Amazon unveils three satellite user terminals, plans broadband service in 2024

Amazon Kuiper terminals come in three sizes, with max speeds from 100Mbps to 1Gbps.

Three satellite broadband user terminals designed by Amazon.

Enlarge / Satellite broadband user terminals designed by Amazon's Project Kuiper. (credit: Amazon)

Amazon has designed three satellite broadband user terminals and will start offering Internet service in 2024, the company announced today. The standard terminal, designed for residential and small business customers, is expected to cost Amazon less than $400 to make; Amazon did not say what it will charge customers for the terminals or for monthly service plans.

The "standard customer terminal measures less than 11 inches square and 1 inch thick," Amazon said. "It weighs less than five pounds without its mounting bracket. Despite this modest footprint, the device will be one of the most powerful commercially available customer terminals of its size, delivering speeds up to 400 megabits per second (Mbps). Amazon expects to produce these terminals for less than $400 each."

Whether customers actually get those speeds in practice will depend on the satellites and how congested the network is in each region, as the experience of SpaceX Starlink customers shows. Amazon didn't differentiate between download and upload speeds in its announcement, but upload speeds are likely to be slower than downloads.

Read 11 remaining paragraphs | Comments

OpenAI’s GPT-4 exhibits “human-level performance” on professional benchmarks

Multimodal AI model can process images and text; pass bar exams.

A colorful AI-generated image of a radiating silhouette.

Enlarge (credit: Ars Technica)

On Tuesday, OpenAI announced GPT-4, a large multimodal model that can accept text and image inputs while returning text output that "exhibits human-level performance on various professional and academic benchmarks," according to OpenAI. Also on Tuesday, Microsoft announced that Bing Chat has been running on GPT-4 all along.

If it performs as claimed, GPT-4 potentially represents the opening of a new era in artificial intelligence. "It passes a simulated bar exam with a score around the top 10% of test takers," writes OpenAI in its announcement. "In contrast, GPT-3.5’s score was around the bottom 10%."

OpenAI plans to release GPT-4's text capability through ChatGPT and its commercial API, but with a waitlist at first. GPT-4 is currently available to subscribers of ChatGPT Plus. Also, the firm is testing GPT-4's image input capability with a single partner, Be My Eyes, an upcoming smartphone app that can recognize a scene and describe it.

Read 10 remaining paragraphs | Comments

Do better coders swear more, or does C just do that to good programmers?

For open source C code, curses mean quality, a recent bachelor’s thesis suggests.

A person screaming at his computer.

Enlarge (credit: dasilvafa)

Ever find yourself staring at a tricky coding problem and thinking, “shit”?

If those thoughts make their way into your code or the associated comments, you’re in good company. When undergraduate student Jan Strehmel from Karlsruhe Institute of Technology analyzed open source code written in the programming language C, he found no shortage of obscenity. While that might be expected, Strehmel’s overall finding might not be: The average quality of code containing swears was significantly higher than the average quality of code that did not.

“The results are quite surprising!” Strehmel said. Programmers and scientists may have a lot of follow-up questions. Are the researchers sure there aren’t certain profanity-prone programmers skewing the results? What about other programming languages? And, most importantly, why would swears correlate with high-quality code? The work is ongoing, but even without all the answers, one thing’s for sure: Strehmel just wrote one hell of a bachelor’s thesis.

Read 10 remaining paragraphs | Comments

Did Oregon once host a nesting colony of pterosaurs?

A fossil site may contain guano washed into the oceans from a pterosaur colony.

The Oregon pterosaur <em>Bennettazhia oregonensis</em>, with 4 meter wingspan, reconstructed independently in seagull colors by Midiaou Diallo and reproduced with permission.

Enlarge / The Oregon pterosaur Bennettazhia oregonensis, with 4 meter wingspan, reconstructed independently in seagull colors by Midiaou Diallo and reproduced with permission. (credit: Midiaou Diallo)

A single fossil toe is all we have of the ‘Mitchell ornithopod,’ the nickname of the first early Cretaceous dinosaur fossil found in Oregon in 2018. Ornithopods were enormous herbivores such as duck-billed dinosaurs and iguanodons, and Gregory J. Retallack, lead author of that discovery, wanted to find more of its skeleton. Three years later, he returned to the site, aided by over 80 volunteers who helped excavate in more detail.

No further ornithopod bones—indeed, no substantial dinosaur bones of any kind—were retrieved after two weeks of digging. It was, he said, “a failure” in that regard. What they found instead was a complete mess, a jumble of the remains of land-based and aquatic animals. And lots and lots of guano.

That came from extinct flying reptiles, known collectively as pterosaurs, and suggests these animals may have flocked together on the cliffs above the coast of Oregon.

Read 28 remaining paragraphs | Comments

Google shows off what ChatGPT would be like in Gmail and Google Docs

Google will bring generative AI to Workspace, but a public launch sounds far off.

