When talking about risk during a press conference on Thursday, the NASA officials in charge of the upcoming Artemis II Moon mission hedged their answers.

Reporters' questions on the risks were certainly valid and appropriate. In an open society, it is vital to set expectations for any hazardous venture such as spaceflight—most importantly for the astronauts actually making the journey, but also for NASA's workforce, the White House, lawmakers, and members of the public paying for the endeavor.

What's more, Artemis II will be the first mission since 1972 to fly humans to the vicinity of the Moon. This is not following the well-trodden yet perilous path that astronauts take to reach the International Space Station, just a few hundred miles above Earth.

Read full article

Comments

A 58-year-old woman in Greece appears to hold the record for growing a parasitic sheep bot fly in her nose the longest, almost creating a snot rocket that could literally fly.

Usually, when the sheep bot fly accidentally nosedives into a human's schnoz, the first-stage larvae they deliver don’t actually develop. In contrast, in its normal target—a sheep's nose— the larvae would move up into the sinuses, feed, grow, and molt into second- and third-stage larvae. From there, the flies ( Oestrus ovis ) drip from the nose onto the ground, burrow into the soil, pupate, and emerge as adult flies.

For a long time, experts thought that the flies couldn't complete their development in humans beyond the first larval stage. But a few human cases have been reported in recent decades involving the second- and third-stage larvae. The woman's case, reported in the Journal of Emerging Infectious Diseases by a medical entomologist and colleagues, goes the furthest yet, finding pupa and a puparium—the hard casing of a pupa—in the woman's nose.

Read full article

Comments

Do you remember the joyful satisfaction you felt when you really started to understand Slay the Spire ?

This isn’t a totally rhetorical question. If you’re reading this piece about Slay the Spire 2 —published roughly a week into what promises to be a lengthy Early Access period —I have to assume you’ve put in dozens, if not hundreds (or thousands?) of hours with the original Slay the Spire . At this point, the game probably feels less like a game and more like a comfortable old pair of sneakers. You probably have a favorite character, a preferred set of card synergies to focus on building for that character, and a set of alternative strategies to aim for when the vagaries of chance make that preferred strategy impossible. The game’s plentiful randomization makes each run feel a bit different, but the contours of those runs start to feel a little common to anyone who has tinkered with the game for years.

But think back, if you can, to when Slay the Spire was an exciting new challenge. Remember those first few runs, when you were still deep in the trial-and-error phase of your Slay the Spire journey. You still had to read each new card carefully as it appeared, developing potential strategies on the fly and weighing key deckbuilding and power-building decisions for minutes at a time to maximize your chance of survival. Sure, you failed a lot. But you got a little more confident each time, and a little farther every few sessions, and just a little more knowledgeable about and immersed in the game’s intricate, well-balanced systems.

Read full article

Comments

With its Alpha series of game-playing AIs, Google's DeepMind group seemed to have found a way for its AIs to tackle any game, mastering games like chess and Go by repeatedly playing itself during training. But then some odd things happened as people started identifying Go positions that would lose against relative newcomers to the game but easily defeat a similar Go -playing AI .

While beating an AI at a board game may seem relatively trivial, it can help us identify failure modes of the AI, or ways in which we can improve their training to avoid having them develop these blind spots in the first place—things that may become critical as people rely on AI input for a growing range of problems.

A recent paper published in Machine Learning describes an entire category of games where the method used to train AlphaGo and AlphaChess fails. The games in question can be remarkably simple, as exemplified by the one the researchers worked with: Nim , which involves two players taking turns removing matchsticks from a pyramid-shaped board until one is left without a legal move.

Read full article

Comments

Google Fiber, now officially called GFiber, is being sold to private equity firm Stonepeak and will be combined with cable-and-fiber firm Astound Broadband to create a larger Internet service provider.

Google owner Alphabet announced Wednesday that it will keep only a minority stake in the fiber ISP that launched with grand ambitions in 2012 but scaled back its expansion plans in 2016 . Alphabet and Astound owner Stonepeak announced "an agreement to combine GFiber with Astound Broadband, creating a leading independent fiber provider," with the merged company to be "majority owned by Stonepeak, an investment firm specializing in infrastructure and real assets."

