News: Technology News

World's largest academic publisher says ChatGPT can't be credited as an author

Fri, 27 Jan 2023 19:24:25 +0000

In a recent development, the world's largest academic publisher, Springer Nature, has stated that the popular language model, ChatGPT, cannot be credited as an author. This decision has sparked a debate about the role of AI in the writing process and the ethical implications of giving credit to machines.

Springer Nature is a leading academic publisher that produces over 2,900 journals and more than 11,000 new books each year. The company is known for publishing high-quality research in a wide range of scientific, technical, and medical fields.

The problem with ChatGPT is that it is a machine-learning model and does not have the ability to understand the context or meaning of the text it generates. Therefore, Springer Nature argues that it cannot be credited as an author as it does not have the capacity for original thought or creativity.

This issue highlights the growing concern surrounding the use of AI in writing. While machine learning models like ChatGPT have the ability to produce large amounts of text quickly and efficiently, they lack the ability to understand the meaning and context of what they are writing. This raises questions about the ethics of using AI in writing, especially when it comes to giving credit to machines.

However, some argue that as AI technology advances, it will be able to understand the context and meaning of the text and may eventually be able to produce truly original content. This raises the question of how we should view AI-generated writing in the future.

For now, though, the debate is a little like saying academic authors should credit the pen they use to write with when publishing reports or books. At the moment, that’s all these machines are although, admittedly, it can be difficult to think of them like that considering how impressive the outputs can be.

With ChatGPT taking the world by storm, AI has been all the rage recently. However, in other AI news, one video game developer has been working hard to stop AI’s pervasive influence from ruining its gameplay.

World's largest academic publisher says ChatGPT can't be credited as an author

Do mechanical keyboards really need arrow keys?

Fri, 27 Jan 2023 08:25:44 +0000

Angry Miao replaces bulky arrow buttons with a 2×0.5-inch capacitive touchpad.

Which keys are absolutely essential to a keyboard? Many will tell you the entire numpad is, while others demand macro keys. I personally insist on a volume knob for my home office setup. And as someone who has tested 60 percent keyboards, which have no numpad or arrow keys, I'd add that for productivity and my sanity, arrow keys are also mandatory.

Arrow-less keyboards have their market, but for the vast majority, no arrows on a keyboard is a deal-breaker. A mechanical keyboard Angry Miao announced today asks us to consider an alternative, though. Instead of arrow keys or relying on a key combo for arrow input (like most 60 percent keyboard users do), it has a capacitive touch panel on the front edge for inputting arrow and other functions with your thumbs.

Is Angry Miao on to something here?

It’s not about the keyboard

Angry Miao's AM Compact Touch saves space by replacing arrow keys with a thumb-controlled touchpad.

Angry Miao/Indiegogo

Before we dive into this intriguing touchpad, let's get something out of the way: You probably don't want to buy the Angry Miao AM Compact Touch wireless keyboard. For one, the peripheral is still in the crowdfunding stage (Indiegogo launches on February 2, according to The Verge). Angry Miao is a known company among mechanical keyboard enthusiasts with several high-end limited product releases in recent years and successful crowdfunding campaigns. Other mechanical keyboard companies, like Keychron, tend to start products through crowdfunding, and some publications, including The Verge and TechCrunch, have tested AM Compact Touch samples. But investing money in any crowdfunding campaign comes with risks.

Beyond that, the AM Compact Touch has a niche layout with no arrow keys, numpad, or function row. Those still interested will also face a steep price, reportedly $398 to $615.

Still, I couldn't help being fascinated by the AM Compact Touch because it challenges how we normally navigate a keyboard. It also bestows function to a part of the keyboard that's virtually never used.

Do I find arrow keys necessary on a keyboard because they're imperative for daily computing or because that's what I'm used to? Now's as good a time as ever to consider if keyboards need a shake-up.

Are touchpads the next arrows?

The Compact Touch uses a capacitive touch panel that relies on a custom PCB and software the company says it spent three months optimizing. The touchpad is positioned to be accessible by two thumbs and is supposed to support four-way swipes and is approximately 2×0.5 inches, according to The Verge.

The panels add further control with support for long-pressing in four directions, Angry Miao claims. This enables actions like slide left and long-press for rewind, or swipe up and long-press to turn up the volume.

Below is a video demoing the feature. Angry Miao published the video, though, so keep that in mind when considering its apparent ease of usability.

Angry Miao's new layout: Touch panel hands-on.

Thinking about applying this concept to other keyboards, the touchpad's location seems wise, since it looks like it would be within reach of both thumbs, save for times when you're using a numpad (if the keyboard has one). Angry Miao says this means you can use the touchpad without moving your wrist, and less physical movement can be helpful for ergonomics.

A touchpad that's a successful arrow replacement could make 60 percent keyboards more appealing, especially for travel and freeing up desk space. The design could even provide "arrow" access with the left hand, which you don't get with a traditional keyboard layout.

But there's one thing painfully missing from touch panel input that mechanical keyboard users adore: tactility. Mechanical keyboard loyalists like input with some oomph behind it, whether that oomph is tactile travel, a discernible bump, and/or a proud clicking noise. Using a flat touch panel in lieu of addictive tactile keys seems counterintuitive.

Mechanical keyboards with high price tags also are wise to be as versatile as possible, and the touch panel on the Compact Touch seems limited when it comes to tasks like repetitive line-by-line scrolling and advanced movements, like steering a car in a game.

Early hands-on reviews suggest the Compact Touch's touchpad is decently, but not absolutely, reliable. TechCrunch reported that it largely worked as expected and was "easy enough to go back a few letters" when writing, but that for other tasks, it's "a guessing game whether you’ll be able to time things right to stop the cursor where you need to." The reviewer ultimately ended up programming arrow input into a secondary layer on the keyboard, suggesting they didn't want to use the touchpad long-term.

TechCrunch's article also pointed out that the location of the touch panel makes it harder to use a wrist rest.

The Verge said there were "many occasions" when the reviewer's cursor went up when they intended for it to go left and down when they wanted it to move to the right. They also said swipe and holding was "easy to overshoot," noting that "since Windows and Mac operating systems are not built for this," this is something you'd have to deal with sometimes.

The final product may arrive more polished. But these early touchpad experiences suggest it's reasonable to expect inherent frustration in the immediate term and greater obstacles than just personal adjustment in the long term.

Often imitated, never duplicated

Early reviews suggest that Angry Miao's specific take on the touch panel has room for improvement. Dealing with potentially wonky OS support is a no-go for impatient people like me, but, as with foldable PCs, that doesn't automatically mean we shouldn't explore the idea further. Greater interest could eventually yield greater support.

However, a touch panel's inferior tactility compared to keys is hard to move past. So many things in modern tech, like button-less smartphones and tablets, demand you swipe. Mechanical keyboard people love the feeling of a great keypress, and a flat touchpad just can't duplicate that.

Angry Miao isn't the only one with a keyboard using imperfect capacitive touch. When I tested Asus' ROG Falchion NX (which has a capacitive touch panel running up the left side for things like volume control) for our best wireless mechanical keyboards roundup, it was frustrating to do something simple like setting the exact volume level I wanted.

Of course, there are other types of keyboards besides mechanical ones, and some, like the Logitech K400 Plus with a touchpad where the numpad typically is and Lenovo's ThinkPad Keyboard with a red nub, have their own arrow alternatives. But those keyboards also include arrows still, broadening appeal.

Logitech's K400 Plus keyboard.

Logitech

With support for long-presses, a keyboard touchpad could also provide a broader range of input than standard arrow keys. The touchpad could perform double duty, like serving as a volume knob. That frees up room for other keys, like macro or media ones or even standard keys like function rows, while still conserving space.

