Victoria Fox's Blog, page 114

"Star Wars Battlefront Classic Collection": Everything about this game is like 2004. And I love it

Once again, two old Star Wars games have been re-released together in a so-called Classic Collection, and I could now say: It costs 35 euros and plays clunky. I could say online matches, even cutscenes in the offline campaign, are full of bugs. Servers crash. The computer opponents couldn’t be more stupid. But since the release, I have played Star Wars Battlefront I and II (PC, Switch, Playstation 4/5, Xbox One/Series) almost continuously for over a week. And I love her very much.

SOURCE

NYC’s government chatbot is lying about city laws and regulations

Enlarge / Has a government employee checked all those zeroes and ones floating above the skyline?Getty Images

If you follow generative AI news at all, you’re probably familiar with LLM chatbots’ tendency to “confabulate” incorrect information while presenting that information as authoritatively true. That tendency seems poised to cause some serious problems now that a chatbot run by the New York City government is making up incorrect answers to some important questions of local law and municipal policy.

NYC’s “MyCity” ChatBot launched as a “pilot” program last October. The announcement touted the ChatBot as a way for business owners to “save … time and money by instantly providing them with actionable and trusted information from more than 2,000 NYC Business webpages and articles on topics such as compliance with codes and regulations, available business incentives, and best practices to avoid violations and fines.”

But a new report from The Markup and local nonprofit news site The City found the MyCity chatbot giving dangerously wrong information about some pretty basic city policies. To cite just one example, the bot said that NYC buildings “are not required to accept Section 8 vouchers,” when an NYC government info page says clearly that Section 8 housing subsidies are one of many lawful sources of income that landlords are required to accept without discrimination. The Markup also received incorrect information in response to chatbot queries regarding worker pay and work hour regulations, as well as industry-specific information like funeral home pricing.

Enlarge / Welcome news for people who think the rent is too damn high, courtesy of the MyCity chatbot.Kathryn Tewson / BlueSky

Further testing from BlueSky user Kathryn Tewson shows the MyCity chatbot giving some dangerously wrong answers regarding the treatment of workplace whistleblowers, as well as some hilariously bad answers regarding the need to pay rent.

This is going to keep happening

The result isn’t too surprising if you dig into the token-based predictive models that power these kinds of chatbots. MyCity’s Microsoft Azure-powered chatbot uses a complex process of statistical associations across millions of tokens to essentially guess at the most likely next word in any given sequence, without any real understanding of the underlying information being conveyed.

That can cause problems when a single factual answer to a question might not be reflected precisely in the training data. In fact, The Markup said that at least one of its tests resulted in the correct answer on the same query about accepting Section 8 housing vouchers (even as “ten separate Markup staffers” got the incorrect answer when repeating the same question).

The MyCity Chatbot—which is prominently labeled as a “Beta” product—tells users who bother to read the warnings that it “may occasionally produce incorrect, harmful or biased content” and that users should “not rely on its responses as a substitute for professional advice.” But the page also states front and center that it is “trained to provide you official NYC Business information” and is being sold as a way “to help business owners navigate government.”

Andrew Rigie, executive director of the NYC Hospitality Alliance, told The Markup that he had encountered inaccuracies from the bot himself and had received reports of the same from at least one local business owner. But NYC Office of Technology and Innovation Spokesperson Leslie Brown told The Markup that the bot “has already provided thousands of people with timely, accurate answers” and that “we will continue to focus on upgrading this tool so that we can better support small businesses across the city.”

NYC Mayor Eric Adams touts the MyCity chatbot in an October announcement event.

The Markup’s report highlights the danger of governments and corporations rolling out chatbots to the public before their accuracy and reliability have been fully vetted. Last month, a court forced Air Canada to honor a fraudulent refund policy invented by a chatbot available on its website. A recent Washington Post report found that chatbots integrated into major tax preparation software provides “random, misleading, or inaccurate … answers” to many tax queries. And some crafty prompt engineers have reportedly been able to trick car dealership chatbots into accepting a “legally binding offer – no take backsies” for a $1 car.

