Watch along today as NASA's Boeing Starliner Crew Flight Test finally — most likely — blasts off to the International Space Station (ISS). NASA should start streaming its coverage at 6:30PM ET on its YouTube channel, with the official launch set for 10:34PM ET. The spacecraft will carry two astronauts, Suni Williams and Butch Wilmore.

We say "most likely" because the road to this day has not been a smooth one. It was a decade ago that NASA first chose Boeing and Space X to construct spacecraft that would fly from the United States to the ISS. Boeing received a $4.2 billion contract, while NASA gave SpaceX $2.6 billion. Yet, the latter had its first successful crewed flight in 2020 and has replicated it about a dozen times since.

Boeing's Starliner failed to reach orbit during its first uncrewed orbital test flight in 2019 due to too much fuel burning. A follow-up flight was scheduled for August 2021 but was scrapped due to a valve issue, with Boeing finally reaching the ISS in spring 2022 with an uncrewed vessel. Two plans for crewed flights came and went, amongst faults in aspects such as the parachute system. Last August, Boeing announced it should have these issues straightened out by March 2024.

It's now two months later than that initial goal and Boeing, Williams and Wilmore seem prepared for take off. "We are ready, the spacecraft's ready and the teams are ready," Wilmore told the press. NASA associate administrator Jim Free added: "The first crewed flight of a new spacecraft is an absolutely critical milestone. The lives of our crewmembers Suni Williams and Butch Wilmore are at stake — we don't take that lightly at all."

Facebook is a hotbed for conspiracy theories and dangerous organizing at critical moments, like the 650,000-plus posts arguing against President Biden's victory between the 2020 general election and the January 6 insurrection. Some users scattered following the latter and subsequent prosecutions, but a new report first published by Wired shows a resurgence, identifying about 200 groups and profiles across the platform organizing militia activity nationwide.

The research, conducted by the Tech Transparency Project, found these groups have ties to organizations such as the Three Percenters militia network, dubbed by Meta as an "armed militia group" in its 2021 Dangerous Individuals and Organizations List. Yet, groups such as the Free American Army have urged users to join their local militia or the Three Percenters without consequence (Meta took down the Free American Army group only after Wired enquired about it, calling Facebook an "adversarial space" that requires regular investment to stay safe).

Katie Paul, director of the Tech Transparency Project, has watched hundreds of these groups and people since 2021 and has seen an increased seriousness and focus on organizing over the previous year. "Many of these groups are no longer fractured sets of localized militia but coalitions formed between multiple militia groups, many with Three Percenters at the helm," Paul told Wired. "Facebook remains the largest gathering place for extremists and militia movements to cast a wide net and funnel users to more private chats, including on the platform, where they can plan and coordinate with impunity."

The Tech Transparency Project found users seek out "active patriots" to discuss anti-government ideology, attend meetings and take combat training. The latter lends itself to a common theme: being prepared for standing up against or even going to war against enemies such as drag queens, pro-Palestine college students and the government itself.

Take a recent post by the administrator of a group called the Pennsylvania Light Foot, which has over 1,000 members: "In light of the violence and uncertainty in the world, Covid 19 shortages, civil unrest, and potential for terrorist attacks and natural calamity, we exist to equip our members. Our aim is to equip them with the ability to defend themselves, whether it be a mugger on the street or foreign soldier on our lawn." These sentiments are echoed by other extremist organizers across Facebook. 

Meta has attempted to at least create a facade of action and transparency. In 2019, it launched the Oversight Board as an independent reviewer of its content moderation. While the entity has pointed to Facebook's role in dangerous election rhetoric, including incidents outside the United States, critics argue it hasn't been impactful enough. Now, The Washington Post reports that layoffs at the Oversight Board could be imminent.

On August 14, Meta will shutter CrowdTangle, a tool it bought in 2016 that allowed journalists and academics to see how conspiracy theories and false information moved on Facebook and its sister site Instagram — often showcasing the platforms' shortcomings. The company is replacing it with the Meta Content Library, which not only appears to be less detailed but isn't available to for-profit news organizations.

Audible is testing a new category of book recommendations based on what a user watched recently on Prime Video. Which, as the name suggests, will show you audiobooks based on what you watch on the Amazon-owned service, TechCrunch reports.

