verge-rss

The Verge

Microsoft is preparing to enable Copilot in OneNote to read and analyze handwritten notes. The feature entered beta testing late last month, and will allow OneNote users to write handwritten notes using a stylus and then summarize them, ask questions, or even generate to-do lists based on the notes.
The AI-powered Copilot feature in OneNote will even be able to convert your handwritten notes into text for easy editing and sharing. Microsoft first launched Copilot inside OneNote in November, and this upgrade will be available to existing Copilot for Microsoft 365 subscribers and Copilot Pro users once it rolls out more broadly.

Image: The Verge
Copilot can quickly convert your handwritten notes into easier to read text.

I’ve been briefly testing Copilot’s ability to read my handwriting, and I’m impressed that it’s even able to decipher it. Summarization works well across small handwritten notes and even larger ones. I asked Copilot to rewrite an entire paragraph of handwritten notes which it managed to format into easy to read text that was true to the original but also a bit more jovial. That’s impressive given generative AI models have a tendency to just make stuff up half the time.
If you use OneNote for handwritten to-do lists, then this feature definitely makes it easier to convert those into text at a later stage. I created a handwritten list and Copilot was able to accurately convert this into a text list within seconds.

Image: The Verge
A to-do list that’s been converted from handwritten notes.

I’m not sure how Copilot will handle even worse handwriting, though. Samsung’s Galaxy AI has a similar feature that can automatically format handwritten lists, but it struggles with some of the worst handwriting I’ve ever seen (sorry, Allison!).
If you want to test Copilot’s handwritten note recognition you’ll need to be a Microsoft 365 Insider running the latest OneNote on Windows build (17628.20006 or later) and have a Copilot Pro or Copilot for Microsoft 365 subscription.

The Verge

Microsoft is preparing to enable Copilot in OneNote to read and analyze handwritten notes. The feature entered beta testing late last month, and will allow OneNote users to write handwritten notes using a stylus and then summarize them, ask questions, or even generate to-do lists based on the notes.

The AI-powered Copilot feature in OneNote will even be able to convert your handwritten notes into text for easy editing and sharing. Microsoft first launched Copilot inside OneNote in November, and this upgrade will be available to existing Copilot for Microsoft 365 subscribers and Copilot Pro users once it rolls out more broadly.

Image: The Verge
Copilot can quickly convert your handwritten notes into easier to read text.

I’ve been briefly testing Copilot’s ability to read my handwriting, and I’m impressed that it’s even able to decipher it. Summarization works well across small handwritten notes and even larger ones. I asked Copilot to rewrite an entire paragraph of handwritten notes which it managed to format into easy to read text that was true to the original but also a bit more jovial. That’s impressive given generative AI models have a tendency to just make stuff up half the time.

If you use OneNote for handwritten to-do lists, then this feature definitely makes it easier to convert those into text at a later stage. I created a handwritten list and Copilot was able to accurately convert this into a text list within seconds.

Image: The Verge
A to-do list that’s been converted from handwritten notes.

I’m not sure how Copilot will handle even worse handwriting, though. Samsung’s Galaxy AI has a similar feature that can automatically format handwritten lists, but it struggles with some of the worst handwriting I’ve ever seen (sorry, Allison!).

If you want to test Copilot’s handwritten note recognition you’ll need to be a Microsoft 365 Insider running the latest OneNote on Windows build (17628.20006 or later) and have a Copilot Pro or Copilot for Microsoft 365 subscription.

Read More 

Would you confront Ganondorf ten times over to acquire such a relic? | Image: Nintendo

Preorders have just opened for the ‘Hyrule Edition’ Nintendo Switch Lite, which commemorates the forthcoming launch of The Legend of Zelda: Echoes of Wisdom. Best Buy is currently accepting orders for the limited edition console for $209.99 ahead of its September 26th release date. You’ll still have to purchase Echoes of Wisdom separately, but the special edition console does include a 12-month individual subscription to Nintendo Switch Online + Expansion Pack (a $49.99 value).

The console comes dipped in a golden hue with black trim — an iconic combo befitting Hylian royalty. The noble house’s signature crest graces the rear, and you’ll find a smaller Triforce symbol stamped on the front in the lower-right corner. There’s not much more setting it apart from any other Switch Lite, but considering Echoes of Wisdom is the first Zelda title to actually star Princess Zelda as the primary character, the occasion may move you to pick it up even if you already own Nintendo’s dedicated handheld.
Nintendo’s premium subscription tier, meanwhile, provides access to online play, cloud-based saves, and the ability to play older classics from the Game Boy, Game Boy Advance, NES, SNES, N64, and even the Sega Genesis eras. This includes heralded Zelda titles like The Legend of Zelda: Ocarina of Time, The Legend of Zelda: Majora’s Mask, and The Legend of Zelda: A Link to the Past. Needless to say, it’s a great way to revisit the series’ roots whenever you’re feeling nostalgic.

Would you confront Ganondorf ten times over to acquire such a relic? | Image: Nintendo

Preorders have just opened for the ‘Hyrule Edition’ Nintendo Switch Lite, which commemorates the forthcoming launch of The Legend of Zelda: Echoes of Wisdom. Best Buy is currently accepting orders for the limited edition console for $209.99 ahead of its September 26th release date. You’ll still have to purchase Echoes of Wisdom separately, but the special edition console does include a 12-month individual subscription to Nintendo Switch Online + Expansion Pack (a $49.99 value).

