on Engadget at https://www.engadget.com/big-tech/uk-parliament-opts-not-to-hold-ai-companies-accountable-over-copyright-material-180234550.html?src=rss 

TikTok may have run afoul of European regulators over advertising transparency, in the latest bout between EU regulators and big tech companies. The European Commission has formally warned TikTok that its advertising transparency tools don’t go far enough, falling short of EU rules requiring online platforms to maintain a publicly available repository of all paid ads being shown to users.

Regulators are taking issue with the data, or lack thereof, being provided in the repository that TikTok has curated. They claim that TikTok has failed to provide information on who paid for ads, what audiences they are targeting and the precise product or service those ads are promoting.

This information is all used by regulators and researchers to help weed out harmful content like scams, disinformation or coordinated influence campaigns. These rules fall under the Digital Services Act (DSA), an EU regulation adopted in 2022 that governs illegal content, transparent advertising and disinformation. Enforcement for Very Large Online Platforms (VLOPs) like TikTok began in the summer of 2023, and the Commission opened formal proceedings to assess whether TikTok had violated the DSA in February of 2024.

In a statement, European Commission Tech VP Henna Virkkunen said, “Transparency in online advertising — who pays and how audiences are targeted — is essential to safeguarding the public interest.Whether we are defending the integrity of our democratic elections, protecting public health or protecting consumers from scam ads, citizens have a right to know who is behind the messages they see.”

In a statement to the press, a TikTok spokesperson disagreed with the findings, saying “While we support the goals of the regulation and continue to improve our ad transparency tools, we disagree with some of the Commission’s interpretations and note that guidance is being delivered via preliminary findings rather than clear, public guidelines.” This isn’t the first time that TikTok has found itself in hot water with internet safety regulators in Europe.

Under the rules of the DSA, TikTok parent company ByteDance could be fined as much as 6 percent of its total worldwide revenue and be subjected to an enhanced supervision period where regulators ensure that any required changes are being made.

This article originally appeared on Engadget at https://www.engadget.com/big-tech/eu-tech-chiefs-believe-tiktok-is-breaking-ad-transparency-rules-181706441.html?src=rss 

The well-reviewed Stellar Blade will be available for PC players on June 11, after launching for PS5 in April of last year. This release coincides with a system update that brings new outfits and “exciting additional content.” The update will be available for both PC and PS5.

The PC version of the game includes some graphical upgrades. It supports NVIDIA DLSS 4 resolution scaling and frame generation, NVIDIA DLAA image enhancement and NVIDIA Reflex latency reduction. It also supports AMD FSR 3 and offers the ability to unlock the frame rate to better suit the needs of a particular PC. It can handle both 21:9 and 32:9 ultrawide resolutions.

Stellar Blade was formerly a PS5 exclusive, so the PC version works with the iconic the DualSense controller. This allows for haptic feedback and adaptive trigger features. For everyone else, the game offers customization options for gamepads, keyboards and mice.

As for specs, Sony recommends at least 16GB of RAM, an Intel Core i5 processor and a GeForce GTX 1060. However, that’s for playing on the lowest setting.

We now live in a post-exclusive world. Sony has been releasing tons of its first-party games for PC. These include Marvel’s Spider-Man and the sequel, God of War Ragnarök and Horizon Zero Dawn, to name a few.

This article originally appeared on Engadget at https://www.engadget.com/gaming/pc/stellar-blade-heads-to-pc-on-june-11-185536164.html?src=rss 

It’s been over two years since Sony had a new pair of headphones in its 1000X lineup, but the newly announced WH-1000XM6 seem worth the wait. From the outside, the new ANC headphones look similar to the streamlined WH-1000XM5 from 2022, with the only major change being the return of ear cups that rotate and fold flat. Inside, though, Sony’s packed in improvements that put the WH-1000XM6 ahead of the competition in terms of both sound quality and active noise cancellation.

That’s mostly thanks to the new QN3 processor on the inside, the addition of four new built-in microphones (from eight on the 1000XM5 to 12 on the 1000XM6) and an updated system for adapting ANC performance around fit and pressure. All three combine for noticeably better ANC. Those changes haven’t led to shorter battery life either — you’ll get the same 30 hours with ANC activated on the WH-1000XM6 that you were able to get on the WH-1000XM5.

The WH-1000XM6 is not only able to play spatial audio tracks like the WH-100XM5 and AirPods Max can, the headphones can also convert any stereo track you throw at it into 360-degree audio. For most people, though, the more immediately useful new feature will be the ability to keep using the headphones while you charge them, something oddly missing from previous models.

At $450, the WH-100XM6 are a good $50 more expensive than WH-1000XM5 and tiptoeing closer to the exorbitantly priced AirPods Max. The best way to know if Sony’s new headphones are right for you is to read Engadget’s review and ideally, try them for yourself. Until then, we’ve put together an overview of how the WH-1000XM6 compare to the WH-1000XM5 and the AirPods Max.

WH-1000MX6

WH-1000MX5

AirPods Max

Pricing (MSRP)

$450

$400

$549

Active noise cancellation

Yes

Yes

Yes

Transparency mode

Yes

Yes

Yes

Adaptive audio

Yes

Yes

Yes

Multipoint connectivity

Yes

Can switch between 2 Bluetooth connections

Fast Pair with Android devices 

Swift Pair with Windows 11 and Windows 10 devices

Yes

Can switch between 2 Bluetooth connections

Fast Pair with Android devices

Swift Pair with Windows 11 and Windows 10 devices

No

Automatic device switching between Apple devices only

Battery life (rated)

Up to 30 hours with ANC on

Up to 40 hours with ANC off

Up to 30 hours with ANC on

Up to 40 hours with ANC off

Up to 20 hours with ANC on

Up to 20 hours with ANC and Spatial Audio on

Up to 20 hours of talk time

Wired charging

USB-C

USB-C

USB-C

Water & dust resistance

No

No

No

Size & weight

8.96 oz

8.82 oz

13.6 oz

Chip(s)

Sony QN3

Sony QN1 + V1

Apple H1

Bluetooth

Bluetooth 5.3

Bluetooth 5.2

Bluetooth 5.0

Bluetooth codecs

SBC, AAC, LDAC, LC3

SBC, AAC, LDAC

AAC, SBC

Spatial audio

Yes

360 Audio

360 Spatial Sound Upmix for converting stereo audio

Yes

360 Audio

Yes

Control type

Touch controls for playback and volume

NC/AMB button for ANC

Touch controls for playback and volume

NC/AMB button

Digital Crown dial for playback and volume

Noise control button for ANC

Mics & sensors

12 microphones total: six microphones for calls

Touch sensors

Accelerometer

Gyroscope

Eight microphones total: four microphones for calls

Touch sensors

Accelerometer

Gyroscope

Nine microphones total: eight ANC mics, three mics for voice pickup (two shared for ANC)

Optical sensor

Position sensor

Case-detect sensor

Accelerometer

Gyroscope

This article originally appeared on Engadget at https://www.engadget.com/audio/headphones/heres-how-the-sony-wh-1000xm6-compare-to-the-wh-1000xm5-and-airpods-max-193706730.html?src=rss