Anthropic’s AI models could potentially help spies analyze classified documents, but the company draws the line at domestic surveillance. That restriction is reportedly making the Trump administration angry.
On Tuesday, Semafor reported that Anthropic faces growing hostility from the Trump administration over the AI company’s restrictions on law enforcement uses of its Claude models. Two senior White House officials told the outlet that federal contractors working with agencies like the FBI and Secret Service have run into roadblocks when attempting to use Claude for surveillance tasks.
The friction stems from Anthropic’s usage policies that prohibit domestic surveillance applications. The officials, who spoke to Semafor anonymously, said they worry that Anthropic enforces its policies selectively based on politics and uses vague terminology that allows for a broad interpretation of its rules.
Garmin’s newest watch is an update to its Venu line. The Venu 3 was one of the brand’s best lifestyle watches, and the Venu X1, launched earlier this year, was a premium version. Now, there is a Venu 4, available on September 22—and it has a flashlight.
Pricing and essential features
The Venu 4 comes in two sizes (41 and 45 millimeters), the same as the sizes of the Venu 3 and 3S respectively. It costs $100 more than the older version, with the Venu 4 coming in at $549.99 compared to the Venu 3 at $449.99.
Like its predecessor, the Venu 4 has an AMOLED touchscreen, stainless steel bezel, Gorilla Glass 3 face, and a speaker and microphone that allow it to take voice calls if your phone is within range.
It adds training status, training readiness, and Garmin Coach for four different activities: running, cycling, strength, and a new fitness coach. Here’s a rundown of some of the notable new features:
Flashlight
It’s a simple thing, but a big one for a lot of people. Garmin used to reserve LED flashlights for its more outdoorsy adventure watches—the Instinct on the low end and the Fenix on the high end—but the Forerunner 970 has one, and it seems to have gone over well. Even Amazfit added a flashlight to its T-Rex 3 Pro. The Venu 3 didn’t have a flashlight, but the premium Venu X1 does, and now the Venu 4 does, too.
Health status
A new health status metric combines your nighttime vitals (heart rate, HRV, skin temperature, and pulse oxygen) and tells you whether these are consistent with what’s normal for you, or whether they’re trending away from your usual baseline.
Lifestyle logging
Garmin has historically just collected your data and displayed it, without the kind of lifestyle analysis provided by others like Whoop and Bevel. But the Venu 4 comes with a lifestyle logging feature, where you can take note of habits and behaviors and see whether they affect your metrics. (One example Garmin gives is logging when you have caffeine late in the day, so you can see how much it affects your sleep metrics.)
Fitness coach
Garmin’s daily suggested workouts aren’t just for runners and cyclists anymore. There is a relatively new strength coach (I’ve had it on my Forerunner watches for a while now) and the Venu 4 adds a fitness coach that can recommend workouts for you from any of 25 of activities. A given day’s workout might recommend that you go for a walk, do a HIIT session, or hop on the rowing machine.
Accessibility features
The Venu was already one of Garmin’s more disability-friendly watches, featuring a wheelchair mode in addition to walking and running workouts. The Venu 4 adds a spoken watch face option, that will read you the time using the speaker, and an option to filter the colors on the display.
An anonymous reader quotes a report from Polygon: The CEOs of Discord, Steam, Twitch, and Reddit have been called to Congress to testify about the “radicalization of online forum users” on those platforms, the House Oversight and Government Reform Committee announced Wednesday. “Congress has a duty to oversee the online platforms that radicals have used to advance political violence,” said chairman of the House Oversight Committee James Comer, a Republican from Kentucky, in a statement. “To prevent future radicalization and violence, the CEOs of Discord, Steam, Twitch, and Reddit must appear before the Oversight Committee and explain what actions they will take to ensure their platforms are not exploited for nefarious purposes.”
Letters from the House Oversight Committee have been sent to Humam Sakhnini, CEO of Discord; Gabe Newell, president of Steam maker Valve; Dan Clancy, CEO of Twitch; and Steve Huffman, CEO of Reddit, requesting their testimony on Oct. 8. “The hearing will examine radicalization of online forum users, including incidents of open incitement to commit violent politically motivated acts,” Comer said in a letter to each CEO. […] Discord, Steam, Twitch, and Reddit execs will have the chance to deliver five-minute opening statements prior to answering questions posed by members of the committee during October’s testimony.