The console comes dipped in a golden hue with black trim — an iconic combo befitting Hylian royalty. The noble house’s signature crest graces the rear, and you’ll find a smaller Triforce symbol stamped on the front in the lower-right corner. There’s not much more setting it apart from any other Switch Lite, but considering Echoes of Wisdom is the first Zelda title to actually star Princess Zelda as the primary character, the occasion may move you to pick it up even if you already own Nintendo’s dedicated handheld.

Nintendo’s premium subscription tier, meanwhile, provides access to online play, cloud-based saves, and the ability to play older classics from the Game Boy, Game Boy Advance, NES, SNES, N64, and even the Sega Genesis eras. This includes heralded Zelda titles like The Legend of Zelda: Ocarina of Time, The Legend of Zelda: Majora’s Mask, and The Legend of Zelda: A Link to the Past. Needless to say, it’s a great way to revisit the series’ roots whenever you’re feeling nostalgic.

Read More 

Illustration: The Verge

OpenAI has created an internal scale to track the progress its large language models are making toward artificial general intelligence, or AI with human-like intelligence, a spokesperson told Bloomberg.
Today’s chatbots, like ChatGPT, are at Level 1. OpenAI claims it is nearing Level 2, defined as a system that can solve basic problems at the level of a person with a PhD. Level 3 refers to AI agents capable of taking actions on a user’s behalf. Level 4 involves AI that can create new innovations. Level 5, the final step to achieving AGI, is AI that can perform the work of entire organizations of people. OpenAI has previously defined AGI as “a highly autonomous system surpassing humans in most economically valuable tasks.”
OpenAI’s unique structure is centered around its mission of achieving AGI, and how OpenAI defines AGI is important. The company has said that “if a value-aligned, safety-conscious project comes close to building AGI” before OpenAI does, it commits to not competing with the project and dropping everything to assist. The phrasing of this in OpenAI’s charter is vague, leaving room for the judgment of the for-profit entity (governed by the nonprofit), but a scale that OpenAI can test itself and competitors on could help dictate when AGI is reached in clearer terms.
Still, AGI is still quite a ways away: it will take billions upon billions of dollars worth of computing power to reach AGI, if at all. Timelines from experts, and even at OpenAI, vary wildly. In October 2023, OpenAI CEO Sam Altman said we are “five years, give or take,” before reaching AGI.
This new grading scale, though still under development, was introduced a day after OpenAI announced its collaboration with Los Alamos National Laboratory, which aims to explore how advanced AI models like GPT-4o can safely assist in bioscientific research. A program manager at Los Alamos, responsible for the national security biology portfolio and instrumental in securing the OpenAI partnership, told The Verge that the goal is to test GPT-4o’s capabilities and establish a set of safety and other factors for the US government. Eventually, public or private models can be tested against these factors to evaluate their own models.
In May, OpenAI dissolved its safety team after the group’s leader, OpenAI cofounder Ilya Sutskever, left the company. Jan Leike, a key OpenAI researcher, resigned shortly after claiming in a post that “safety culture and processes have taken a backseat to shiny products” at the company. While OpenAI denied that was the case, some are concerned about what this means if the company does in fact reach AGI.
OpenAI hasn’t provided details on how it assigns models to these internal levels (and declined The Verge’s request for comment). However, company leaders demonstrated a research project using the GPT-4 AI model during an all-hands meeting on Thursday and believe this project showcases some new skills that exhibit human-like reasoning, according to Bloomberg.
This scale could help provide a strict definition of progress, rather than leaving it up for interpretation. For instance, OpenAI CTO Mira Murati said in an interview in June that the models in its labs are not much better than what the public has already. Meanwhile, CEO Sam Altman said late last year that the company recently “pushed the veil of ignorance back,” meaning the models are remarkably more intelligent.

Illustration: The Verge

OpenAI has created an internal scale to track the progress its large language models are making toward artificial general intelligence, or AI with human-like intelligence, a spokesperson told Bloomberg.

Today’s chatbots, like ChatGPT, are at Level 1. OpenAI claims it is nearing Level 2, defined as a system that can solve basic problems at the level of a person with a PhD. Level 3 refers to AI agents capable of taking actions on a user’s behalf. Level 4 involves AI that can create new innovations. Level 5, the final step to achieving AGI, is AI that can perform the work of entire organizations of people. OpenAI has previously defined AGI as “a highly autonomous system surpassing humans in most economically valuable tasks.”

OpenAI’s unique structure is centered around its mission of achieving AGI, and how OpenAI defines AGI is important. The company has said that “if a value-aligned, safety-conscious project comes close to building AGI” before OpenAI does, it commits to not competing with the project and dropping everything to assist. The phrasing of this in OpenAI’s charter is vague, leaving room for the judgment of the for-profit entity (governed by the nonprofit), but a scale that OpenAI can test itself and competitors on could help dictate when AGI is reached in clearer terms.

Still, AGI is still quite a ways away: it will take billions upon billions of dollars worth of computing power to reach AGI, if at all. Timelines from experts, and even at OpenAI, vary wildly. In October 2023, OpenAI CEO Sam Altman said we are “five years, give or take,” before reaching AGI.