These kinds of issues are already leading some companies away from more generalized LLM-powered chatbots and toward more specifically trained Retrieval-Augmented Generation models, which have been tuned only on a small set of relevant information. That kind of focus could become that much more important if the FTC is successful in its efforts to make chatbots liable for “false, misleading, or disparaging” information.

SOURCE

Playboy image from 1972 gets ban from IEEE computer journals

EnlargeAurich Lawson | Getty Image

On Wednesday, the IEEE Computer Society announced to members that, after April 1, it would no longer accept papers that include a frequently used image of a 1972 Playboy model named Lena Forsén. The so-called “Lenna image,” (Forsén added an extra “n” to her name in her Playboy appearance to aid pronunciation) has been used in image processing research since 1973 and has attracted criticism for making some women feel unwelcome in the field.

In an email from the IEEE Computer Society sent to members on Wednesday, Technical & Conference Activities Vice President Terry Benzel wrote, “IEEE’s diversity statement and supporting policies such as the IEEE Code of Ethics speak to IEEE’s commitment to promoting an including and equitable culture that welcomes all. In alignment with this culture and with respect to the wishes of the subject of the image, Lena Forsén, IEEE will no longer accept submitted papers which include the ‘Lena image.'”

An uncropped version of the 512×512-pixel test image originally appeared as the centerfold picture for the December 1972 issue of Playboy Magazine. Usage of the Lenna image in image processing began in June or July 1973 when an assistant professor named Alexander Sawchuck and a graduate student at the University of Southern California Signal and Image Processing Institute scanned a square portion of the centerfold image with a primitive drum scanner, omitting nudity present in the original image. They scanned it for a colleague’s conference paper, and after that, others began to use the image as well.

The original 512×512 “Lenna” test image, which is a cropped portion of a 1972 Playboy centerfold.Playboy / Wikipedia

The image’s use spread in other papers throughout the 1970s, 80s, and 90s, and it caught Playboy’s attention, but the company decided to overlook the copyright violations. In 1997, Playboy helped track down Forsén, who appeared at the 50th Annual Conference of the Society for Imaging Science in Technology, signing autographs for fans. “They must be so tired of me … looking at the same picture for all these years!” she said at the time. VP of new media at Playboy Eileen Kent told Wired, “We decided we should exploit this, because it is a phenomenon.”

The image, which features Forsén’s face and bare shoulder as she wears a hat with a purple feather, was reportedly ideal for testing image processing systems in the early years of digital image technology due to its high contrast and varied detail. It is also a sexually suggestive photo of an attractive woman, and its use by men in the computer field has garnered criticism over the decades, especially from female scientists and engineers who felt that the image (especially related to its association with the Playboy brand) objectified women and created an academic climate where they did not feel entirely welcome.

Due to some of this criticism, which dates back to at least 1996, the journal Nature banned the use of the Lena image in paper submissions in 2018.

The comp.compression Usenet newsgroup FAQ document claims that in 1988, a Swedish publication asked Forsén if she minded her image being used in computer science, and she was reportedly pleasantly amused. In a 2019 Wired article, Linda Kinstler wrote that Forsén did not harbor resentment about the image, but she regretted that she wasn’t paid better for it originally. “I’m really proud of that picture,” she told Kinstler at the time.

Since then, Forsén has apparently changed her mind. In 2019, Creatable and Code Like a Girl created an advertising documentary titled Losing Lena, which was part of a promotional campaign aimed at removing the Lena image from use in tech and the image processing field. In a press release for the campaign and film, Forsén is quoted as saying, “I retired from modelling a long time ago. It’s time I retired from tech, too. We can make a simple change today that creates a lasting change for tomorrow. Let’s commit to losing me.”