The new carousel should appear on mobile and web apps for about half of users who have Amazon Prime Video and Audible subscriptions. You might see recommendations as straightforward as the book a movie you watched is based on or titles with storylines or authors that users with similar preferences to you have enjoyed.

Audible claims the decision came due to the uptick it saw in users accessing titles recently released as shows or movies. "There is a natural synergy between TV, movies, and books, and we see that clearly in how our customers engage with content on Audible," Andy Tsao, chief product and analytics officer at Audible, said in a statement. The company gives examples such as Reacher, which came out on Amazon Prime in 2022. Audible claims that the listenership of author Lee Child's books rose by almost 80 percent daily in the two weeks after its release.

Another day, another company tests Apple's resolve. This time, it's TikTok allegedly violating the company's rules for apps, with the video platform seemingly allowing some users to purchase its coins directly from its website. TikTok has apparently given some iOS users the option to "Try recharging on tiktok.com to avoid in-app service fees" — namely Apple's 30 percent commission on purchases.

According to photos shared on X (formerly Twitter) by David Tesler, co-founder of the app Sendit, TikTok is prompting users to save around 25 percent when purchasing coins (used to tip creators) thanks to lower third-party service fees. They can then use Apple Pay, PayPal or a credit or debit card to complete their transaction. It's unclear why only some users have access to this circumnavigation; one hypothesis is the feature was turned on for individuals who previously purchased a large number of coins.

Apple notably kicked Fortnite off its app store in 2020 after Epic Games introduced discounts on the game's currency for anyone who directly purchased it. The incident set off a multi-year legal battle, with Apple reinstating Epic Games' developer account in March after the European Union began looking into the situation. More recently, Apple has faced pushback from Spotify and rejected updates that would have displayed the music streamer's pricing and allowed in-app plan purchases. 

Meta is bringing some of its existing safety features to Threads, including the Hidden Words tool. The platform — which recently hit more than 150 million users — will now allow you to filter out words, phrases and even emojis, just like you can on Instagram.

Threads will have the setting on by default and filter out anything that it deems offensive, along with content that could be irrelevant or uninteresting. You can then add other things that you want to be filtered out, such as "weight loss," "assault" or a derogatory term. To add your own options, go to "manage custom words and phrases" and type in anything you're sick of seeing on Threads.

The platform is also testing two additional features: muting and quote controls. The first would allow you to mute notifications specifically for any interactions with your posts. This way, you could still get alerts about new followers or tags without seeing every response. Quote controls would let you choose who can quote your post and even let you unquote yourself. Meta has yet to announce when these two features might be available to all Threads users.

It's a post-Digital Markets Act (DMA) world, and Spotify continues to test what that means for its iOS app. The music streamer announced that it submitted an update for Apple's approval that would allow Spotify to display "basic pricing and website information" on its app in Europe and "the bare minimum outlined under the European Commission's ruling in its music streaming case." 

In the news, shared in a post on X (formerly Twitter), Spotify's chief public affairs officer Dustee Jenkins further stated, "By charging developers to communicate with consumers through in-app links, Apple continues to break European law. It's past time for the Commission to enforce its decision so that consumers can see real, positive benefits."

Apple and Spotify have consistently butted heads over what the latter can and can't do with its iOS app. Following the DMA going into effect, Spotify submitted an update to Apple that would have allowed users to purchase plans directly from the app, but Apple rejected it. Apple did so even though the European Union had just hit it with a nearly $2 billion fine for "blocking" alternative music apps. The EU is also investigating Apple, Meta and Google for self-preferencing and charging developers additional fees. As for how Apple will react to Spotify's latest test, we'll just have to wait and see.

As learning language models (LLMs) continue to advance, so do questions about how they can benefit society in areas such as the medical field. A recent study from the University of Cambridge's School of Clinical Medicine found that OpenAI's GPT-4 performed nearly as well in an ophthalmology assessment as experts in the field, the Financial Times first reported.

In the study, published in PLOS Digital Health, researchers tested the LLM, its predecessor GPT-3.5, Google's PaLM 2 and Meta's LLaMA with 87 multiple choice questions. Five expert ophthalmologists, three trainee ophthalmologists and two unspecialized junior doctors received the same mock exam. The questions came from a textbook for trialing trainees on everything from light sensitivity to lesions. The contents aren't publicly available, so the researchers believe LLMs couldn't have been trained on them previously. ChatGPT, equipped with GPT-4 or GPT-3.5, was given three chances to answer definitively or its response was marked as null. 