And for those who already don't require arrow keys on their keyboards—yes, 60 percent keyboards exist for someone—most claim that using a key combination instead somehow becomes second nature eventually. Personally, a keyboard with an assumedly perfectly reliable touchpad and no arrow keys would appeal to me more than a keyboard that solely relies on a key combo for arrows. And with the space savings, there's potential for an extremely portable travel keyboard.

For portability and maximizing desk space, a capacitive touch panel is an interesting challenger to arrow keys. If fine-tuned to work smoothly with every swipe and long-press, it could enable unique keyboard designs and give the 60 percent design a broader appeal.

But for those who prefer more traditional forms of input (like mechanical switches) over newer ones (like touchscreens and voice command) and who don't have the patience for frequent key combinations, dedicated arrow keys will continue being a critical, preferred part of daily computing for the foreseeable future.

Ars Technica may earn compensation for sales from links on this post through affiliate programs.

Do mechanical keyboards really need arrow keys?

AirPods are earplugs now

Thu, 26 Jan 2023 18:58:07 +0000

Noise-canceling wireless earbuds aren’t just great for listening to music or podcasts — they are also great for turning down the volume of the world around you.

Lately, I’ve been using my AirPods to hear nothing.

My usage of Apple’s AirPods Pro and other noise-canceling wireless earbuds has changed from when I first started wearing them half a decade or so ago. Back then, I’d pop in earbuds to listen to music, stream a podcast, or watch video on my phone or tablet while in a public place. I’d take meetings and calls with them or put them in when doing yard work for some motivational metal. On walks around the neighborhood, my earbuds would accompany me with audio to keep me entertained and moving.

Now, I most often don’t listen to anything when wearing them. I’ll pop the AirPods or another noise-canceling earbud in my ears when sitting in a coffee shop where the music and chatter is too loud; I’ll wear them in my home office to cut the sound of my fan or air purifier. I’ll even use them when I’m sitting in my favorite chair reading a book so I’m not distracted by kids wreaking havoc a room over. I’m not playing music or a podcast or anything in those instances; I’m just wearing the AirPods with their noise-cancellation features enabled.

It’s undeniable that modern life has gotten very loud. Urban environments are noisy, coffee shops and restaurants routinely turn music up to club levels, and open office plans are widely criticized for how much noisier they are than cubicle farms. Anyone who lives in a home with an open floor plan and more than one other person can attest to how noisy the space becomes when you can’t shut doors to separate rooms.

That noise primarily makes it hard for me and a lot of other easily distracted people to relax or focus, but it can also be detrimental to your hearing over time. Noise pollution has become such a topic of concern that the Apple Watch will even give you a warning when you’ve been exposed to high decibels for an extended period.

Now, noise-canceling earbuds are not effective enough to completely mute the world. (In my experience, Samsung’s first-gen Galaxy Buds Pro are the most comfortable and effective at canceling noise such as conversation, but since they don’t integrate as nicely as the AirPods Pro do with my daily devices, AirPods it is for me.) They also aren’t my choice for long travel on an airplane — over-ear headphones are much more effective for that.

Bose, Apple, Sony, and others all have effective enough noise-cancellation features.

Photo by Chris Welch / The Verge

But wireless earbuds are good at reducing the amount of noise that gets to my eardrums. It’s like turning the world down from an 11 to a more reasonable four or five. If a person speaks to me directly, I can generally hear them, but if they are having a conversation with someone else nearby, I’m not distracted by what they are talking about. I can still hear enough of my kids’ play to know they are still alive (or if someone has gotten hurt and needs an adult), but I don’t get irritated by every little scream and shriek that emanates from wherever they are. (Any parent can tell you, little kids scream and shriek A Lot.) They are also more convenient to have with me anywhere than over-ear headphones.

Wireless earbuds are also easy to briefly remove when I do need to hear someone clearer or want to listen for traffic or another safety concern. Popping an AirPod out of my ear is typically so much quicker and easier than trying to switch between noise-cancellation and transparency modes that I almost never use the latter.

At this point, you’re probably wondering why I’m using expensive technology that needs to be charged and cared for to do something that a cheap foam earplug can do just as well. And to that I say, fair. You got me. But I already have the AirPods, and when they are in my ears, I can choose to either listen to something or not, which isn’t an option with foam earplugs.

It’s also interesting that my usage of AirPods dovetails with their recently acquired ability to amplify the noise around you as a stand-in for hearing aids. That’s certainly helpful for those that have hearing loss, but for me, for now, I appreciate their cancellation abilities more.

So instead of shaking my fist and yelling at everyone to turn it down, I pop my earbuds in and go on about my day. After all, silence is bliss.

AirPods are earplugs now

Why are so many tech companies laying people off right now?

Thu, 26 Jan 2023 18:55:23 +0000

Didn’t they just have record-breaking profits?

There’s an eerie similarity to the statements tech companies have made about their recent layoffs. Mainly, if the press releases are to be believed, the C-suite of every Big Tech company on Earth — well, with the notable exception of Apple, which has not announced layoffs — figured no one would ever go outside or spend money offline again after the pandemic and their various online businesses would stay just as big as they were during the heights of covid.

I do love a heavily lawyered statement that was clearly written by the public relations department! In fact, these are all so similar that they might as well have come from the same PR person. It kind of seems like tech firms are laying off workers because… other tech firms are laying off workers.

Let’s be real, none of these companies are teetering on the edge of bankruptcy — in fact, they were recently minting money. That money did not evaporate. And as any person who’s been through job cuts can tell you, it’s generally not about performance, either! Essentially, someone went through a budget and zeroed out a bunch of line items that happened to be, you know, people’s jobs. The question, then, is why a company might make job cuts that don’t seem especially necessary.

The answer is that investors have changed how they’re evaluating companies, says Michael Cusumano, the deputy dean at the MIT Sloan School of Management. Generally, when companies are growing really fast — like when revenue is shooting up 20 percent or 30 percent a year — nobody cares about profits, Cusumano says. But we’re not in a growth period right now, so investors are being more cautious.

What tech companies said about their layoffs

Meta:

“At the start of Covid, the world rapidly moved online and the surge of e-commerce led to outsized revenue growth. Many people predicted this would be a permanent acceleration that would continue even after the pandemic ended. I did too, so I made the decision to significantly increase our investments. Unfortunately, this did not play out the way I expected.”

Google:

“Over the past two years we’ve seen periods of dramatic growth. To match and fuel that growth, we hired for a different economic reality than the one we face today.”

Microsoft:

“As we saw customers accelerate their digital spend during the pandemic, we’re now seeing them optimize their digital spend to do more with less.”

Salesforce:

“As our revenue accelerated through the pandemic, we hired too many people leading into this economic downturn we’re now facing, and I take responsibility for that.”

Amazon (1):

“As you know, we continue to face an unusual and uncertain macroeconomic environment. In light of this, we’ve been working over the last few months to further prioritize what matters most to our customers and the business. After a deep set of reviews, we recently decided to consolidate some teams and programs.”

Amazon (2):

“This year’s review has been more difficult given the uncertain economy and that we’ve hired rapidly over the last several years ... Today, I wanted to share the outcome of these further reviews, which is the difficult decision to eliminate additional roles. Between the reductions we made in November and the ones we’re sharing today, we plan to eliminate just over 18,000 roles.”

Spotify:

“Like many other leaders, I hoped to sustain the strong tailwinds from the pandemic and believed that our broad global business and lower risk to the impact of a slowdown in ads would insulate us. In hindsight, I was too ambitious in investing ahead of our revenue growth. And for this reason, today, we are reducing our employee base by about 6% across the company.”