It seems like that commitment is now being granted. The ban in IEEE publications, which have been historically important journals for computer imaging development, will likely further set a precedent toward removing the Lenna image from common use. In his email, the IEEE’s Benzel recommended wider sensitivity about the issue, writing, “In order to raise awareness of and increase author compliance with this new policy, program committee members and reviewers should look for inclusion of this image, and if present, should ask authors to replace the Lena image with an alternative.”

SOURCE

AT&T investigates the publication of millions of customers' data on the 'dark web'

The company AT&T reported today that it is investigating an incident in which information of millions of customers, such as their address and social security number, were published two weeks ago on the “deep web” (Darkweb), which is accessed in most cases through specific applications and browsers.

The company indicated in a press release that although it took that action, “it is not yet known whether the data in those fields originated from AT&T or one of its providers” and also includes former customers.

The investigation is being conducted by internal and external cybersecurity experts to determine how the data of approximately 7.6 million customers and approximately 65.4 million former account holders ended up there.

The preliminary analysis carried out by the company indicates that these data are from 2019 or earlier , adds the statement echoed by the Efe agency.

“AT&T has no evidence of unauthorized access to its systems that resulted in data breaches,” the company said.

The company also indicated that it is communicating with those affected and urged them to visit www.att.com/accountsafety for more information.

Last February, AT&T apologized to thousands of users affected by a service outage caused by technical problems that affected thousands of people in Atlanta, Los Angeles and New York, and that occurred due to a technical problem while the company was trying to expand its network. .

SOURCE

Artificial Intelligence: OpenAI introduces voice cloning technology

The US company OpenAI has presented a new program for cloning real voices. OpenAI presents several voice recordings on its website that were created using the Voice Engine software. Accordingly, after uploading a 15-second audio of a human voice, Voice Engine was able to imitate it, read a text and also translate it. This would allow users to, among other things, create an audio recording of their voice in Chinese or French.

In September, the company, which also developed the chatbot ChatGPT, introduced HeyGen, a tool that can translate videos using artificial intelligence and adjust lip movements. According to OpenAI, HeyGen uses Voice Engine technology .

In view of the risk of misuse, Voice Engine will not initially be freely accessible, according to the company. OpenAI wants to test the product with testers.

“Serious risks” in the election year

“We understand that generating speech that resembles people’s voices has serious risks that are particularly in focus in an election year,” the company said. A new president will be elected in the USA on November 5th. Important elections are also taking place in numerous other countries this year.

In the US state of New Hampshire, authorities are currently investigating calls in which President Joe Biden’s voice was imitated using AI to deter people from voting in the local primary election in January.

The US company OpenAI has presented a new program for cloning real voices. OpenAI presents several voice recordings on its website that were created using the Voice Engine software. Accordingly, after uploading a 15-second audio of a human voice, Voice Engine was able to imitate it, read a text and also translate it. This would allow users to, among other things, create an audio recording of their voice in Chinese or French.

In September, the company, which also developed the chatbot ChatGPT, introduced HeyGen, a tool that can translate videos using artificial intelligence and adjust lip movements. According to OpenAI, HeyGen uses Voice Engine technology .

SOURCE

New Federal Forest Law: Foresters against bicycles

397 meters above sea level, a small earth wall has become a stumbling block. At first glance it really doesn’t look like anything more, just a bit of piled up old earth covered by leaves. About 2,000 years ago, the Limes, the border line with which the Romans defined their empire, ran here, and it also ran through this piece of forest up here on the Gaulskopf in the eastern Hintertaunus. That’s why the small earth wall is a UNESCO World Heritage Site. And that’s why he needs a protector.

Birgit Wetzel is a forester for the Maibach district and watches over 3,000 hectares of forest and therefore also the Limes. And she is worried. Because apparently mountain bikers ride around on the Limes, even though it’s forbidden, and thus wear away its top layer. A monument that is thousands of years old could be gradually destroyed.