Senator Elizabeth Warren (D-MA) and Representative Melanie Stansbury (D-N.M.) are leading a group of congressional Democrats in investigating White House Special Advisor David Sacks for possibly serving in his position for longer than he’s allowed. Sacks, a former PayPal executive and venture capitalist at Craft Ventures, was originally picked by President Donald Trump to be the “White House A.I. & Crypto Czar” in 2024.
“Any effort to stay beyond the time limits imposed on you as a Special Government Employee (SGE) would raise additional ethics concerns for you and the Trump Administration,” the group writes in a letter to Sacks,”particularly as it moves to implement recently enacted cryptocurrency legislation and put in place new rules for the crypto industry.”
Besides being friendly with the Trump campaign and allies like Elon Musk, Sacks was given his position because of his knowledge of the crypto and AI industries as an investor. That poses an obvious conflict of interest, something that’s only waived during the 130-day limit that SGEs are supposed to serve. As Warren and the other Democrats backing the investigation note, though, it’s possible Sacks has been working in his role for longer than that.
“If you have worked every calendar day since the presidential inauguration, your 130th day of work in this role was May 29, 2025,” the group writes. “If you have worked every business day, your 130th day was July 25, 2025. As of the date of this letter, it is the 167th business day of this Administration.”
As part of the investigation, Sacks is expected to offer a more detailed account of when and how he works in his advisory role, including if he answers government emails while working in Silicon Valley. Congressional Democrats are trying to verify if norms have been violated to make sure that they won’t be violated in the future, but there are larger ethical concerns to contend with, too.
The second Trump administration has been friendly to the crypto industry, likely thanks in part to the influence of Sacks. Trump signed an executive order calling for the creation of a federal Bitcoin stockpile and signed the GENIUS Act into law in July, establishing a regulatory framework for stablecoins, a type of cryptocurrency typically tied to the value of the US dollar. Continuing to serve in his role without leaving his position at Craft Ventures or disclosing his investments would only raise more questions about how Sacks stands to benefit from advising on regulation.
This article originally appeared on Engadget at https://www.engadget.com/general/democrats-are-investigating-trump-crypto-advisor-david-sacks-over-a-possible-sge-violation-215758087.html?src=rss
Health secretary and anti-vaccine activist Robert F. Kennedy Jr. plans to roll back access to lifesaving vaccines for children, and has refused to even speak with staff scientists and subject-matter experts at the Centers for Disease Control and Prevention about evidence-based recommendations. That’s according to former CDC officials who testified before the Senate on Wednesday.
The Senate Committee on Health, Education, Labor, and Pensions (HELP) called ex-CDC director Susan Monarez to review the chaos that has engulfed the public health agency under Kennedy. Monarez, a microbiologist and long-serving federal employee, led the CDC as the first Senate-confirmed director for just 29 days before her dramatic ouster last month. She appeared before the HELP committee alongside Debra Houry, the former chief medical officer for the CDC. Houry had worked at the agency for a decade—spanning four administrations and six directors— before resigning in protest against Kennedy’s leadership soon after Monarez’s ouster.
Monarez’s ouster
Much of their testimony today was alarming, but not surprising. Upon her exit, Monarez claimed that she was fired because she refused Kennedy’s demand that she agree in advance to approve changes to the CDC’s childhood vaccine recommendations regardless of whether any scientific evidence supported the changes. She also claimed that Kennedy demanded that she fire CDC scientific leadership without cause, which she also refused to do. Similarly, when Houry resigned, she said Kennedy was censoring science, steamrolling CDC experts, and spreading misinformation. In the hearing today, the two stood by their previous comments and provided more details.
Apple makes good devices and accessories, but then do tend to be more expensive than the competition, even for something as basic as a charging cable. It’s certainly true of Apple Watch bands—I’m not sure why a braided strip of polyester and silicone costs $99, but that’s what the company charges. But right now, Woot is selling a whole rainbow of Apple Watch bands for up to 70% off. It’s a great (and rare) opportunity to stock up on a few different styles or colors to outfit your smartwatch.