Stripe:

“At the outset of the pandemic in 2020, the world rotated overnight towards e-commerce. We witnessed significantly higher growth rates over the course of 2020 and 2021 compared to what we had seen previously. As an organization, we transitioned into a new operating mode and both our revenue and payment volume have since grown more than 3x. The world is now shifting again.”

Tech companies have “tens of billions, often hundreds of billions of dollars, collectively, in reserves,” Cusumano says. “But they don’t really use that to support operations.” When an investor is reading an earnings statement, those reserves aren’t what they’re thinking about, either. One measure people use for measuring tech companies’ investment value is revenue per employee — and having hired all this staff during the pandemic, that means revenue per employee has gone down.

“Software companies like Microsoft, they should have like $500 million per employee,” Cusumano says. “It could be higher than that, but when it starts to get below that, you start to worry that they’ve got too much headcount. So that’s something people look at on a yearly or even quarterly basis.”

The theory behind layoffs is that they save the company money, even though there’s an initial expenditure of millions or billions of dollars in severance. The idea is, with fewer salaries, the company’s costs are lower on an ongoing basis. I asked Cusumano if that was empirically true. He said he wasn’t sure.

So I called up someone who’s been studying this kind of thing for a long time: Jeffrey Pfeffer, a professor at the Stanford Graduate School of Business. When I asked him about the similarities in the company statements, his answer was succinct: the tech companies are copying each other. “I think Peter Drucker [who is widely known as a father of management thinking] was quoted as saying something to the effect of ... thinking is hard work, which is why most managers don’t do it,” Pfeffer told me.

Layoffs probably don’t cut costs, Pfeffer says. In fact, there is little empirical evidence that layoffs help improve profitability, and some evidence they actually hurt profitability, he says. “Oftentimes, companies don’t have a cost problem,” Pfeffer says. “They have a revenue problem. And cutting employees will not increase your revenue. It will probably decrease it.”

The literature on whether layoffs actually work to boost stock price is mixed: in one study, companies that closed plants and did layoffs had better returns than companies that only did layoffs. During the 2020 coronavirus pandemic, layoffs had no effect on stock prices at all.

Layoffs do have one concrete impact. Pfeffer’s research has found that layoffs literally kill people — by increasing the risk that someone will die by suicide and by levering up stress, both among people laid off and among those who remain. Layoffs may also reduce productivity among those who remain employed.

So why do layoffs at all if they don’t actually work? “People do all kinds of stupid things all the time,” Pfeffer says. “I don’t know why you’d expect managers to be any different.”

Why are so many tech companies laying people off right now?

IBM cuts 3,900 jobs

Thu, 26 Jan 2023 18:43:58 +0000

IBM announced today that it will lay off about 3900 workers, which is less than 2% of the global workforce. The company is the last technology company to lay off workers. IBM's workforce reduction is not as large as that of other major technology companies. Google-parent Alphabet announced 12000 job cuts, Microsoft 10000 and Amazon 18000.

The layoffs are mostly the results of earlier asset sales and less a result of the current market climate, according to IBM. IBM sold or restructured several assets in the last two years. The company's legacy technology services business was spun off in 2021 and its health care data analytics business was sold in 2022.

The layoffs will cost IBM about $300 million to pay for employee severance costs. The layoffs were announced during an investor conference call for the financial results of the fourth quarter of 2022. IBM is still committed to hiring "for client-facing research and development" according to the company's CFO.

The company's revenue and operating profits matched the predictions of analysts. IBM managed to earn $16.7 billion in revenue in the quarter, which was $0.3 billion higher than what Wall Street hat estimated. Earnings per share was $3.60, which beat analyst estimates by $0.02 per share. The operating pre-tax income was $3.8 billion.

CEO Arvind Krishna attributed the success to IBM's hybrid cloud and artificial intelligence strategy. Software revenue was up 8%, consulting 9% and infrastructure 7% according to Jim Kavanaugh, IBM's Chief Financial Officer.

Revenue by division for the fourth quarter of 2022 according to the published earnings report.

Software: $7.3 billion
Consulting: $4.8 billion
Infrastructure: $4.5 billion

The company's shares fell more than 4% today and are trading at $134 at the time of writing.

IBM expects that its revenue will grow in the mid-single digits and that its free cash flow will grow to $10.5 billion in fiscal 2023. The free cash flow would increase by more than $1 billion if the prediction holds.

Chief Financial Officer James Kavanaugh revealed on Wednesday that the weak Dollar was helping IBM but that the currency effect would be neutral in 2023 overall. IBM plans to expand its strategic partnerships and increase investments in specific growth markets in 2023.

A full transcript of the earnings call is available here.

Source

A robot was scheduled to argue in court, then came the jail threats

Thu, 26 Jan 2023 15:54:57 +0000

A British man who planned to have a "robot lawyer" help a defendant fight a traffic ticket has dropped the effort after receiving threats of possible prosecution and jail time.

Joshua Browder, the CEO of the New York-based startup DoNotPay, created a way for people contesting traffic tickets to use arguments in court generated by artificial intelligence.

Here's how it was supposed to work: The person challenging a speeding ticket would wear smart glasses that both record court proceedings and dictate responses into the defendant's ear from a small speaker. The system relied on a few leading AI text generators, including ChatGPT and DaVinci.

The first-ever AI-powered legal defense was set to take place in California on Feb. 22, but not anymore.

As word got out, an uneasy buzz began to swirl among various state bar officials, according to Browder. He says angry letters began to pour in.

"Multiple state bar associations have threatened us," Browder said. "One even said a referral to the district attorney's office and prosecution and prison time would be possible."

In particular, Browder said one state bar official noted that the unauthorized practice of law is a misdemeanor in some states punishable up to six months in county jail.

"Even if it wouldn't happen, the threat of criminal charges was enough to give it up," he said. "The letters have become so frequent that we thought it was just a distraction and that we should move on."

State bar associations license and regulate attorneys, as a way to ensure people hire lawyers who understand the law.

Browder refused to cite which state bar associations in particular sent letters, and what official made the threat of possible prosecution, saying his startup, DoNotPay, is under investigation by multiple state bar associations, including California's.

In a statement, State Bar of California Chief Trial Counsel George Cardona declined to comment on the probe into DoNotPay but said the organization has a duty to investigative possible instances of unauthorized practice of law.

"We regularly let potential violators know that they could face prosecution in civil or criminal court, which is entirely up to law enforcement," Cardona said in a statement.

Leah Wilson, the State Bar of California's executive director, told NPR that there has been a recent surge in poor-quality legal representation that has emerged to fill a void in affordable legal advice.

"In 2023, we are seeing well-funded, unregulated providers rushing into the market for low-cost legal representation, raising questions again about whether and how these services should be regulated," Wilson said.

Even if the use of AI in court was not being challenged, some observers have questioned just how effective DoNotPay's AI tools would be for people in need of legal services, with some having mixed to shoddy results attempting to use its basic features.

Browder has been known for drumming up attention with stunts. Earlier this month, he claimed on Twitter that the company would pay any lawyer $1 million to argue in front of the U.S. Supreme Court wearing AirPods that would pipe AI-generated arguments from its "robot lawyer."

Founded in 2015, DoNotPay has raised $28 million, including funding from prominent venture capital firm, Andreessen Horowitz, according to analytics firm PitchBook.

Pivoting away from AI legal defense amid threats

Instead of trying to help those accused of traffic violations use AI in the courtroom, Browder said DoNotPay will train its focus on assisting people dealing with expensive medical bills, unwanted subscriptions and issues with credit reporting agencies.

Browder also still hopes it is not the end of the road for AI in the courtroom.

"The truth is, most people can't afford lawyers," he said. "This could've shifted the balance and allowed people to use tools like ChatGPT in the courtroom that maybe could've helped them win cases."