SOURCE

The entire state of Illinois is going to be crawling with cicadas

EnlargeEd Reschke via Getty

Brace yourselves, Illinoisans: A truly shocking number of cicadas are about to live, make sweet love, and die in a tree near you. Two broods of periodical cicadas—Brood XIX on a 13-year cycle and Brood XIII on a 17-year cycle—are slated to emerge together in central Illinois this summer for the first time in over two centuries. To most humans, they’re an ephemeral spectacle and an ear-splitting nuisance, and then they’re gone. To many other Midwestern animals, plants, and microbes, they’re a rare feast, bringing new life to forests long past their death.

From Nebraska to New York, 15 broods of periodical cicadas grow underground, quietly sipping watery sap from tree roots. After 13 or 17 years (depending on the brood), countless inch-long adults dig themselves out in sync, crawling out of the ground en masse for a monthlong summer orgy. After mating, they lay eggs in forest trees and die, leaving their tree-born babies to fall to the forest floor and begin the cycle anew. Cicadas don’t fly far from their birthplace, so each brood occupies a distinct patch of the US. “They form a mosaic on the landscape,” says Chris Simon, senior research scientist in ecology and evolutionary biology at the University of Connecticut.

Most years, at least one of these 15 broods emerges (annual cicadas, not to be confused with their smaller periodical cousins, pop up separately every summer). Sometimes two broods emerge at the same time. It’s also not unheard of for multiple broods to coexist in the same place. “What’s unusual is that these two broods are adjacent,” says John Lill, insect ecologist at George Washington University. “Illinois is going to be ground zero. From the very top to the very bottom of the state, it’s going to be covered in cicadas.” The last time that these broods swarmed aboveground together, Thomas Jefferson was president and the city of Chicago had yet to exist.

Entomologists around the world already have their flights booked for May. “We’re like cicada groupies,” Lill says. He promises that this once-in-a-generation spectacle will be even better than April’s total solar eclipse. During 2004’s Brood X emergence, Lill remembers walking outside at midnight. “For two seconds, I was like, ‘Wow, I didn’t know it was raining,’ because I saw water flowing down the street. As my eyes focused, I realized it was literally just thousands of cicadas crawling across the street.”

Some cicada devotees, like author and entomologist Greg Kritsky, have already witnessed Brood XIII emerge a couple of times. But for most of their predators, a brood emergence happens once in a lifetime, and it’s always an extremely pleasant surprise. “It’s a food bonanza,” Kritsky says, “like if you walked outside and found the whole world swarming with flying Hershey’s Kisses.”

Cicadas are shockingly chill, protein-packed, and taste like high-end shrimp—easy, delicious prey. “Periodical cicadas are sitting ducks,” says Lill. They don’t bite, sting, or poison anyone, and they’re totally unbothered by being handled. Dogs, raccoons, birds, and other generalist predators will gorge themselves on this flying feast until they’re stuffed, and it barely makes a dent in the cicada population. It’s their secret weapon, Lill says: In the absence of other defense mechanisms, “they just overwhelm predators by their sheer abundance.”

Much like an unexpected free dinner will distract you from the leftovers sitting in your fridge, this summer’s cicada emergence will turn predators away from their usual prey. During the 2021 Brood X emergence, Zoe Getman-Pickering, a scientist in Lill’s research group, found that as birds swooped in on cicadas, caterpillar populations exploded. Spared from birds, caterpillars chomped on twice as many oak leaves as normal—and the chain of effects went on and on. Scientists can’t possibly study them all. “The ecosystem gets a swift kick, with this unexpected perturbation that changes a lot of things at once,” says Louie Yang, an ecologist and professor of entomology at UC Davis.

From birth to death, these insects shape the forest around them. As temperatures rise in late April, pale, red-eyed cicada nymphs begin clawing pinky-sized holes in the ground, preparing for their grand May entrance. All of these tunnels make it easier for rainwater to move through the soil, where it can then be used by plants and other dirt-inhabiting microbes. Once fully grown and aboveground, adult cicadas shed their exoskeletons, unfurl their wings, and fly off to spend their remaining four to six weeks on Earth singing (if they’re male), listening for the sexiest songs (if they’re female), and mating.