This new grading scale, though still under development, was introduced a day after OpenAI announced its collaboration with Los Alamos National Laboratory, which aims to explore how advanced AI models like GPT-4o can safely assist in bioscientific research. A program manager at Los Alamos, responsible for the national security biology portfolio and instrumental in securing the OpenAI partnership, told The Verge that the goal is to test GPT-4o’s capabilities and establish a set of safety and other factors for the US government. Eventually, public or private models can be tested against these factors to evaluate their own models.

In May, OpenAI dissolved its safety team after the group’s leader, OpenAI cofounder Ilya Sutskever, left the company. Jan Leike, a key OpenAI researcher, resigned shortly after claiming in a post that “safety culture and processes have taken a backseat to shiny products” at the company. While OpenAI denied that was the case, some are concerned about what this means if the company does in fact reach AGI.

OpenAI hasn’t provided details on how it assigns models to these internal levels (and declined The Verge’s request for comment). However, company leaders demonstrated a research project using the GPT-4 AI model during an all-hands meeting on Thursday and believe this project showcases some new skills that exhibit human-like reasoning, according to Bloomberg.

This scale could help provide a strict definition of progress, rather than leaving it up for interpretation. For instance, OpenAI CTO Mira Murati said in an interview in June that the models in its labs are not much better than what the public has already. Meanwhile, CEO Sam Altman said late last year that the company recently “pushed the veil of ignorance back,” meaning the models are remarkably more intelligent.

Read More 

Image: Frontier Developments

Planet Coaster’s sequel isn’t just about building the best theme park anymore: it’s about water parks, too. The sequel, coming this fall, will finally let you construct everything from wave pools and lazy rivers to custom-built waterslides.
You’ll also get to outfit your aquatic experience with changing rooms, sunbeds, lifeguards, and more. Planet Coaster 2 builds upon its predecessor with the addition of new building and pathing tools, as well as an advanced event sequencer system that can give your guests new experiences. There are also new rides, scenery pieces, and themes.
Another neat feature coming to Planet Coaster 2 is the ability to create shared theme parks with friends — but you can’t all build in the park at the same time. The developer says you can “jump one-at-a-time” into a shared park and then “save your progress to allow your fellow thrill-seekers to build upon your changes,” which is a bit of a bummer.

Image: Frontier Developments

Just like its predecessor, Planet Coaster 2 will come with three game modes: career, sandbox, and challenge mode. It will also be available on PC, Xbox Series X | S, and PlayStation 5 when it launches later this year.
If you haven’t tried Planet Coaster, Planet Zoo, or even Jurassic World Evolution 2 from Frontier Developments yet, I’d highly recommend them. You may not be able to get away with some of the things you could do in Roller Coaster (or Zoo) Tycoon, but they’re just as easy to get lost in.

Image: Frontier Developments

Planet Coaster’s sequel isn’t just about building the best theme park anymore: it’s about water parks, too. The sequel, coming this fall, will finally let you construct everything from wave pools and lazy rivers to custom-built waterslides.

You’ll also get to outfit your aquatic experience with changing rooms, sunbeds, lifeguards, and more. Planet Coaster 2 builds upon its predecessor with the addition of new building and pathing tools, as well as an advanced event sequencer system that can give your guests new experiences. There are also new rides, scenery pieces, and themes.

Another neat feature coming to Planet Coaster 2 is the ability to create shared theme parks with friends — but you can’t all build in the park at the same time. The developer says you can “jump one-at-a-time” into a shared park and then “save your progress to allow your fellow thrill-seekers to build upon your changes,” which is a bit of a bummer.

Image: Frontier Developments

Just like its predecessor, Planet Coaster 2 will come with three game modes: career, sandbox, and challenge mode. It will also be available on PC, Xbox Series X | S, and PlayStation 5 when it launches later this year.

If you haven’t tried Planet Coaster, Planet Zoo, or even Jurassic World Evolution 2 from Frontier Developments yet, I’d highly recommend them. You may not be able to get away with some of the things you could do in Roller Coaster (or Zoo) Tycoon, but they’re just as easy to get lost in.

Read More 

The leaked Pixel 9, 9 Pro, and 9 Pro XL. | Image: Rozetked

Pixels are leaking again, and this time, it looks like a larger Pixel 9 Pro ‘XL’ is coming and coming fast in August. We may never know why Google can’t keep its Pixel phones under wraps until launch, but here we are again in 2024 with a new set of leaks. They started back in January with the new Pixel 9, which apparently will come in three sizes. Then, in February, we got the Pixel Fold 2, which might be called the Pixel 9 Pro Fold.
The official word is that Google’s holding a “Made by Google” Pixel event on August 13th. The timing indicates Google might launch the new Pixel 9 and 9 Pro series well ahead of the usual October release, which could give it a buzzy lead ahead of the iPhone 16 expected in September. However, it would mean the current flagship Pixel 8 lineup’s time in the spotlight would get cut short — it feels like it was just reviewed!
Google seems to be bringing back the “XL” moniker for a larger Pixel 9 Pro, something it had dropped with the 2020 launch of the Pixel 5. From leaked images, it looks like the Pixel 9 Pro and 9 Pro XL will have the same camera systems, so unlike previous years, those who prefer an easier-to-hold device don’t have to opt for the non-Pro Pixel with a lesser camera.
The prices for all the new hardware might get a slight bump, too, according to reports from Dealabs and Android Authority.
Scroll down to see all the rumors, including a future tablet, the inevitable Pixel 9A, and watches.