The future of robot lawyers faces uncertainty for another reason that is far simpler than the bar associations' existential questions: courtroom rules.

Recording audio during a live legal proceeding is not permitted in federal court and is often prohibited in state courts. The AI tools developed by DoNotPay, which remain completely untested in actual courtrooms, require recording audio of arguments in order for the machine-learning algorithm to generate responses.

"I think calling the tool a 'robot lawyer' really riled a lot of lawyers up," Browder said. "But I think they're missing the forest for the trees. Technology is advancing and courtroom rules are very outdated."

Source

Nvidia GeForce RTX 3060 Was The Most Googled Card In 2022

Wed, 25 Jan 2023 20:26:42 +0000

As per a report, Nvidia’s mid-range GeForce RTX 3000 series offerings were the most Googled graphics cards last year. GeForce RTX 4090 came fifth in the list.

If one looks at all the reviews and discussions happening on sites, one would think that $1000 and above graphics cards would be the most sold ones out there. That’s far from reality.

While most sites and advertisements concentrate on the flagship graphics cards and even CPUs, it’s the budget and mid-range ones which are the most purchased ones by the common people.

Reasons are simple. Expensive products have the highest profits. So it’s understandable for companies like Nvidia and AMD to market them more. What gets ignored that most people in the world cannot buy them and would rather prefer to game on cheap graphics cards.

Going by Google searches, that seems to be the case.

Nvidia GeForce RTX 3060 and RTX 3050 Most Searched Graphics Cards in 2022

Most searched graphics cards in 2022. Credit: Raul Bilc for razzem.com.

As per a study conducted by Raul Bilc for razzem.com, Nvidia GeForce RTX 3060 was the most searched graphics card on Google in 2022. Launched in 2021, RTX 3060 was searched 1,467,000 times, totaling almost 9% of all the search done in the study.

Next in the list was the $249 Nvidia GeForce RTX 3050, with 1,161,800 searches. Amounting to almost 7% of searches.

Third on the list was the $699 Nvidia GeForce RTX 3080, with 1,006,500 searches. Amounting to exactly 6% of searches.

Fourth on the list was the $499 Nvidia GeForce RTX 3070, with 973,100 searches. Amounting to 5.80% of searches.

The most interesting was the fifth entry in the list. The $1,599 Nvidia GeForce RTX 4090, with 822,600 searches done. Amounting to 4.90% of the searches done in the study. The RTX 4090 was released in October, but was in a lot of news for various reasons, including cable faults.

It was followed by the likes of Nvidia GeForce RTX 2060, GTX 1650, RTX 3090, RTX 3060 TI and others.

The Radeon RX 6600 was the only card from AMD in the top ten.

In total, more than 60 graphics cards were tested, which gives us a detailed perspective. Those who want to check the full table in detail with are advised to check the study.

There are few conditions to the study methodology, though. The author specifies that the study was done using 3 different (highly trusted) SEO tools. Additionally, it was done on worldwide search results, which we think it’s great. All search results were from only from Google. Only Nvidia GeForce cards from 10 series onward and AMD Radeon cards from 500 series onward were checked. Memory sizes were not considered to not complicate any results. Among many other things.

We don’t find anything wrong in the research methodology. So it seems quite credible.

Steam Hardware Survey Results Show Something Similar

Steam Hardware Survey, December 2022. Credit: Valve Corporation.

The Steam Hardware Survey from December 2022 shows something similar in the graphics card usage results. If we ignore NVIDIA GeForce GTX 1650, GTX 1060 and GTX 1050 Ti which are used mostly by pre-built PC makers, the most used cards include NVIDIA GeForce RTX 3060 and GeForce RTX 3070 in top used cards.

Sure, searching for cards is different from buying them. But it gives us a good idea. The results again show that budget gamers, often sidelined and ignored, buy the most graphics cards.

Nvidia GeForce RTX 3060 Was The Most Googled Card In 2022

Artificial Intelligence May Hit The Singularity Within 7 Years, Researchers Claim

Wed, 25 Jan 2023 20:22:26 +0000

Researchers have claimed that artificial intelligence (AI) will reach the singularity within seven years, after attempting to quantify its progress.

Translation company Translated, presenting their work at an Association for Machine Translation in the Americas conference, explained that they first began testing machine translation technology in 2011. The team settled on a metric to measure AI progress, which they've called "Time to Edit" (TTE). Simply put, it is the time it takes a human translator to edit a translation produced by another human or an AI.

Over the years, the TTE for AI-translated texts has come down fairly consistently, leading Translated to predict the date when AI hits the singularity, when the time is equivalent to human translators.

"When plotted graphically, the TTE data shows a surprisingly linear trend," the team wrote on their website. "If this trend in TTE continues its decline at the same rate as it has since 2014, TTE is projected to decrease to one second within the next several years, approaching a point where MT [machine translation] would provide what could be called 'a perfect translation'."

The team highlights other measures the AI is improving on, such as linguistic errors per 1,000 translated words. Currently, industry AI makes around 50 errors per 1,000 translated words, with the team expecting that to improve at roughly the same rate as TTE.

The team believes that the ability to translate, a complex task for a machine, is a good measure of when technology reaches singularity with humans.

"Many AI researchers believe that solving the language translation problem is the closest thing to producing Artificial General Intelligence (AGI)," the team writes. "This is because natural language is by far the most complex problem we have in AI. It requires accurate modeling of reality in order to work, more so than any other narrow AI."

The term "singularity" used by the team appears to mean "on par with humans", and may be an indication of AGI, rather the traditional meaning – the point where technological growth is out of our control, and irreversible. One popular version of the singularity is the "intelligence explosion", where AI is able to improve itself more and more with each generation cycle, making it far surpass human intelligence.

Nevertheless, tasks such as translation on par with humans would be an important milestone in AI, even if we won't be seeing machines grow out of our control by 2030. It could be used as a part of a test for AGI, though it's important to remember that language ability does not make a machine conscious and language models are essentially extremely good predictive texts, rather than intelligence that understands the meaning.

However, tracking the progress of AI is still pretty cool, and will only become more important as we grow closer to that theoretical singularity point.

"All of us understand that we are approaching singularity in AI," Translated CEO Marco Trombetti said of the research. "For the first time, we have been able to quantify the speed at which we are progressing toward it."

Source

Microsoft to face antitrust probe over Slack complaint about Teams

Wed, 25 Jan 2023 20:21:30 +0000

Last year, we heard that Microsoft could be the target of an antitrust probe from the European Commission (EC) due to a complaint filed by Slack regarding Teams. Now, a new report claims that this is definitely happening and that the Redmond tech giant is facing formal scrutiny soon.

A report from Politico notes that multiple EC sources have confirmed that Microsoft will be facing an antitrust probe soon regarding bundling Teams with its other Microsoft 365 software. Although the details are currently unclear, the regulator is reportedly planning a "statement of objections" regarding how Microsoft treats its competition unfairly in an anti-competitive manner. An EC spokesperson also confirmed that an assessment regarding the probe is ongoing.

For those unaware, the root of the probe is a complaint filed by Slack in 2020 in which it stated that Microsoft hides the "true cost" from enterprise customers when it bundles Teams with the Microsoft 365 suite of products and forces it to be installed on machines. Back then, it urged the regulator to force Microsoft to remove Teams from its Microsoft 365 suite and offer it separately at "fair" commercial prices. The EC already sent out questionnaires to concerned parties to gain more insight into the matter.

Needless to say, this is not a good time for Microsoft to be facing yet another antitrust probe. The company has already been under scrutiny for quite some time from some regulators regarding its ongoing $69 billion Activision Blizzard acquisition.