Mother cicadas use the metal-enhanced saws built into their abdomens—wood-drilling shafts layered with elements like aluminum, copper, and iron—to slice pockets into tree branches, where they’ll lay roughly 500 eggs each. Sometimes, all of these cuts cause twigs to wither or snap, killing leaves. While this could permanently damage a very young sapling, mature trees simply shed the slashed branches and carry on. “It’s like natural pruning,” Kritsky says, which keeps hearty trees strong, prevents disease, and promotes flower growth.

Once mating season winds down, so does the cicada’s life. “In late summer, everybody forgets about cicadas,” Lill says. “They all die. They all rot in the ground. And then they’re gone.” By late June, there will be millions of pounds of cicadas piling up at the base of trees, decomposing. The smell, Kritsky says, “is a sentient memory you will never forget—like rancid Limburger cheese.”

But these stinky carcasses send a massive pulse of food to scavengers in the soil. “The cicadas serve as reservoirs of nutrients,” Yang says. “When they come out, they release all this stored energy into the ecosystem,” giving their bodies back to the plants that raised them. In the short term, dead cicadas have a fertilizing effect, feeding microbes in the soil and helping plants grow larger. And as their remnants make their way into woodland ponds and streams, cicada nutrients are carried downstream, where they may strengthen aquatic ecosystems far beyond their home tree.

They may smell like bad hamburgers, but Yang says that if you’re lucky enough to host a tree full of cicadas this year, it’s best to just leave their bodies alone to decompose naturally. “They’ll be gone soon enough,” he says. If the pileup is especially obtrusive, simply sweep them out of the way and let nature do the rest.

The thought of billions of screeching insects in your backyard might make your skin crawl, but you don’t need to be a passive observer when they arrive. Researchers are clamoring for citizen scientists to send in photos of their local cicadas to help map the upcoming emergence. The Cicada Safari app, developed by Kritsky, received and verified 561,000 cicada pics during the 2021 Brood X emergence—he hopes to get even more this time around.

“This is an amazing natural phenomenon to wonder about,” Lill says, “not something to be afraid of.”

This story originally appeared on wired.com.

SOURCE

Proteins let cells remember how well their last division went

EnlargeMartin Barraud

When we talk about memories in biology, we tend to focus on the brain and the storage of information in neurons. But there are lots of other memories that persist within our cells. Cells remember their developmental history, whether they’ve been exposed to pathogens, and so on. And that raises a question that has been challenging to answer: How does something as fundamental as a cell hold on to information across multiple divisions?

There’s no one answer, and the details are really difficult to work out in many cases. But scientists have now worked out one memory system in detail. Cells are able to remember when their parent had a difficult time dividing—a problem that’s often associated with DNA damage and cancer. And, if the problems are substantial enough, the two cells that result from a division will stop dividing themselves.

Setting a timer

In multicellular organisms, cell division is very carefully regulated. Uncontrolled division is the hallmark of cancers. But problems with the individual segments of division—things like copying DNA, repairing any damage, making sure each daughter cell gets the right number of chromosomes—can lead to mutations. So, the cell division process includes lots of checkpoints where the cell makes sure everything has worked properly.

But if a cell makes it through all the checkpoints, it’s presumably all good, right? Not entirely, as it turns out.

Mitosis is the portion of cell division where the duplicated chromosomes get separated out to each of the daughter cells. Spending a lot of time in mitosis can mean that the chromosomes have picked up damage, which may cause problems in the future. And prior research found that some cells derived from the retina will register when mitosis takes too long, and the daughter cells will stop dividing.

The new work, done by a team of researchers in Okinawa, Japan, and San Diego, started by showing that this behavior wasn’t limited to retinal cells—it seems to be a general response to a slow mitosis. Careful timing experiments showed that the longer cells spent trying to undergo mitosis, the more likely the daughter cells would be to stop dividing. The researchers term this system a “mitotic stopwatch.”