The leaked Pixel 9, 9 Pro, and 9 Pro XL. | Image: Rozetked

Pixels are leaking again, and this time, it looks like a larger Pixel 9 Pro ‘XL’ is coming and coming fast in August.

We may never know why Google can’t keep its Pixel phones under wraps until launch, but here we are again in 2024 with a new set of leaks. They started back in January with the new Pixel 9, which apparently will come in three sizes. Then, in February, we got the Pixel Fold 2, which might be called the Pixel 9 Pro Fold.

The official word is that Google’s holding a “Made by Google” Pixel event on August 13th. The timing indicates Google might launch the new Pixel 9 and 9 Pro series well ahead of the usual October release, which could give it a buzzy lead ahead of the iPhone 16 expected in September. However, it would mean the current flagship Pixel 8 lineup’s time in the spotlight would get cut short — it feels like it was just reviewed!

Google seems to be bringing back the “XL” moniker for a larger Pixel 9 Pro, something it had dropped with the 2020 launch of the Pixel 5. From leaked images, it looks like the Pixel 9 Pro and 9 Pro XL will have the same camera systems, so unlike previous years, those who prefer an easier-to-hold device don’t have to opt for the non-Pro Pixel with a lesser camera.

The prices for all the new hardware might get a slight bump, too, according to reports from Dealabs and Android Authority.

Scroll down to see all the rumors, including a future tablet, the inevitable Pixel 9A, and watches.

Read More 

It’s a big price for big sound, but the AirPods Max are a little more reasonable at $398. | Photo by Chris Welch / The Verge

The high-end pricing on Apple’s first-gen products never ceases to amaze me — just look at the $3,500 Apple Vision Pro or its $5,000 display. The AirPods Max were also probably a bit more expensive than they should have been at their $549 launch price, but that’s relatively tame compared to the competition Apple is targeting with its luxe noise-canceling headphones. Thankfully, they’re now down to $398 ($151 off) in select colors at Amazon and Walmart, which is a record low.

Some may feel the AirPods Max are getting a bit long in the tooth since we first reviewed them in 2020, but Apple’s wireless headphones are still our top recommendation for Apple users and some of the best headphones overall. They sound nearly as good as the Sony WH-1000XM5 and Bose QuietComfort Ultra, perhaps only yielding to more niche audiophile options that can cost three times as much.
In terms of features, the Max are built with Apple users in mind. They support a well-balanced soundstage and Apple’s immersive spatial audio feature, not to mention great noise cancellation that rivals premium headphones from Sony and Bose. Their transparency mode is also second to none given how natural it sounds. Apple users even get built-in Find My tracking, one-touch pairing, and fast switching between Apple devices — features you won’t find on many pairs of headphones, aside from perhaps the Beats Studio Pro.

The luxurious metal build further justifies the premium price tag, whereas similar options from Bose and Sony can feel a bit plasticky (only the new Sonos Ace really compare). They also sport large, comfortable ear cups that magnetically detach and a handy digital crown, the latter of which lets you manage volume, calls, playback, and Siri easier than with gesture or swipe-based controls. That said, you might find the Max bulkier than the competition, which can make them annoying to travel with.
Their biggest downside, other than the fact that they lack a USB-C port, is that there’s no 3.5mm audio jack. That means you’ll need to pick up the optional $35 Lightning to 3.5mm cable to use them on a plane or take advantage of hi-res audio codecs in supported apps.

Read our AirPods Max review.

It’s a big price for big sound, but the AirPods Max are a little more reasonable at $398. | Photo by Chris Welch / The Verge

The high-end pricing on Apple’s first-gen products never ceases to amaze me — just look at the $3,500 Apple Vision Pro or its $5,000 display. The AirPods Max were also probably a bit more expensive than they should have been at their $549 launch price, but that’s relatively tame compared to the competition Apple is targeting with its luxe noise-canceling headphones. Thankfully, they’re now down to $398 ($151 off) in select colors at Amazon and Walmart, which is a record low.

Some may feel the AirPods Max are getting a bit long in the tooth since we first reviewed them in 2020, but Apple’s wireless headphones are still our top recommendation for Apple users and some of the best headphones overall. They sound nearly as good as the Sony WH-1000XM5 and Bose QuietComfort Ultra, perhaps only yielding to more niche audiophile options that can cost three times as much.

In terms of features, the Max are built with Apple users in mind. They support a well-balanced soundstage and Apple’s immersive spatial audio feature, not to mention great noise cancellation that rivals premium headphones from Sony and Bose. Their transparency mode is also second to none given how natural it sounds. Apple users even get built-in Find My tracking, one-touch pairing, and fast switching between Apple devices — features you won’t find on many pairs of headphones, aside from perhaps the Beats Studio Pro.

The luxurious metal build further justifies the premium price tag, whereas similar options from Bose and Sony can feel a bit plasticky (only the new Sonos Ace really compare). They also sport large, comfortable ear cups that magnetically detach and a handy digital crown, the latter of which lets you manage volume, calls, playback, and Siri easier than with gesture or swipe-based controls. That said, you might find the Max bulkier than the competition, which can make them annoying to travel with.