Source: Politico

Microsoft to face antitrust probe over Slack complaint about Teams

Forspoken on PCIe 5.0 NVMe hints at the true power of Microsoft DirectStorage on Windows 11

Wed, 25 Jan 2023 20:20:24 +0000

Luminous Studios' Forspoken is out on PC and a big reason for its anticipation among fans and enthusiasts was the debut of Microsoft's DirectStorage API. The feature promised enormous improvements in performance over legacy technology and preliminary tests indeed show extremely impressive results.

While the previous coverage focused on loading time improvements, some more testing by the YouTube channel Compusemble adds details like disk usage and disk speeds, which shows the SSD controller usage percentage and the disk data reading speeds. It also gives us an idea about how DirectStorage can affect frame pacing. And just like load times, the frametimes in Forspoken also appear to be vastly improved as a result of the technology. The outcome is a generally stutter-free experience even when the game is streaming assets on the fly.

The images below are examples of such instances when the game is streaming in data but the frame pacing is quite even with no hitches or stutters which are characterized by large spikes in the frame time graph. One of the scenes below receives data at 830.7MB/s but the frame timings are fairly smooth.

While part of this performance credit goes to DirectStrorage, some of it is also due to the latest generation PCIe 5.0 SSD that was used in this test. The YouTuber says that a Gen5 Engineering Sample NVMe SSD with the Phison E26 controller was used for this.

However, despite the positives, there certainly are aspects of the game that may need a bit more polish. For example, the screenshots below show a sequence of scenes where Chapter 2 "Stuck" loads up. As you may note, it exhibits a few very noticeable frametime spikes and takes nearly eight seconds to load up. This is despite the read speeds not really being that high as the Gen5 SSD here is capable of much more, hence suggesting there may be issues elsewhere in the optimization pipeline.

Perhaps the biggest complaint one might have about Forspoken's DirectStorage implementation is the absence of the feature on Windows 10. However, Microsoft has already confirmed that Windows 11 features "additional optimizations in the IO stack" which may imply that the new OS would have been the better choice anyway.

Source: Compusemble (YouTube)

Forspoken on PCIe 5.0 NVMe hints at the true power of Microsoft DirectStorage on Windows 11

Shape-Shifting Robot Escapes Miniature Prison Cell By Liquifying Before Re-solidifying

Wed, 25 Jan 2023 20:20:03 +0000

If your name is John or Sarah Connor, this might be the moment to start covering your tracks.

A robot that can shift between solid and liquid states has been filmed escaping from a miniature jail cell with bars too close together to allow it to leave in solid form. The creators claim they were inspired by sea cucumbers’ capacity to alter their tissue stiffness – but the scene is just a little too similar to Robert Patrick liquifying his way through the mental hospital bars for us to believe them. We even see the famous reabsorption of the little bit left behind.

Hard-bodied robots are common, even if they have yet to reach the capacities of science fiction films. Their soft-bodied counterparts can get into tight spaces, but what they can do there is limited, and they are also difficult to control.

A team led by Dr Chengfeng Pan of the Chinese University of Hong Kong has made a robot that can swap states to whichever is most needed, with a video that sums it up. The prison escape may trigger our fears, but robots like these could also provide lifesaving services others cannot.

“Giving robots the ability to switch between liquid and solid states endows them with more functionality,” Pan said in a statement.

The achievement relies on a material that can shift between solid and liquid under the influence of a magnetic field, which the authors call a “magnetoactive solid-liquid phase transitional machine.”

To make it, they needed a metal that turns to liquid close to room temperature. Mercury's conditions to solidify are too cold, and most other metals require too much heat to liquefy. However, gallium fits the bill, with a melting point of just 29.8 °C (85.6 °F).

The team added neodymium-iron-boron magnetic microparticles to the gallium, which Professor Carmel Majidi of Carnegie Mellon University said serves two roles. “They make the material responsive to an alternating magnetic field, so you can, through induction, heat up the material and cause the phase change.” The induction warmed the gallium from 25 to 35 °C (77 to 95 °F), “But the magnetic particles also give the robots mobility and the ability to move in response to the magnetic field,” Majidi added.

Phase-shifting materials have been made before, but needed external heat sources or electrical currents to transform. Neither is ideal if you want to send the robot somewhere hard to reach, such as inside the human body.

The paper reports that, in addition to the achievements in the video, their little robot can jump 21-millimeter (0.8-inch) moats and climb walls when solid, but subdivide to get around objects before rejoining when liquid. “Now, we’re pushing this material system in more practical ways to solve some very specific medical and engineering problems,” Pan said.

Human body temperature means a metal with a higher melting point than gallium will be required, and the phase change takes longer in liquids than in air because it loses heat faster to the environment. Nevertheless, the potential exists to rapidly deliver drugs to, or remove foreign objects from, the stomach.

Even in the solid phase, the robot only reaches 1.5 m/s (5.4 km/hr) so most people could outrun it if it turned nasty. What the paper describes as “high load capacity” is about 30 kilograms (66 pounds), which is impressive but also not very threatening – but it might be wise to prepare some liquid nitrogen or molten steel for the future.

The paper is published in Matter.

Source

Microsoft posts $52.7 billion revenue in latest quarter with weak Xbox and Windows results

Wed, 25 Jan 2023 20:16:56 +0000

It's that time of the year again, where Microsoft shares its latest earnings for the benefit of investors and the awareness of the public. While it has experienced a relatively strong quarter primarily due to strong growth in cloud and enterprise sectors (Office, LinkedIn, Dynamics, Server), there are a few worrying aspects in the latest results as well.

Rather than making you go through a long-form review of Microsoft's financial figures for FY23 Q2 ending on December 31, 2022, we have decided to list the highlights in bullets this time for better readability. You can view these below:

Revenue: $52.7 billion (+2% compared to same quarter last year)
- Productivity and Business Processes: $17.0bn (+7%)
- Intelligent Cloud: $21.5bn (+18%)
- More Personal Computing: $14.2bn (-19%)
Gross margin: $35.3bn (+1%)
Operating income: $20.4bn (-8%)
Net income: $16.4bn (-12%)
Diluted earnings per share: $2.20 (-11%)

As can be seen, despite exhibiting strong growth in cloud and business, its revenue gains are mostly offset by the comparative decrease in the More Personal Computing category, also resulting in a lower net income and diluted earnings per share. It is important to understand that Microsoft isn't losing money, its growth is just lower compared to the same period in the last fiscal year.

Perhaps most interesting is the lack of growth in the More Personal Computing sector, which primarily comprises of consumer devices and services:

Windows OEM and Commercial revenue down by 39% and 3% respectively due to weak PC demand and low standalone product sales
Xbox content and services decreased by 12% due to lack of first-party content despite growth in Xbox Game Pass
Devices revenue reduced by 39% due to weak PC demand and "execution challenges" when it comes to launching new products
Search and news advertising revenue excluding traffic acquisition costs up by 10% because of Xandr acquisition and higher search volumes

CEO Satya Nadella highlighted the increased focus in AI moving forward, also driven by Microsoft's OpenAI investments:

The next major wave of computing is being born, as the Microsoft Cloud turns the world’s most advanced AI models into a new computing platform. We are committed to helping our customers use our platforms and tools to do more with less today and innovate for the future in the new era of AI.

In the next quarter, Microsoft expects revenue to be around $50.5 billion to $51.5 billion with operating margin decreasing by 2% YoY, which means that investors can expect similar results after three months too. Most notably, Microsoft expects its More Personal Computing revenue to decline further to $11.9-12.3 billion. Although the situation isn't really dire, it's clear that with recent layoffs, Microsoft has its work cut out.