The "put generative AI in everything" era is kicking off at Google. In addition to the ChatGPT-style features that are eventually coming to Google Search, today Google announced a round of generative AI features for Google Docs and Gmail. Basically, Google plans to eventually, someday, release a text bot that will do all the writing for you. It can reply to emails and make presentations with just a text prompt. As usual for Google and AI, this is not out yet, and the company says it is only "sharing our broader vision" with this blog post.

Just like the rise of Facebook and Google's hyper-aggressive response with Google+, Google is in a total panic over the rise of ChatGPT and AI-powered text. Just like how Google put social features into every product back in the G+ days, the plan going forward is to build ChatGPT-style generative text into every Google product. Google's blog post backs this up by framing this announcement as part of a larger plan, saying, "To start, we’re introducing a first set of AI-powered writing features in Docs and Gmail." So far, the company has also promised to put AI into its health care offerings and opened up API access to a language model, but we've yet to see a real consumer product launch.

The company says it's building a "collaborative AI partner" into Workspace and has a breakdown of the future features:

Read 10 remaining paragraphs | Comments

Ransomware attacks have entered a heinous new phase

Cybercriminal gangs now releasing stolen photos of cancer patients, student records.

row of lockers

Enlarge (credit: Don Farrall/Getty Images)

In February, attackers from the Russia-based BlackCat ransomware group hit a physician practice in Lackawanna County, Pennsylvania, that's part of the Lehigh Valley Health Network (LVHN). At the time, LVHN said that the attack “involved” a patient photo system related to radiation oncology treatment. The health care group said that BlackCat had issued a ransom demand, “but LVHN refused to pay this criminal enterprise.”

After a couple of weeks, BlackCat threatened to publish data stolen from the system. “Our blog is followed by a lot of world media, the case will be widely publicized and will cause significant damage to your business,” BlackCat wrote on their dark-web extortion site. “Your time is running out. We are ready to unleash our full power on you!” The attackers then released three screenshots of cancer patients receiving radiation treatment and seven documents that included patient information.

Read 10 remaining paragraphs | Comments

Banana Pi BPI-R3 Mini is a tiny router board with two 2.5 GbE Ethernet ports and an MT7986 chip

The latest member of the Banana Pi line of single-board computers is a tiny model designed for networking applications. The Banana Pi BPI-R3 Mini has two 2.5 GbE Ethernet ports, two M.2 connectors, and support for WiFi 6E, but there’s no video o…

The latest member of the Banana Pi line of single-board computers is a tiny model designed for networking applications. The Banana Pi BPI-R3 Mini has two 2.5 GbE Ethernet ports, two M.2 connectors, and support for WiFi 6E, but there’s no video output. So you could use the board as a router, wireless repeater, gateway, […]

The post Banana Pi BPI-R3 Mini is a tiny router board with two 2.5 GbE Ethernet ports and an MT7986 chip appeared first on Liliputing.

System Shock gets May 30 PC release date after successful demo

It’s been a long road since the project’s 2016 Kickstarter funding.

Floor full of dismembered bodies.

Enlarge / Nearly 30 years later, a room full of dismembered bodies looks a lot more vivid.

We'll have to wait just a bit longer for Nightdive Studios' long-awaited remake of System Shock. The PC version of the game, which was expected to launch this month, is now targeting a May 30 release date on Steam, GOG, and the Epic Games Store, according to an announcement that came in the voice of series antagonist AI Shodan. Nightdive added that the PC release will be followed by versions for PlayStation and Xbox consoles "in due course."

"We had hoped to bring the game to market by the end of March, but that turned out to be just beyond our reach," the developers wrote. "We are after all merely human (unlike Shodan!)."

The long road to this System Shock remake started back in 2016, when Nightdive raised $1.35 million via Kickstarter for what it called "a complete remake of the genre-defining classic from 1994, rebuilt from the ground up with the Unity Engine." Active development later moved to the Unreal Engine but was put on hiatus in early 2018, shortly after the team missed its initial December 2017 target.

Read 3 remaining paragraphs | Comments

Meta to lay off another 10,000 as Zuckerberg celebrates “year of efficiency”

After over-hiring, Zuckerberg says staff cuts taught him that “leaner is better.”

Mark Zuckerberg walks away from a courthouse while wearing a suit; he carries a phone in his left hand and a mask in his right hand.

Enlarge / Meta CEO Mark Zuckerberg (right) leaves federal court in San Jose, California, on Dec. 20, 2022, after testifying in an antitrust case brought by the Federal Trade Commission. (credit: Getty Images | Bloomberg)

Meta plans to lay off another 10,000 employees and will stop trying to fill 5,000 open roles, CEO Mark Zuckerberg told staff in a memo today.

Zuckerberg titled the memo "Update on Meta's Year of Efficiency" and used the first two paragraphs to tout improvements to the Facebook, Instagram, and WhatsApp owner's operations. In the third paragraph, employees reading the memo found out that 10,000 of them will lose their jobs in the coming months.

"Overall, we expect to reduce our team size by around 10,000 people and to close around 5,000 additional open roles that we haven't yet hired," Zuckerberg wrote.

Read 17 remaining paragraphs | Comments