You can choose from the Solo Loop or the Braided Solo Loop designs, which normally go for $49.99 and $99.99, respectively. Solo Loops are on sale for $14.99, and Braided Solo Loops are $29.99. You can find them in virtually every color— even the rainbow Pride edition of the Braided Solo Loop.
All of these bands are compatible with Apple Watch SE, Apple Watch Series 4 or later, and Apple Watch Ultra or later. They aren’t adjustable, so you’ll need to select the right size, with the most popular sizes likely to run out first, so maybe set an alarm to remind you to place an order before they do.
The past few months we have been intrigued by an AMD GFX1250 target added to the LLVM codebase for the AMDGPU shader compiler back-end. GFX12 is RDNA4 and GFX1250 is presumably some “RDNA 4.5” / “RDNA Refresh” part akin to GFX1150 having been for the RDNA 3.5 parts with Strix Halo / Strix Point. The prior LLVM code confirmed GFX1250 is in APU form factor but product details beyond that have been scarce. Today a new AMD GFX1251 target was merged to LLVM…
Microsoft is now prioritizing Anthropic’s Claude 4 over OpenAI’s GPT-5 in Visual Studio Code’s auto model feature, signaling a quiet but clear shift in preference. The Verge reports: “Based on internal benchmarks, Claude Sonnet 4 is our recommended model for GitHub Copilot,” said Julia Liuson, head of Microsoft’s developer division, in an internal email in June. While that guidance was issued ahead of the GPT-5 release, I understand Microsoft’s model guidance hasn’t changed.
Microsoft is also making “significant investments” in training its own AI models. “We’re also going to be making significant investments in our own cluster. So today, MAI-1-preview was only trained on 15,000 H100s, a tiny cluster in the grand scheme of things,” said Microsoft AI chief Mustafa Suleyman, in an employee-only town hall last week.
Microsoft is also reportedly planning to use Anthropic’s AI models for some features in its Microsoft 365 apps soon. The Information reports that the Microsoft 365 Copilot will be “partly powered by Anthropic models,” after Microsoft found that some of these models outperformed OpenAI in Excel and PowerPoint.
When I was a child, SimCity 2000 felt like a fun, animated set of urban-themed Lego blocks to tinker with. Revisiting the game roughly three decades later, though, I’ve found the weight of my adult responsibilities tempering my role as god-mayor of a tiny metropolis.
The tough economics of establishing a thriving city barely concerned me as a child. Rather than building up a durable tax base from a slowly growing city of happy citizens, I’d usually type in an infinite money cheat or load up the handy Urban Renewal Kit expansion to build whatever I wanted, wherever I wanted, as quickly as possible.
A blank canvas, ready for me to destroy.
Credit:
Maxis
Thus unleashed, my childhood self would go mad with unchecked power, petulantly turning dials just to see what happened to the citizens in my virtual ant farm. Sometimes I’d try to arrange a repeating grid of fancy arcologies and police stations, trying to create a regimented utopia out of the game’s most expensive (and therefore “best”) building type. More often, I’d play with the far edges of the simulation, crowding residential areas next to polluting heavy industry or letting entire neighborhoods go without fire protection and waiting to see how long it took for things to fall apart.
The long-anticipated LMDE (Linux Mint Debian Edition) 7 (codename Gigi) operating system is now available for public beta testing, featuring the latest Cinnamon desktop environment and a Debian Trixie base.
Google and Apple’s app stores both have a reputation for being pretty trustworthy these days. It’s easy to assume that if an app is in either the iPhone App Store or the Google Play Store, it’s safe to download. But a new Google crackdown this week reminds us that this isn’t so clear cut.
Yesterday, a third-party security report revealed that Google had recently removed 224 malicious apps from the Android Play Store. Dubbed “SlopAds” apps by security company Human, which discovered the apps and wrote the report, these apps evaded Google’s usual security procedures and instead used a clever workaround to secretly install malware on users’ devices, even when downloaded straight from Google’s servers.