The deal is subject to regulatory approvals and other closing conditions, with an expected closing date in Q4 of this year. The sale price was not disclosed. The deal will help GFiber take "a major step toward its goal of operational and financial independence" and obtain the "external capital and strategic focus needed to accelerate its next phase of growth," the announcement said.

Read full article

Comments

Researchers say they’ve discovered a supply-chain attack flooding repositories with malicious packages that contain invisible code, a technique that’s flummoxing traditional defenses designed to detect such threats.

The researchers, from firm Aikido Security, said Friday that they found 151 malicious packages that were uploaded to GitHub from March 3 to March 9. Such supply-chain attacks have been common for nearly a decade . They usually work by uploading malicious packages with code and names that closely resemble those of widely used code libraries, with the objective of tricking developers into mistakenly incorporating the former into their software. In some cases, these malicious packages are downloaded thousands of times.

Defenses see nothing. Decoders see executable code

The packages Aikido found this month have adopted a newer technique: selective use of code that isn’t visible when loaded into virtually all editors, terminals, and code review interfaces. While most of the code appears in normal, readable form, malicious functions and payloads—the usual telltale signs of malice—are rendered in unicode characters that are invisible to the human eye. The tactic, which Aikido said it first spotted last year, makes manual code reviews and other traditional defenses nearly useless. Other repositories hit in these attacks include NPM and Open VSX.

Read full article

Comments

Canceling a software subscription is supposed to be easy—that's what US law dictates. Adobe, however, has played fast and loose with its Creative Cloud subscriptions in the past. The company was sued by the Department of Justice in 2024 due to its practice of hiding hefty termination fees when customers signed up. The case has now been settled, with Adobe agreeing to a $75 million fine and matching free services to users of its products.

Turning software into a monthly subscription is all the rage these days, but Adobe was way ahead of the curve. The company began offering its suite of editing tools, like Photoshop and Illustrator, as a monthly subscription back in 2013, and most of its customers migrated to the new system.

It was easy for Adobe to get away with that shift because CS6, the last perpetual license offered for its editing tools, started at $700 and went up to more than $2,600 for all apps. By contrast, paying between $10 and $70 per month seems like a good deal, and it might be in the short term. Although anyone who has been paying monthly since the change has spent thousands of dollars on Adobe software. And when people noticed that and decided they wanted to cancel, many of them were frustrated with the outcome.

Read full article

Comments

For more than a decade, most electric vehicles have shared the same electrical backbone: a battery pack operating at roughly 400 V. It’s the invisible standard behind everything from early compliance cars to today’s bestselling EVs. But over the past few years, a growing number of automakers have doubled that number, moving to 800 V architectures and promising dramatically faster charging, better performance, and improved efficiency.

Cars like the Porsche Taycan and Hyundai Ioniq 5 helped push 800 V into the mainstream conversation, touting 18-minute charging sessions and sustained high-speed performance. On paper, doubling the voltage sounds like a simple upgrade. In reality, it reshapes everything from cable thickness and thermal management to semiconductor choice and charging infrastructure compatibility.

The physics: Why higher voltage matters

Understanding why higher voltage matters is as important as the hardware that carries it.

Read full article

Comments

If you're an AT&T FirstNet customer and suddenly get hit with a $6,200 charge, the good news is that it's probably a mistake and can be corrected. But actually getting the wrong charge wiped out might not be so easy.

This has now happened at least twice. In December 2024, a Texas police officer received a $6,223 bill with a $6,194 charge for using 3.1GB of data. He said he had unlimited data but was charged incorrectly after moving a line to AT&T's FirstNet service for first responders. He called AT&T and went to an AT&T store but only got the bill reversed after contacting the AT&T president’s office.

An AT&T spokesperson told Ars at the time that it was "investigating to determine what caused this system error." But AT&T never revealed exactly what caused it, and now another FirstNet user has gone through an almost identical ordeal.

Read full article

Comments