Their biggest downside, other than the fact that they lack a USB-C port, is that there’s no 3.5mm audio jack. That means you’ll need to pick up the optional $35 Lightning to 3.5mm cable to use them on a plane or take advantage of hi-res audio codecs in supported apps.

Read our AirPods Max review.

Read More 

Photo by Amelia Holowaty Krales / The Verge

A group of self-proclaimed “gay furry hackers” says it breached the Heritage Foundation earlier this month, releasing two gigabytes of the right-wing think tank’s internal data on Tuesday. On its Telegram channel, the hacktivist collective SiegedSec — which has previously claimed responsibility for hacking NATO’s computer systems — said the Heritage hack was part of its #OpTransRights campaign, which also targeted the far-right media outlet Real America’s Voice and the Hillsong megachurch. The group also cited their objections to Project 2025, the Heritage Foundation’s policy proposal for a second term for former President Donald Trump, as a motivating factor.
In an email to The Verge, Heritage Foundation spokesperson Noah Weinrich denied that Heritage had been hacked, calling it a “false narrative and an exaggeration by a group of criminal trolls trying to get attention.”
“An organized group stumbled upon a two-year-old archive of The Daily Signal website that was available on a public-facing website owned by a contractor,” Weinrich said in an emailed statement. “The information obtained was limited to usernames, names, email addresses, and incomplete password information of both Heritage and non-Heritage content contributors, as well as article comments and the IP address of the commenter. No Heritage systems were breached at any time, and all Heritage databases and websites remain secure, including Project 2025. The data at issue has been taken down, and additional security steps have since been taken as a precaution.”
The Heritage Foundation claims the alleged hack was an exaggeration for attention, but in the meantime, Mike Howell, the executive director of the foundation’s Oversight Project, has triumphantly taken credit for disbanding “the Gay Furry Hackers.”

COMPLETE AND TOTAL VICTORY I have forced the Gay Furry Hackers to DISBAND https://t.co/TSi2qVCi7j— Mike Howell (@MHowellTweets) July 10, 2024

SiegedSec has released chat logs of a conversation one of its members claimed to have with Howell on Signal. The chat logs show a person who claims to be Howell asking a SiegedSec member why the group hacked the Heritage Foundation and threatening to expose the hackers.

We are in the process of identifying and outting members of your group
Reputations and lives will be destroyed
Closeted Furries will be presented to the world for the degenerate perverts they are
You cannot hide Your means are miniscule compared to mine. You now can either turn yourself in or you can cooperate

Howell confirmed the legitimacy of the messages in an X exchange with a Daily Dot reporter.
Weinrich did not comment on the alleged messages between Howell and SiegedSec.
The messages also show Howell claiming to be “tied up with the fbi issuing a 2702 order” on SiegedSec’s social media. SiegedSec has indeed disbanded, a decision its members attribute to their “own mental health, the stress of mass publicity, and to avoid the eye of the FBI.”
“Mike’s threats and insults showed anger that confirmed what Heritage denied.”
A SiegedSec representative who goes by vio told The Verge they “completely expected” Heritage to deny that it had been hacked. “Many companies try denial to save face,” vio said. “The server we hacked was linked to The Daily Signal, and the server was named ‘first-heritage-foundation’. Clearly, Heritage was genuinely hacked.”
“Mike’s threats and insults showed anger that confirmed what Heritage denied,” vio said.
In a statement on Telegram, SiegedSec said the goal of the hack was to draw attention to — and combat — the Heritage Foundation’s anti-LGBT and anti-abortion policy proposals.
“The Heritage Foundation is a conservative think tank in America, among the most influential public policy organizations,” one SiegedSec member wrote on Telegram. “This organization is responsible for leading Project 2025, an authoritarian Christian nationalist plan to reform the United States government.” The hacked data, which was reviewed by CyberScoop, includes Heritage Foundation blogs and material related to The Daily Signal, a news website affiliated with the organization.
Heritage published Project 2025, its sweeping recommendations for a second Trump term, in April 2023, with a “broad coalition” of more than 100 other conservative organizations. The 900-plus-page “mandate for leadership” touches on virtually every sector of the executive branch, from the White House to the bevy of federal agencies under the president’s control.
Broadly speaking, its recommendations involve expanding presidential power, purging federal agencies of career employees, and replacing them with Trump loyalists. The mandate calls for dismantling entire federal departments — including the departments of Commerce, Education, and Homeland Security, the latter of which would be replaced with a new agency that is more extreme in its mission and less subject to oversight. Project 2025 also urges Trump to reverse the Food and Drug Administration’s approval of abortion pills, eliminate policies that promote “abortion as health care,” outlaw pornography, and shut down “telecommunications and technology firms that facilitate its spread.”
The mandate’s chapter on the Federal Communications Commission — written by Brendan Carr, the agency’s head under Trump — calls for the imposition of “transparency rules on Big Tech” and an overhaul of Section 230 of the Communications Decency Act, the law that states that “interactive computer services” can’t be treated as the publishers or speakers of third-party content published on their platforms. Carr claims that Section 230 allows tech companies to “censor protected speech,” echoing claims that major social media companies suppress right-wing viewpoints.
Trump recently attempted to distance himself from Project 2025, claiming he has “no idea” who is behind the plan — even though a CNN analysis found that more than 140 former members of his administration were involved in drafting the mandate.