The way these apps worked was that, if you downloaded them by searching for them through the Google Play Store, they would work as advertised with no malware dragging them down. Generally, these apps were pitched as simple utilities, or attempted to pass themselves off as more popular programs like ChatGPT, to try to trick users into downloading them. Not the best tools, certainly, but if accessed directly through Google, they wouldn’t hurt you.
But the trick is that, if you downloaded one of these apps after arriving at the Play Store via one of SlopAds ad campaigns, it would also secretly download an encrypted configuration file that, after a few post-download checks, would install malware on your device.
Once a device was infected, the app would then steal its information, and start using it to generate fake ad impressions on sites run by the scammers, maximizing profit.
It was a clever way to get around Google’s regular review process, and a good reminder that, even as major companies try to make their app stores safe to use, you should still be vigilant while browsing them.
How to avoid installing malware on your device
While SlopAds has been thwarted for now, you should still take a few steps to keep your device safe while downloading new apps, especially since SlopAds isn’t alone in sneaking Malware onto the Play Store. Here are just a few ways to protect your device while browsing for new apps.
Download your apps directly from the Play Store
Android is different from iOS in that it allows you to sideload apps onto your device. This can be convenient when working with smaller developers, who might not have the resources to get their programs on the Play Store. But downloading an app that hasn’t been verified by Google opens you up to extra risk. Always ensure you trust a developer and the specific APK file you’re using before sideloading an app. Google is currently working to block sideloading unless a developer is verified (which has been controversial, despite the extra security it gives you), although these changes aren’t set to hit most of the world until 2027.
Find apps through the Play Store
As SlopAds proves, navigating to an app through an outside source can flag it to download extra files to your device that you won’t get if you find the app through the Play Store’s own search functionality. Always be cautious about links to apps that you find on suspicious websites, and especially in ads. Using the Play Store to discover new apps instead could save you some headaches down the line, especially as Google’s search is less likely to send you to suspicious apps than it is popular apps that have been verified as safe by other users.
Check user reviews and permissions
If you scroll down before downloading an app through the Play Store, you’ll see which permissions an app needs to do its work on your phone, and you’ll also be able to read user reviews. This can be handy if there are known issues with an app, or the requested permissions seem to be a bit too generous for what the app claims to do. However, this isn’t a catch-all solution—SlopAds did its malicious behavior in the background without needing any sort of permissions, and depending on where a user downloaded a SlopAd app from, it might not have even installed malware on their device. It’s possible apps with good reviews could still have harmful software included on the sly.
Turn on Google Play Protect
While SlopAds was able to bypass Google Play Protect, it’s still a good idea to turn it on if it isn’t enabled on your device already. This will scan an app for known malware before downloading it, giving you an extra layer of protection. It will also periodically scan apps already installed on your device. To ensure it’s turned on, open the Play Store, click your profile icon in the top right corner, and navigate to Play Protect > Settings. To scan sideloaded apps, you can also turn on the Improve harmful app detection setting, which is in the same spot.
Run a Google Security Checkup
Finally, you can run a Google Security Checkup on your device via a web browser. This will help you tighten your online security, encouraging you to take security steps like adding a recovery email or phone number to your Google account. It’ll also list recent security activity, and will generally ensure that, even if a malicious app steals your data, you’ll be able to lock it out of your account with the least issue possible.
Governments worldwide are implementing heat protection laws as 2.4 billion workers face extreme temperature exposure and 19,000 die annually from heat-related workplace injuries, according to a World Health Organization and World Meteorological Organization report.
Japan imposed $3,400 fines for employers failing to provide cooling measures when wet-bulb temperatures reach 28C. Singapore mandated hourly temperature sensors at large outdoor sites and requires 15-minute breaks every hour at 33C wet-bulb readings. Southern European nations ordered afternoon work stoppages this summer when temperatures exceeded 115F across Greece, Italy and Spain.
The United States lacks federal heat standards; only California, Colorado, Maryland, Minnesota, Nevada, Oregon and Washington have state-level protections. Boston passed requirements for heat illness prevention plans on city projects. Enforcement remains inconsistent — Singapore inspectors found nearly one-third of 70 sites violated the 2023 law. Texas and Florida prohibit local governments from mandating rest and water breaks.