GPT-4 scored higher than the trainees and junior doctors, getting 60 of the 87 questions right. While this was significantly higher than the junior doctors' average of 37 correct answers, it just beat out the three trainees' average of 59.7. While one expert ophthalmologist only answered 56 questions accurately, the five had an average score of 66.4 right answers, beating the machine. PaLM 2 scored a 49, and GPT-3.5 scored a 42. LLaMa scored the lowest at 28, falling below the junior doctors. Notably, these trials occurred in mid-2023. 

While these results have potential benefits, there are also quite a few risks and concerns. Researchers noted that the study offered a limited number of questions, especially in certain categories, meaning the actual results might be varied. LLMs also have a tendency to "hallucinate" or make things up. That's one thing if its an irrelevant fact but claiming there's a cataract or cancer is another story. As is the case in many instances of LLM use, the systems also lack nuance, creating further opportunities for inaccuracy.

The European Union doesn't think you should have to choose between giving Meta and other major players your data or your money. In a statement, the European Data Protection Board (EDPB) stated that "consent or pay" models often don't "comply with the requirements for valid consent" when a person must choose between providing their data for behavioral advertising purposes or pay for privacy.

The EDPB argues that only offering a paid alternative to data collection shouldn't be the default for large online platforms. It doesn't issue a mandate but stresses that these platforms should "give significant consideration" to providing a free option that doesn't involve data processing (or at least not as much). "Controllers should take care at all times to avoid transforming the fundamental right to data protection into a feature that individuals have to pay to enjoy," EDPB Chair Anu Talus said. "Individuals should be made fully aware of the value and the consequences of their choices."

Currently, EU users must pay €10 ($11) monthly for an ad-free subscription or be forced to share their data. The EU is already investigating if this system complies with the Digital Markets Act, which went into effect at the beginning of March.

TVs might have gotten much less bulky in recent years, but most are still just big black screens. If you've been holding out for something a bit prettier, check out Amazon's sale on Samsung's QLED 4K The Frame LS03B Series. Most of the sizes are down to record lows, but the 75-inch model has the best deal thanks to a 37 percent discount. It's still going to cost you a pretty penny, but the framed TV is down to $1,895 from $2,998 — a $1,103 savings. 

We've been big fans of Samsung's frame TVs for some time and love not having to stare at a blank screen. This model has features such as an anti-reflection and matte display, so it doesn't give itself away as a TV hiding along your actual artwork. You can display one of over 2,000 beautiful works from across the world, ranging from The Starry Night to The Kiss by Gustav Klimt. However, on top of the money you shelled out on the TV, there's also a $6 monthly subscription to the Samsung Art Store. You can opt to add your own photos instead (just make sure they're prepped for such a large format and don't get all pixelated). 

The Samsung Frame TV does include a slim wall mount and lets you customize the bezel's type and color. Quality-wise, it's a QLED 4K TV with Quantum HDR. If the size — or price — feels like too much, then check out other sizes, such as the 55-inch Samsung Frame TV. This model is down to $998 from $1,498 — a 33 percent discount and an all-time low price. 

Follow @EngadgetDeals on Twitter and subscribe to the Engadget Deals newsletter for the latest tech deals and buying advice.

It's becoming increasingly difficult to determine what is AI-created as tools' capabilities continue to improve. Marking AI-created images is one solution and an option Snapchat is expanding on. The platform is introducing a watermark on AI-generated content that is saved to your camera roll or exported.

Snapchat

In a surprise to absolutely no one, Snapchat's AI watermark will be a ghost with sparkles next to it. There doesn't seem to be any way to control where the icon appears (as is typical with watermarks), but we'll know more when Snapchat adds the feature "soon."

The addition of an AI marker for externally seen images follows Snapchat's steps to label in-app content. There's the extend tool, which creates the effect of a zoomed out image, and has a sparkle icon to indicate it's an AI feature. Then there's the Dreams feature, which uses generative AI to turn selfies into "fantastical images that transform their persona into new identities." Anyone who receives a Dream image also gets a context card explaining its AI use.