So, how does a cell operate a stopwatch? It’s not like it can ask Siri to set a timer—it’s largely stuck working with nucleic acids and proteins.

It turns out that, like many things relayed to cell division, the answer comes down to a protein named p53. It’s a protein that’s key to many pathways that detect damage to cells and stop them from dividing if there are problems. (You may recall it from our recent coverage of the development of elephant stem cells.)

A stopwatch made of proteins

The researchers found that, while mitosis was going on, p53 started showing up in a complex with two other proteins (ubiquitin-specific protease 28 and the creatively named p53-binding protein 1). If you made mutations in one of the proteins that blocked this complex from forming, the mitotic stopwatch stopped ticking. This three-protein complex only started building up to significant levels if mitosis took longer than usual, and it remained stable once it formed so that it would get passed on to the daughter cells once cell division was completed.

So, why does this complex form only when mitosis takes longer than usual? The key turned out to be a protein called a kinase, which attaches a phosphate to other proteins. The researchers screened chemicals that inhibit specific kinases that are active during mitosis and DNA repair, and found a specific one that was needed for the mitotic stopwatch. In the absence of this kinase (PLK1, for the curious), the three-protein complex doesn’t form.

So, the researchers think that the stopwatch looks like this: during mitosis, the kinase slowly attaches a phosphate to one of the proteins, allowing it to form the three-protein complex. If mitosis gets done quickly enough, the levels of this complex don’t get very high, and it has no effect on the cell. But if mitosis goes more slowly, then the complex starts building up, and it’s stable enough that it’s still around in both daughter cells. The existence of the complex helps stabilize the p53 protein, allowing it to stop future cell divisions once it’s present at high enough levels.

Consistent with this idea, all three of the proteins in the complex are tumor suppressors, meaning that mutations in them make tumor formation more likely. The researchers confirmed that the mitotic stopwatch was frequently defective in tumor samples.

So, that’s how individual cells manage to store one of their memories—the memory of problems with cell division. The mitotic stopwatch, however, is just one of the memory storage systems, with completely separate systems handling different memories. And, at the same time this is happening, a large number of other pathways also feed into the activity of p53. So, while the mitotic stopwatch may efficiently handle one specific type of problem, it’s integrated into a lot of additional, complex systems operating in the cell.

Science, 2024. DOI: 10.1126/science.add9528  (About DOIs).

SOURCE

"The Anxious Generation": An insecure generation? Yes, namely those of the parents

Some claims sound so obvious that they pass as general truths. Berlin can’t get anything done. Deutsche Bahn is a juice shop that is constantly delayed. And of course: Smartphones make our children stupid, lonely and dependent.

Anyone who advocates such theses can be sure that many people will nod in agreement, because it reflects a feeling that many people share. This is also the case with the thesis of a new book that has now been published in the USA and will also be available in German from June: The Anxious Generation ( in German: Generation Angst) by the US social psychologist Jonathan Haidt. The book’s intellectual subtitle: How the Great Rewiring of Childhood is Leading to an Epidemic of Mental Illness.

SOURCE

"Dragon's Dogma 2": Now slowly here!

It’s a shame. The desire for even more money undermines the gameplay that makes Dragon’s Dogma 2 so special. The game says: explore me in depth, take the time to really penetrate my world. The publisher Capcom says: or not. If you don’t feel like it, just spend a little money to make everything go faster.

There was a lot of excitement surrounding the release of the role-playing game Dragon’s Dogma 2 (PC, Playstation 5, Xbox Series X/S). The critics’ ratings are excellent. The pre-orders from players are numerous. But then it turns out: There are a number of microtransactions in the game that were not included in the test version of the game. Attempted blinding? Perhaps. But above all, it’s a lot of nonsense. Because if you don’t spend any extra money, you will have more fun playing.

SOURCE