Zwift Racing League (ZRL) is the most popular race series on Zwift, with over 30,000 riders participating in each of the past two seasons! Many new riders join each round. Maybe that’s you?
If so, you may be confused by how Points Races work in ZRL. (Don’t feel bad, we all were when we started!) Here’s a complete answer.
Intermediate Segment Points (FAL and FTS)
Certain sprint and KOM sections on the course are designated as “intermediate” segments in ZRL points races. There are two ways to earn points on these intermediates: First-Across-the-Line (FAL) and Fastest-Through-Segment (FTS).
First-Across-the-Line (FAL)
FAL points are given to everyone who crosses the segment’s finish line each time that segment is attempted, with points based on the number of riders in the race and your position across the line. The first rider across the line gets points equal to the number of riders (eg, 60 riders=60 points) in the race. Using that 60-rider race example, 2nd would get 59 points, 3rd 58 points, etc.
Position
1st
2nd
3rd
4th
5th
6th
7th
8th
9th
…
Last
FAL (per lap)
N
N-1
N-2
N-3
N-4
N-5
N-6
N-7
N-8
…
1
N=number of riders in the race
Fastest-Through Segment (FTS)
FTS points are given to the top 5 fastest times through a segment over the course of the entire race. If a segment is designated multiple times in a race, a single rider could potentially have multiple top-5 FTS times. For example (see table below) a rider who got 1st in FTS on the first lap and 4th on the next lap would earn 10 + 4 = 14 FTS points for the race. In a 5-lap race, a strong rider could (theoretically) grab all 5 FTS spots!
Position
1st
2nd
3rd
4th
5th
FTS (whole race)
10
8
6
4
2
Simply put, the highest FAL points go to whoever is at the front of the race on designated segments. FTS points, on the other hand, are given to the fastest times on given segments, and your overall position across the segment finish line is of no consequence.
Finish Line Points (FIN and PBP)
Along with intermediate segment points, riders earn points based on their position across the finish line. Known as FIN points in the world of ZRL, these are computed just like FAL points above:
Position
1st
2nd
3rd
4th
5th
6th
7th
8th
9th
…
Last
FIN
N
N-1
N-2
N-3
N-4
N-5
N-6
N-7
N-8
…
1
Along with FIN points, additional Podium Bonus Points (PBP) are given to the top 5 finishers:
Position
1st
2nd
3rd
4th
5th
PBP
10
8
6
4
2
Race Example
This week, A/B riders in ZRL are racing four laps of Makuri Islands’ Fine and Sandy route. This route has two intermediate segments which will be awarding both FTS and FAL points:
On the first Boardwalk Sprint, every rider earns FAL points. The first rider across the line earns points equal to the total number of riders in the race (60 riders means 1st place earns 60 points), 2nd rider across the line earns 59, all the way down to the last-placed rider earning just 1 point. Everyone earns FAL points each time they pass through the arch.
Additionally, on the first Boardwalk Sprint, the top 5 segment times are slotted into the leaderboard for FTS. If the race was to end right now, those 5 riders would each FTS points of 10 for 1st place, 8 for 2nd, down to 2 for 5th.
But the race isn’t over yet. We’ll be going through the Boardwalk Sprint three more times!
Riders will hit the Tidepool Sprint next, and it works just like the Boardwalk Sprint. (Intermediate segments are never related in any way, so whether there is just one intermediate or several, they all act independently, awarding FAL points each time through, and building a top-5 leaderboard across the duration of the event.)
On the next lap, FAL points work just like they did on the first lap. First rider across the line earns points equal to the total number of riders in the race, 2nd rider earns one point less, etc.
But FTS points are awarded to the 5 fastest times through the segment across the duration of the race. Think of it as a leaderboard, why whenever a rider sets a segment time faster than one currently on the leaderboard, that rider’s time gets added to the leaderboard, and the slowest time gets bumped so only the top 5 are ever listed.
On the second lap, a strong rider who set the fastest FTS time on the first lap might better their time and end up with the 1st and 2nd place listings on the invisible FTS leaderboard.
And so it continues, with each lap handing out a pile of FAL points, and new fast FTS times replacing slower ones on the top-5 leaderboard.