Photo by Amelia Holowaty Krales / The Verge

A group of self-proclaimed “gay furry hackers” says it breached the Heritage Foundation earlier this month, releasing two gigabytes of the right-wing think tank’s internal data on Tuesday. On its Telegram channel, the hacktivist collective SiegedSec — which has previously claimed responsibility for hacking NATO’s computer systems — said the Heritage hack was part of its #OpTransRights campaign, which also targeted the far-right media outlet Real America’s Voice and the Hillsong megachurch. The group also cited their objections to Project 2025, the Heritage Foundation’s policy proposal for a second term for former President Donald Trump, as a motivating factor.

In an email to The Verge, Heritage Foundation spokesperson Noah Weinrich denied that Heritage had been hacked, calling it a “false narrative and an exaggeration by a group of criminal trolls trying to get attention.”

“An organized group stumbled upon a two-year-old archive of The Daily Signal website that was available on a public-facing website owned by a contractor,” Weinrich said in an emailed statement. “The information obtained was limited to usernames, names, email addresses, and incomplete password information of both Heritage and non-Heritage content contributors, as well as article comments and the IP address of the commenter. No Heritage systems were breached at any time, and all Heritage databases and websites remain secure, including Project 2025. The data at issue has been taken down, and additional security steps have since been taken as a precaution.”

The Heritage Foundation claims the alleged hack was an exaggeration for attention, but in the meantime, Mike Howell, the executive director of the foundation’s Oversight Project, has triumphantly taken credit for disbanding “the Gay Furry Hackers.”

SiegedSec has released chat logs of a conversation one of its members claimed to have with Howell on Signal. The chat logs show a person who claims to be Howell asking a SiegedSec member why the group hacked the Heritage Foundation and threatening to expose the hackers.

We are in the process of identifying and outting members of your group

Reputations and lives will be destroyed

Closeted Furries will be presented to the world for the degenerate perverts they are

You cannot hide Your means are miniscule compared to mine. You now can either turn yourself in or you can cooperate

Howell confirmed the legitimacy of the messages in an X exchange with a Daily Dot reporter.

Weinrich did not comment on the alleged messages between Howell and SiegedSec.

The messages also show Howell claiming to be “tied up with the fbi issuing a 2702 order” on SiegedSec’s social media. SiegedSec has indeed disbanded, a decision its members attribute to their “own mental health, the stress of mass publicity, and to avoid the eye of the FBI.”

A SiegedSec representative who goes by vio told The Verge they “completely expected” Heritage to deny that it had been hacked. “Many companies try denial to save face,” vio said. “The server we hacked was linked to The Daily Signal, and the server was named ‘first-heritage-foundation’. Clearly, Heritage was genuinely hacked.”

“Mike’s threats and insults showed anger that confirmed what Heritage denied,” vio said.

In a statement on Telegram, SiegedSec said the goal of the hack was to draw attention to — and combat — the Heritage Foundation’s anti-LGBT and anti-abortion policy proposals.

“The Heritage Foundation is a conservative think tank in America, among the most influential public policy organizations,” one SiegedSec member wrote on Telegram. “This organization is responsible for leading Project 2025, an authoritarian Christian nationalist plan to reform the United States government.” The hacked data, which was reviewed by CyberScoop, includes Heritage Foundation blogs and material related to The Daily Signal, a news website affiliated with the organization.

Heritage published Project 2025, its sweeping recommendations for a second Trump term, in April 2023, with a “broad coalition” of more than 100 other conservative organizations. The 900-plus-page “mandate for leadership” touches on virtually every sector of the executive branch, from the White House to the bevy of federal agencies under the president’s control.

Broadly speaking, its recommendations involve expanding presidential power, purging federal agencies of career employees, and replacing them with Trump loyalists. The mandate calls for dismantling entire federal departments — including the departments of Commerce, Education, and Homeland Security, the latter of which would be replaced with a new agency that is more extreme in its mission and less subject to oversight. Project 2025 also urges Trump to reverse the Food and Drug Administration’s approval of abortion pills, eliminate policies that promote “abortion as health care,” outlaw pornography, and shut down “telecommunications and technology firms that facilitate its spread.”

The mandate’s chapter on the Federal Communications Commission — written by Brendan Carr, the agency’s head under Trump — calls for the imposition of “transparency rules on Big Tech” and an overhaul of Section 230 of the Communications Decency Act, the law that states that “interactive computer services” can’t be treated as the publishers or speakers of third-party content published on their platforms. Carr claims that Section 230 allows tech companies to “censor protected speech,” echoing claims that major social media companies suppress right-wing viewpoints.

Trump recently attempted to distance himself from Project 2025, claiming he has “no idea” who is behind the plan — even though a CNN analysis found that more than 140 former members of his administration were involved in drafting the mandate.