Microsoft posts $52.7 billion revenue in latest quarter with weak Xbox and Windows results

How to restore Wikipedia's old design

Wed, 25 Jan 2023 20:14:09 +0000

Last week, Wikipedia rolled out a change an updated version of its desktop interface. It has a modern look, and is actually one of the biggest changes made to the website in over a decade.

The new design is called Vector 2022, the old one was Vector 2010 and is now called Legacy Vector. I use Wikipedia from time to time as a reference source for work, to learn about various things like history, sports, etc., just like many people do. If you like the new design, that's cool. But I'm sure there are users who dislike it. I'm not a fan of the redesign, I think it looks ridiculous on a large monitor with a lot of wasted space on the sides.

You can click the maximize button that appears in the bottom right corner of the pages to expand the view, but it still doesn't offer the same experience as before. The official explanation given by MediaWiki (which Wikipedia uses) for the change is that shorter lines make for easy reading, as users don't have to move their eyes a lot.

If you don't mind the new Wikipedia design, but want a persistent sidebar (h/t reddit) to appear when you scroll up or down a page, you can add the following rule to uBlock Origin's filters. wikipedia.org###vector-toc-pinned-container:style(position: static !important;)

Thankfully, there are a few ways to restore the old Wikipedia design. One option is to sign up for a free account on Wikipedia, and enable the Legacy Vector theme from the Settings.

How to restore Wikipedia's old design

Go to Wikipedia and log in to your account.
Click on Preferences > Appearance, or just go to this URL. https://en.wikipedia.org/wiki/Special:Preferences#mw-prefsection-rendering
The Skin section has the Vector 2022 option set as the default experience, click on Vector 2010 to set it as the preferred design.
Scroll down the web page, and click the save button to apply the changes.

Not everyone may want to sign up for an account just to change a single setting, right? In that case, you can add the following parameter at the end of Wikipedia URLs. ?useskin=vector

For example, https://en.wikipedia.org/wiki/Microsoft_Windows is the default link of the page. Add the suffix, and it becomes https://en.wikipedia.org/wiki/Microsoft_Windows?useskin=vector

This allows you to use the site's old style, but the problem is you will have to add the parameter to every link that you open. And on pages that are divided by sections, and tables, you will need to add it after the URL, but before the # part. That's a tedious way to browse the site, but you cannot save your preferences, since the settings are not available for users without an account. So this is the only option that you have without relying on external sources.

If you know your way around UserCSS, you can play with the styles here to tailor the appearance to your liking.

Tip: You can use email masking services like Firefox Relay, DuckDuckGo Email Protection, SimpleLogin or an email alias service of your choice to hide your real email ID, and use the one provided by the service to create a Wikipedia account for personal usage.

Browser extensions to restore the old Wikipedia interface

Wikipedia is suggesting users to use browser extensions to customize their user experience. A few add-ons have popped up which offer an easy way to restore Wikipedia's legacy look. These are of course made by third-parties, indie developers.

Old Wiki is the only one I found on the Chrome Web Store, it works fine on Wikipedia, and Wikimedia sites. All you have to do is install the plugin and let it do the job. Here is its GitHub page.

Firefox users have a few options to choose from. Wikipedia Vector Skin is one such add-on. An alternative is to use the Classic mode for Wikipedia add-on, the advantage here is that unlike Wikipedia Vector Skin, the former supports a few more Wiki-websites.

All these extensions basically just add the ?useskin=vector parameter that I mentioned above to the URL, but since they do it automatically, they are quite convenient to have.

How to restore Wikipedia's old design

Samsung’s new touch tech enables thinner, lighter OLED laptops

Wed, 25 Jan 2023 20:09:32 +0000

Incorporating a design phones have used for over a decade.

As plenty of laptop makers proved at CES earlier this month, there will be plenty of OLED laptops to choose from in 2023. The technology's high-contrast, rich image quality has tempted creatives and power users alike. A new year means new ways to entice shoppers, though, so one approach Samsung plans to take with its next series of Galaxy Books concerns the screen's physical build.

On Tuesday, Samsung Display, which makes display panels for various companies, including parent company Samsung Electronics, announced that it started mass production of a new touch-capable OLED laptop screen. Samsung Display is calling it the first On Cell Touch AMOLED (OCTA) screen for laptops. Per a Google translation of the South Korea-based company's announcement, these screens use embedded touch sensors, whereas other OLED laptop panels use adhesive to apply a film screen with touch sensors. According to Samsung Display, the film represents 6 to 11 percent of the screen's thickness.

Because it uses fewer materials, Samsung Display also claims its OCTA technology is more environmentally friendly than the previous design and will weigh less due to a "simplified" panel structure.

Samsung Display is clearly targeting the panels at the high end by including specs like a 16:10 aspect ratio and "3K" resolution, which we expect to mean 2880×1800. Samsung Display's announcement included an image of a 16-inch panel in a concept laptop. With a 16-inch screen and 2880×1800 pixels, the OLED screen would have a pixel density of 212.3 pixels per inch.

The screens will also support a 120 Hz refresh rate. High refresh rates already tax battery life, but with the panels also being OLED and potentially aimed at thinner laptop designs with less room for a beefy battery and cooling, battery life will be a top concern for PCs adopting OCTA.

Samsung Display said that it expects the panels to debut in Samsung's next series of Galaxy Book laptops, which will come out in February and will be "highly portable." Samsung is expected to announce its new Galaxy Book PCs at its Galaxy Unpacked Event on February 1. Beyond Samsung, the OLED screens are supposed to be appropriate for "medium-sized" laptops.

The new laptop touchscreens with OLED computing displays are finally starting to improve selection. In the laptop realm, some of the most extreme designs anticipated for 2023 include the Lenovo Yoga Book 9i, which has two 13.3-inch OLED screens; and the Asus ProArt Studiobook 16 3D OLED (H7604) and Vivobook Pro 16 3D OLED (K6604), both of which have glasses-free 3D OLED. In fact, pretty much every laptop in Asus' 2023 lineup will have an OLED option. There will also be more budget OLED options this year, and the technology is starting to make a play for gamers, with 240 Hz OLED laptop panels (also made by Samsung Display).

Samsung Display's OCTA OLED has primarily appeared in smartphones since the manufacturer debuted it in 2010. If the design proves successful in laptops, companies could gain a little more space for things like longer-lasting batteries, improved cooling for boosted production, or, dare we dream, better port selection than what is currently available in OLED laptops.

Samsung’s new touch tech enables thinner, lighter OLED laptops

NASA And DARPA Want To Use Nuclear Rockets To Blast Humans To Mars

Wed, 25 Jan 2023 16:07:24 +0000

Sending humans to Mars is a mammoth task, but nuclear thermal propulsion technology could make it easier.

NASA and its eccentric cousin DARPA are teaming up to test nuclear-powered rockets that could someday blast astronauts to Mars. If all goes to plan, we could see this tech being demonstrated in the next four to five years.

Time is the main push to develop nuclear rockets for space exploration. In theory, nuclear-powered rockets will be able to travel much faster than conventional rockets, thereby reducing the time it will take for astronauts to reach Mars.

Mars is around 480 million kilometers (300 million miles) away. With current rockets that cruise at around 39,600 kilometers (24,600 miles) per hour, the journey would take seven months or so.

In their announcement on Tuesday, NASA didn’t speculate how fast the nuclear-powered rockets could fly, although their previous nuclear-propulsion projects have suggested they could make the journey to Mars in just 45 days.

“NASA will work with our long-term partner, DARPA, to develop and demonstrate advanced nuclear thermal propulsion technology as soon as 2027. With the help of this new technology, astronauts could journey to and from deep space faster than ever – a major capability to prepare for crewed missions to Mars,” Bill Nelson, NASA Administrator, said in a statement.