Eventually, riders will have finished all the intermediate segments, and only the finish line will be up the road. Riders will jostle for position, go all-in on the final sprint, and when the dust clears, FIN and PBP will be given to riders based on their position across the finish line.
The first rider across will earn FIN points equal to the number of riders in the race plus 10 PBP, the second rider will earn 1 less FIN point and 8 PBP, all the way down to the last-placed rider earning a single FIN point.
Best-Case Scenario
If you understand the explanation above, then you can understand how we get our Points Distribution charts, like this one for this week’s race:
Points Distributions, Fine and Sandy (4 Laps)
This chart shows the maximum points a team of 6 could earn in the race with 60 participants. No team will ever do this well, of course, but it lets you see, at a glance, the importance of segment points vs finishing points for a particular race. Here’s how we calculated the values for this particular chart:
If a team of 6 crossed the Boardwalk Sprint finish line in front of everyone else every lap, they would grab 60+59+58+57+56+55=345 points each lap, for a total of 4×345=1380 Sprint FAL points
If a team of 6 crossed the Tidepool Sprint finish line in front of everyone else every lap, they would grab 60+59+58+57+56+55=345 points each lap, for 4×345=1380 additional Sprint FAL points
If the team set all 5 top FTS times for the Boardwalk Sprint, they would earn 10+8+6+4+2=30 FTS points
If the team set all 5 top FTS times for the Tidepool Sprint, they would earn 10+8+6+4+2=30 additional FTS points
If that same team of 6 were the first across the finish line, they would earn 70+67+64+61+58+55=375 FIN+PBP points
That gives us these maximum points for a team of 6:
Sprint FAL Points: 2760
Sprint FTS Points: 60
KOM FAL Points: 0
KOM FTS Points: 0
FIN and PBP Points: 375
Segment Points Minutiae
Which Segments?
With the way WTRL sets up ZRL race routes, they don’t have to give segment points on every sprint or KOM segment on a route. They could designate certain segments as FAL only, FTS only, or as not giving points at all! That said, WTRL normally gives FAL and FTS points on all segments on the route, probably for simplicity’s sake.
Tracking FTS On Leaderboards
The leaderboards on the left side of your screen are a great way to see who grabbed FTS points, but that only works if your race has you going through a particular segment only one time. Because leaderboards only display each rider’s fastest time on the segment, in multi-lap races they probably won’t reflect the actual top 5 fastest times, since a single rider may have multiple top 5 fastest times.
I sure would love to see Zwift modify leaderboard functionality for points races, so we could see our team points totals in real time!
Seeing FTS and FAL Results on ZwiftPower
If you want to see who grabbed segment points after your race, head over to ZwiftPower, pull up your event, and then click on “Primes” at the top. Here you can toggle “First over line” and “Fastest time” to see FAL and FTS times, respectively.
Pro tip: you can actually look at these times mid-race. But you may want to leave that job to your DS!
Annulled Results and FAL
If a rider’s result is annulled (perhaps they don’t finish the race, or they get a DQ), any intermediate points they earned are effectively removed from the race. The annulled rider “keeps” those points – they are not passed on to the next-best rider. But they are also not included in the team total.
Your Thoughts
Watch for a followup post discussing strategic tips for maximizing your segment points.
Until then, if you have questions or segment point tips, share below!
While down to AMD Austin yesterday for the Instinct MI355X and ROCm 7.0 launch, I had the chance to chat again with Anush Elangovan. As the VP of AI Software at AMD, talking with Anush is always insightful and technical in nature. One of the questions I posed him was around the length of hardware support with ROCm…
The ability of AI displace humans at various tasks is accelerating quickly, Anthropic CEO Dario Amodei said at an Axios event on Wednesday. From the report: Amodei and others have previously warned of the possibility that up to half of white-collar jobs could be wiped out by AI over the next five years. The speed of that displacement could require government intervention to help support the workforce, executives said.
“As with most things, when an exponential is moving very quickly, you can’t be sure,” Amodei said. “I think it is likely enough to happen that we felt there was a need to warn the world about it and to speak honestly.” Amodei said the government may need to step in and support people as AI quickly displaces human work.