Read More 

Illustration by Alex Castro / The Verge

Arm is launching its own upscaler for mobile gaming. The chip designer says its Arm Accuracy Super Resolution (ASR) can make games look better, while lowering power consumption on your phone. It’s also making the upscaling technology available to developers under an MIT open-source license.
Arm based its technology on AMD’s FidelityFX Super Resolution 2 (FSR 2), which uses temporal upscaling to make PC games look better and boost frame rates. Unlike spatial upscaling, which upscales an image based on a single frame, temporal upscaling involves using multiple frames to generate a higher-quality image.
Upscaling is especially useful for lower-powered graphics cards that couldn’t otherwise run a game at playable frame rates. AMD FSR, Nvidia DLSS, and Intel XeSS are all upscalers that work on desktop and laptop GPUs. Arm ASR is one of the few that will work on mobile phones, which could have a major impact on battery life.

Image: Arm

You can see just how Arm ASR stacks up to AMD’s FSR 2 and Qualcomm’s GSR tech in the above chart created by Arm. Arm claims ASR produced 53 percent higher frame rates than rendering at native resolution on a device with an Arm Immortalis-G720 GPU and 2800 x 1260 display, beating AMD FSR 2.
It also tested ASR on a device using MediaTek’s Dimensity 9300 chip and found that rendering at 540p and upscaling with ASR used much less power than running a game at native 1080p resolution.

Image: Arm

With Arm-based processors powering Copilot Plus PCs from companies like Microsoft, Dell, Lenovo, and Samsung, we could eventually see Arm’s lightweight ASR upscaler make its way to laptops, too. These laptops already have Microsoft’s new Automatic Super Resolution, which is exclusive to Snapdragon X chips, but Microsoft is working on another upscaling solution for a broader range of Windows devices.

Illustration by Alex Castro / The Verge

Arm is launching its own upscaler for mobile gaming. The chip designer says its Arm Accuracy Super Resolution (ASR) can make games look better, while lowering power consumption on your phone. It’s also making the upscaling technology available to developers under an MIT open-source license.

Arm based its technology on AMD’s FidelityFX Super Resolution 2 (FSR 2), which uses temporal upscaling to make PC games look better and boost frame rates. Unlike spatial upscaling, which upscales an image based on a single frame, temporal upscaling involves using multiple frames to generate a higher-quality image.

Upscaling is especially useful for lower-powered graphics cards that couldn’t otherwise run a game at playable frame rates. AMD FSR, Nvidia DLSS, and Intel XeSS are all upscalers that work on desktop and laptop GPUs. Arm ASR is one of the few that will work on mobile phones, which could have a major impact on battery life.

Image: Arm

You can see just how Arm ASR stacks up to AMD’s FSR 2 and Qualcomm’s GSR tech in the above chart created by Arm. Arm claims ASR produced 53 percent higher frame rates than rendering at native resolution on a device with an Arm Immortalis-G720 GPU and 2800 x 1260 display, beating AMD FSR 2.

It also tested ASR on a device using MediaTek’s Dimensity 9300 chip and found that rendering at 540p and upscaling with ASR used much less power than running a game at native 1080p resolution.

Image: Arm

With Arm-based processors powering Copilot Plus PCs from companies like Microsoft, Dell, Lenovo, and Samsung, we could eventually see Arm’s lightweight ASR upscaler make its way to laptops, too. These laptops already have Microsoft’s new Automatic Super Resolution, which is exclusive to Snapdragon X chips, but Microsoft is working on another upscaling solution for a broader range of Windows devices.

Read More 

Illustration: The Verge

Warner Bros. Discovery has been debuting its newer IP-based series on Max for the past few years in an effort to make it “the one to watch,” but the company is now switching things up with a big plan to rebrand multiple projects as — wait for it — HBO shows.
Though The Penguin and Dune: Prophecy were both initially billed as Max Originals, Deadline reports that Warner Bros. Discovery will now market them as HBO Originals as part of a larger effort to differentiate what kind of content is released under the two brand umbrellas.
Now, the shows will premiere simultaneously on both Max and HBO the way other big tentpoles like House of the Dragon and Succession have in the past. This always kind of seemed like a given because of the success of the Batman and Dune franchises and WBD CEO David Zaslav’s professed desire to flood the space with IP riffs. But what’s interesting is how the move speaks to the way WBD is rethinking its decision to prioritize Max — one of the least inspired brand names imaginable — over HBO, a name synonymous with the birth of cable television.
After revealing last month that the upcoming Harry Potter, Lanterns, and Welcome to Derry Max projects were being rebranded as HBO Originals, content head Casey Bloys was frank with Variety about how using IP as a delineator between the two platforms “started to feel unnecessary” as the shows began staffing up. Bloys explained that in many cases, the creative teams behind this new crop of shows “were using the same methods, the same kind of thinking” as those who have worked on traditional HBO shows. That tracks given how many HBO veterans are in the mix for Harry Potter and how Damon Lindelof’s (of Watchmen fame) Lanterns will take cues from True Detective.
But it also seemed like Bloys was admitting that WBD’s attempt to make Max happen simply by slapping the name onto a series people probably would have watched anyway was a bust and that the company had come around to a more sensible plan to “just call them what they are: HBO shows.”
Despite Zaslav’s best efforts, a lot of people still think of Max as HBO going through a truly wild identity crisis, like a corporation struggling to find itself after its parent company married some weird guy. This pivot back to HBO probably means that Max won’t exclusively premiere WBD’s bigger, more expensive new series that it expects to pull in larger audiences.
That doesn’t exactly mean that Max will be hurting for new material, but it does seem like WBD could have saved itself a lot of time and money by sticking to what was already working.