The Artemis moon missions, which launched in November 2022, can be considered NASA's first step towards landing humans on Mars. Image credit: NASA/Bill Ingalls

NASA has hopes of landing humans on Mars at some time during the 2030s. Part of this plan involves the current Artemis Mission that looks to land the first woman and first person of color on the Moon by the end of this decade. By "returning" to the Moon, they hope to better understand what’s needed for human missions to Mars and beyond.

“The space domain is critical to modern commerce, scientific discovery, and national security. The ability to accomplish leap-ahead advances in space technology through the DRACO nuclear thermal rocket program will be essential for more efficiently and quickly transporting material to the Moon and eventually, people to Mars,” added Dr. Stefanie Tompkins, director of DARPA.

When it comes to future crewed missions to Mars, transit time is currently a major hurdle. The longer the journey, the higher the risk.

Along with the risk of something going wrong with the spacecraft, astronauts will be exposed to huge amounts of cosmic radiation on their travels. There is also the issue of psychological stress on the astronauts, who will be isolated in cramped and uncomfortable conditions with communication delays back home.

Some spacecraft that have ventured into deep space, such as the Voyager probes and Cassini, have used nuclear energy to fuel their journeys. Russian engineers have also hinted they are developing nuclear rockets for the purpose of missions to Mars. However, the idea of nuclear thermal rockets has been floating around since the early days of the Space Race during the Cold War.

Source

Deepfakes: faces created by AI now look more real than genuine photos

Wed, 25 Jan 2023 09:14:12 +0000

These faces may look realistic, but they were generated by a computer. NVIDIA, via thispersondoesnotexist.com

Even if you think you are good at analysing faces, research shows many people cannot reliably distinguish between photos of real faces and images that have been computer-generated. This is particularly problematic now that computer systems can create realistic-looking photos of people who don’t exist.

Recently, a fake LinkedIn profile with a computer-generated profile picture made the news because it successfully connected with US officials and other influential individuals on the networking platform, for example. Counter-intelligence experts even say that spies routinely create phantom profiles with such pictures to home in on foreign targets over social media.

These deep fakes are becoming widespread in everyday culture which means people should be more aware of how they’re being used in marketing, advertising and social media. The images are also being used for malicious purposes, such as political propaganda, espionage and information warfare.

Making them involves something called a deep neural network, a computer system that mimics the way the brain learns. This is “trained” by exposing it to increasingly large data sets of real faces.

In fact, two deep neural networks are set against each other, competing to produce the most realistic images. As a result, the end products are dubbed GAN images, where GAN stands for Generative Adversarial Networks. The process generates novel images that are statistically indistinguishable from the training images.

In our study published in iScience, we showed that a failure to distinguish these artificial faces from the real thing has implications for our online behaviour. Our research suggests the fake images may erode our trust in others and profoundly change the way we communicate online.

My colleagues and I found that people perceived GAN faces to be even more real-looking than genuine photos of actual people’s faces. While it’s not yet clear why this is, this finding does highlight recent advances in the technology used to generate artificial images.

All faces apart from one have been created by a generative adversarial network (GAN). Read to the end of article to find out which one is real. NVIDIA via thispersondoesnotexist.com, Author provided (no reuse)

And we also found an interesting link to attractiveness: faces that were rated as less attractive were also rated as more real. Less attractive faces might be considered more typical and the typical face may be used as a reference against which all faces are evaluated. Therefore, these GAN faces would look more real because they are more similar to mental templates that people have built from everyday life.

But seeing these artificial faces as authentic may also have consequences for the general levels of trust we extend to a circle of unfamiliar people — a concept known as “social trust”.

We often read too much into the faces we see, and the first impressions we form guide our social interactions. In a second experiment that formed part of our latest study, we saw that people were more likely to trust information conveyed by faces they had previously judged to be real, even if they were artificially generated.

It is not surprising that people put more trust in faces they believe to be real. But we found that trust was eroded once people were informed about the potential presence of artificial faces in online interactions. They then showed lower levels of trust, overall — independently of whether the faces were real or not.

This outcome could be regarded as useful in some ways, because it made people more suspicious in an environment where fake users may operate. From another perspective, however, it may gradually erode the very nature of how we communicate.

In general, we tend to operate on a default assumption that other people are basically truthful and trustworthy. The growth in fake profiles and other artificial online content raises the question of how much their presence and our knowledge about them can alter this “truth default” state, eventually eroding social trust.

Changing our defaults

The transition to a world where what’s real is indistinguishable from what’s not could also shift the cultural landscape from being primarily truthful to being primarily artificial and deceptive.

If we are regularly questioning the truthfulness of what we experience online, it might require us to re-deploy our mental effort from the processing of the messages themselves to the processing of the messenger’s identity. In other words, the widespread use of highly realistic, yet artificial, online content could require us to think differently – in ways we hadn’t expected to.

In psychology, we use a term called “reality monitoring” for how we correctly identify whether something is coming from the external world or from within our brains. The advance of technologies that can produce fake, yet highly realistic, faces, images and video calls means reality monitoring must be based on information other than our own judgments. It also calls for a broader discussion of whether humankind can still afford to default to truth.

It’s crucial for people to be more critical when evaluating digital faces. This can include using reverse image searches to check whether photos are genuine, being wary of social media profiles with little personal information or a large number of followers, and being aware of the potential for deepfake technology to be used for nefarious purposes.

The next frontier for this area should be improved algorithms for detecting fake digital faces. These could then be embedded in social media platforms to help us distinguish the real from the fake when it comes to new connections’ faces.

Source

With Nvidia Eye Contact, you’ll never look away from a camera again

Wed, 25 Jan 2023 06:52:33 +0000

Creepy AI-powered eyeballs maintain direct eye contact even if you avert your gaze.

Nvidia's Eye Contact feature automatically maintains eye contact with a camera for you.

Nvidia

Nvidia recently released a beta version of Eye Contact, an AI-powered software video feature that automatically maintains eye contact for you while on-camera by estimating and aligning gaze. It ships with the 1.4 version of its Broadcast app, and the company is seeking feedback on how to improve it. In some ways, the tech may be too good because it never breaks eye contact, which appears unnatural and creepy at times.

To achieve its effect, Eye Contact replaces your eyes in the video stream with software-controlled simulated eyeballs that always stare directly into the camera, even if you're looking away in real life. The fake eyes attempt to replicate your natural eye colour, and they even blink when you do.

So far, the response to Nvidia's new feature on social media has been largely negative. "I too, have always wanted streamers to maintain a terrifying level of unbroken eye contact while reading text that obviously isn't displayed inside their webcams," wrote The D-Pad on Twitter.

An Nvidia press video for Broadcast 1.4 update featuring Eye Contact.

A former TV news anchor named Scott Baker also chimed in about Nvidia Eye Contact with his analysis: "As a TV news anchor for decades ... this is not quite the right approach. To make communication effective ... you have to naturally break eye contact with the camera (just as you would in real life) fairly often. The power of eye contact in human communication is deeply researched. Locking eyes with someone for more than 7-10 seconds is intuitively regarded as creepy or uncomfortable. True across a dinner table, in front of a group, or through a camera."

This isn't the first time a company has used simulated eyeballs to maintain eye contact in video calls or video streams. In 2019, Apple introduced its "Eye Contact" feature in FaceTime that kept your peepers always glued to the camera. Like Nvidia's version of the technology, it also faced a generally negative reception upon launch.

But hey, if non-stop soul-searing eye contact is your thing, you can run Eye Contact yourself by downloading Nvidia Broadcast for free from the company's website. It requires Windows, an Nvidia RTX graphics card, and a deep desire to freak out anyone watching your video.