It is difficult to even comprehend the scale of modern photolithography. Current-generation manufacturing processes are described with terms like “2 nanometer” and “18 angstrom”. Those numbers undersell the actual size of the logic gates, but the point is, we’re working at a scale so small that the wavelengths of light used for the etching
House Oversight and Government Reform Committee Chairman James Comer (R-KY) has asked the CEOs of Discord, Twitch, Valve and Reddit to testify at a hearing on online radicalization. The hearing will be held on October 8, 2025, and is in direct response to the assassination of political activist Charlie Kirk, an event some have tried to connect to the online communities the alleged shooter, Tyler Robinson, participated in.
“Congress has a duty to oversee the online platforms that radicals have used to advance political violence,” Comer shared in the press release announcing the hearing. “To prevent future radicalization and violence, the CEOs of Discord, Steam, Twitch, and Reddit must appear before the Oversight Committee and explain what actions they will take to ensure their platforms are not exploited for nefarious purposes.”
Following Kirk’s death, law enforcement, regulators and the press have exerted significant effort to both understand Robinson’s actions and structure them into a coherent narrative. The fact that he seemingly engraved the bullets he used with references to furry memes and the game Helldivers implies he may have been immersed in online culture, and could have been influenced by who he interacted with there. But a recent report on his Discord chats suggests his motivations are hard to pin down, even for his friends.
In general, online platforms don’t escape scrutiny after tragic events, so these sorts of hearings are to be expected. Twitch and Discord were both investigated by the New York and New Jersey Attorney Generals following a 2022 shooting in Buffalo, New York, for example. Given the reaction to Kirk’s death, though, it remains to be seen how much the House Oversight Committee actually wants answers from online platforms, and how much it wants to lay the blame at their feet.
This article originally appeared on Engadget at https://www.engadget.com/social-media/house-committee-asks-discord-valve-twitch-and-reddit-to-testify-on-online-radicalization-184100470.html?src=rss
The latest Apple Watches now boast a sleep score, which gives you a number on a scale of 1 to 100 for how well you slept. But that’s not all—you can even get this number without the Series 11 watch, as I discovered this morning while waiting for my new watch to be delivered. To get the new sleep score, just make sure you have iOS 26 and watch OS 26 installed.
Where to find the sleep score
You can view the sleep score on your iPhone or on your Apple Watch. On the watch (Series 6 or later), go to the Sleep app. The sleep score is the first thing it will show you. On the iPhone, the sleep score gets its own card in the Apple Health app. If you don’t see it, tap the search icon and then tap on Sleep.
Many devices can give you an Apple sleep score, not just the newest Apple Watch
The sleep score is a data analysis feature that doesn’t require specific Apple Watch hardware, and Apple has said that Apple Watches as old as Series 6 will get it. What’s more, I noticed that my Apple Health app gives me sleep scores for nights I wore other devices to bed, but not an Apple Watch.
According to my settings, my Apple Health app receives sleep data from the Coros, Garmin, Zepp (Amazfit), Oura, Ultrahuman, Whoop, and Withings devices I’ve used. I’ve rarely worn an Apple Watch to bed lately, but I’m always wearing something, usually including an Oura ring. I have sleep scores for all of those nights, not just the nights I wore an Apple Watch.
Your past sleep scores are calculated retroactively
The sleep score is new with iOS 26 and watch OS 26, but the data needed to calculate it was there anytime you wore a supported device to bed. Scrolling back, I can see sleep scores going back to April of 2020.
How sleep scores are calculated
Apple’s sleep score comes with a breakdown of where your 100 potential points come from. This includes:
50 points for how long you slept
30 points for how consistent your bedtime was with previous nights
20 points for minimizing interruptions
My own scores include a lot of “high” and “excellent” ratings, even though I don’t think my sleep has been great. But sleep scores are never totally scientific—even when they’re based on solid data, there’s no such thing as a medical cutoff for what counts as “high” or “excellent” sleep. The World Sleep Society tells us to “focus on trends and patterns, not individual ‘scores.’” If you see your sleep scores trending up (or down), that’s more important than what exactly your number is today.