Illustration: The Verge

Warner Bros. Discovery has been debuting its newer IP-based series on Max for the past few years in an effort to make it “the one to watch,” but the company is now switching things up with a big plan to rebrand multiple projects as — wait for it — HBO shows.

Though The Penguin and Dune: Prophecy were both initially billed as Max Originals, Deadline reports that Warner Bros. Discovery will now market them as HBO Originals as part of a larger effort to differentiate what kind of content is released under the two brand umbrellas.

Now, the shows will premiere simultaneously on both Max and HBO the way other big tentpoles like House of the Dragon and Succession have in the past. This always kind of seemed like a given because of the success of the Batman and Dune franchises and WBD CEO David Zaslav’s professed desire to flood the space with IP riffs. But what’s interesting is how the move speaks to the way WBD is rethinking its decision to prioritize Max — one of the least inspired brand names imaginable — over HBO, a name synonymous with the birth of cable television.

After revealing last month that the upcoming Harry Potter, Lanterns, and Welcome to Derry Max projects were being rebranded as HBO Originals, content head Casey Bloys was frank with Variety about how using IP as a delineator between the two platforms “started to feel unnecessary” as the shows began staffing up. Bloys explained that in many cases, the creative teams behind this new crop of shows “were using the same methods, the same kind of thinking” as those who have worked on traditional HBO shows. That tracks given how many HBO veterans are in the mix for Harry Potter and how Damon Lindelof’s (of Watchmen fame) Lanterns will take cues from True Detective.

But it also seemed like Bloys was admitting that WBD’s attempt to make Max happen simply by slapping the name onto a series people probably would have watched anyway was a bust and that the company had come around to a more sensible plan to “just call them what they are: HBO shows.”

Despite Zaslav’s best efforts, a lot of people still think of Max as HBO going through a truly wild identity crisis, like a corporation struggling to find itself after its parent company married some weird guy. This pivot back to HBO probably means that Max won’t exclusively premiere WBD’s bigger, more expensive new series that it expects to pull in larger audiences.

That doesn’t exactly mean that Max will be hurting for new material, but it does seem like WBD could have saved itself a lot of time and money by sticking to what was already working.

Read More 

Seven different Arduino-based devices can be built with the Plug and Make Kit kit. | Image: Arduino

Arduino’s new starter kit features electronic components that can be connected without having to know anything about soldering or wiring. It’s similar to LittleBits’ electronic building blocks that make it easy for kids to learn about electronics but targets an older audience of beginners who eventually aspire to do more with an Arduino.
The $87.36 Arduino Plug and Make Kit comes with an Arduino UNO R4 WiFi microcomputer featuring Bluetooth and its own built-in 12 x 8 LED matrix display, plus seven different Modulino components that can all be connected using cables that snap into one-way connector ports, ensuring nothing can be wired incorrectly.

Image: Arduino
The kit includes an Arduino microcomputer and seven different electronic components each offering different functionality.

The bundled Modulinos include an adjustable knob, a strip of dimmable LEDs, a movement sensor, a time-of-flight distance sensor, a buzzer, a thermometer and humidity sensor, and a set of three buttons. To help keep things organized, all of the components, including the Arduino, can be attached to a baseplate using included screws.
The Arduino Plug and Make Kit includes tutorials on how to use each component and step-by-step instructions on how to assemble them to build seven devices, such as a video game controller, a synthesizer, a touchless lamp, and an electronic hourglass timer.

The Plug and Make Kit is also compatible with all of the hardware and software in the Arduino ecosystem, including the Arduino Cloud platform, which allows users to share project ideas. Once those seven included projects are complete, they can find inspiration for their next creation.

Seven different Arduino-based devices can be built with the Plug and Make Kit kit. | Image: Arduino

Arduino’s new starter kit features electronic components that can be connected without having to know anything about soldering or wiring. It’s similar to LittleBits’ electronic building blocks that make it easy for kids to learn about electronics but targets an older audience of beginners who eventually aspire to do more with an Arduino.

The $87.36 Arduino Plug and Make Kit comes with an Arduino UNO R4 WiFi microcomputer featuring Bluetooth and its own built-in 12 x 8 LED matrix display, plus seven different Modulino components that can all be connected using cables that snap into one-way connector ports, ensuring nothing can be wired incorrectly.

Image: Arduino
The kit includes an Arduino microcomputer and seven different electronic components each offering different functionality.

The bundled Modulinos include an adjustable knob, a strip of dimmable LEDs, a movement sensor, a time-of-flight distance sensor, a buzzer, a thermometer and humidity sensor, and a set of three buttons. To help keep things organized, all of the components, including the Arduino, can be attached to a baseplate using included screws.

The Arduino Plug and Make Kit includes tutorials on how to use each component and step-by-step instructions on how to assemble them to build seven devices, such as a video game controller, a synthesizer, a touchless lamp, and an electronic hourglass timer.

The Plug and Make Kit is also compatible with all of the hardware and software in the Arduino ecosystem, including the Arduino Cloud platform, which allows users to share project ideas. Once those seven included projects are complete, they can find inspiration for their next creation.

Read More