With Nvidia Eye Contact, you’ll never look away from a camera again

Lawsuit: Twitter stopped paying rent at headquarters after Musk took over

Wed, 25 Jan 2023 06:49:47 +0000

Twitter didn't pay $3.4M rent bills in December and January, building owner says.

Twitter is being sued for alleged nonpayment of rent by the owner of its US headquarters building in San Francisco, the latest of several lawsuits saying Twitter stopped paying bills after Elon Musk bought the company.

The lawsuit, filed Friday by SRI Nine Market Square LLC, says Twitter leases 462,855 square feet on eight floors of the 1355 Market Street building but failed to pay $3.36 million in rent due in December 2022. Twitter also failed to pay rent of $3.43 million in January 2023, according to the complaint filed in California Superior Court in San Francisco County.

SRI had a letter of credit with Twitter amounting to $3.6 million "as security for Defendant's performance of its obligations under the Lease," the lawsuit said. Because Twitter failed to pay rent, SRI drew on the letter of credit, and it is now down to $1, the complaint said. Twitter allegedly failed to replenish the letter of credit as required by the lease.

SRI said the lease also requires Twitter to increase the letter of credit by $10 million upon a transfer of control and that such a transfer occurred when Musk bought Twitter in late October. SRI said it served Twitter with a demand to increase the letter of credit by $10 million in November but that Twitter responded to the notice by claiming it was not required to comply.

The building owner's complaint seeks $3.16 million plus late fees and interest for unpaid rent, additional damages for Twitter's breach of the lease, and a declaration that Twitter is obligated to increase the letter of credit by $10 million.

Twitter is facing a similar complaint in the High Court of London. "Britain's Crown Estate, an independent commercial business that manages the property portfolio belonging to the monarchy, said on Monday it had begun court proceedings against Twitter over alleged unpaid rent on its London headquarters," Reuters reported.

Four other lawsuits allege nonpayment

The SRI complaint is the second lawsuit against Twitter filed by a landlord in the San Francisco County Superior Court. The previous lawsuit, filed by Columbia REIT-650 California LLC, said Twitter failed to pay rent of $136,000 at the 650 California Street office tower in San Francisco. The space was leased by a mobile ad company purchased by Twitter in May 2020.

The latest lawsuit isn't much of a surprise because of an earlier report stating that Twitter wasn't paying rent on the headquarters and other buildings. "To cut costs, Twitter has not paid rent for its San Francisco headquarters or any of its global offices for weeks, three people close to the company said," The New York Times reported on December 13.

Twitter was also sued last week by a consulting firm that says it was never paid the $2.19 million owed for work it did on the lawsuit that forced Musk to complete his purchase of the company. Previous lawsuits alleging Twitter failed to pay bills after Musk took over were filed by a software vendor and a private jet provider.

Twitter faces $1.5 billion in annual interest payments on the $13 billion of debt Musk used to fund his takeover, and the first payment "could be due as soon as the end of January," the Financial Times reported.

Musk said on November 4, 2022, that Twitter was losing over $4 million a day. He raised the prospect of bankruptcy but reportedly said on a podcast in late December, "We've got the expenses reasonably under control, so the company's not in the fast lane to bankruptcy anymore. It has been quite a roller coaster. It has its highs and lows, to say the least, but overall it seems to be going in a good direction."

Besides allegedly not paying bills, Musk has reduced Twitter's expenses by laying off half its staff, terminating thousands of contractors, and issuing an ultimatum that caused many employees to resign.

Lawsuit: Twitter stopped paying rent at headquarters after Musk took over

4A Games opens up Metro Exodus engine for the community to make mods

Tue, 24 Jan 2023 19:54:42 +0000

The powerful tools that produced Metro Exodus are now available to everyone on PC as series developer 4A Games opens the doors wide open to the modding community. Dubbed the Exodus Software Development Kit (SDK), the toolset is described as the "very engine where Metro was born."

"This isn’t just simple mod support – we’re giving you our full Editor as it was the day we released Metro Exodus, with the ability to create standalone content run from a basic executable," says the studio in today's announcement.

All that's required to download the Exodus SDK is a copy of the original base game or the Metro Exodus Enhanced Edition on Steam (tools section), GOG, or the Epic Games Store. Its features include a scene editor, a model viewer/editor, AI navigation tools, particle effects, terrain tools, a weather editor, camera tools for cutscene creation, and even a visual script editor for complex gameplay scenario making.

Documentation for the SDK is here while Mod.io support has also been integrated to make the sharing of created content with the community much easier. The system requirements for using the tools are these:

OS: WINDOWS 10
CPU: INTEL CORE I7-4770K OR EQUIVALENT
MEMORY: 8 GB RAM
GRAPHICS: GTX 1070 / RTX 2060 / AMD RX VEGA 56
DIRECTX: VERSION 12
STORAGE: 31 GB (IN ADDITION TO 70.1 GB FOR METRO EXODUS (77.95 GB WITH ALL DLCS) OR 79.6 GB FOR METRO EXODUS ENHANCED EDITION)

The Ukrainian studio also gave a small update on the next Metro entry, saying that it is embracing themes of the series such as "conflict, power, politics, tyranny, repression" even more as many of its personnel continue to live through the current war in Ukraine. Read the full developer update here.

4A Games opens up Metro Exodus engine for the community to make mods

Appliance makers sad that 50% of customers won’t connect smart appliances

Tue, 24 Jan 2023 19:53:04 +0000

Did users change their Wi-Fi password, or did they see the nature of IoT privacy.

This hypothetical dishwasher owner is one of a minority of smart appliance customers getting the full value of their device, including timely reminders to buy more of the company's recommended dishwasher tabs and cleaning packs.

Dani Serrano/Getty Images

Appliance makers like Whirlpool and LG just can't understand. They added Wi-Fi antennae to their latest dishwashers, ovens, and refrigerators and built apps for them—and yet only 50 percent or fewer of their owners have connected them. What gives?

The issue, according to manufacturers quoted in a Wall Street Journal report (subscription usually required), is that customers just don't know all the things a manufacturer can do if users connect the device that spins their clothes or keeps their food cold—things like "providing manufacturers with data and insights about how customers are using their products" and allowing companies to "send over-the-air updates" and "sell relevant replacement parts or subscription services."

“The challenge is that a consumer doesn’t see the true value that manufacturers see in terms of how that data can help them in the long run. So they don’t really care for spending time to just connect it,” Henry Kim, US director of LG's smart device division ThinQ, told the Journal.

LG told the Journal that fewer than half of its smart appliances—which represent 80–90 percent of its sold appliances—stay connected to the Internet. Whirlpool reported that "more than half" are connected. Wi-Fi-connected smart appliances may be connected when they're first set up, but a new Internet provider, router hardware, or Wi-Fi password could take the device offline. And a smart oven is likely to be far down the list of devices to set up again once that happens.

That means companies like Whirlpool are missing out on services revenue, which is increasingly crucial to manufacturers facing rising input costs, declining replacement purchases, and hungry shareholders. Whirlpool acquired recipe management app Yummly in 2017, and its customers can sync a Yummly Pro subscription to a smart oven so that it follows the recipe's instructions (which must, apparently, sometimes go beyond "heat to this level").

For its part, LG saw an incremental increase in water-filter sales when it tracked water volumes on connected fridges versus non-connected fridges, the company told the Journal. Both companies also suggested that new features, including safety alerts, are issued to connected customers.

Whirlpool told the Journal that customers "have the opportunity to opt in or opt out" of sharing data with the company. LG doesn't offer that option, but Kim told the Journal that "all data is anonymized."

While the manufacturers blame technical constraints, some customers may simply not want to provide companies with vague privacy policies or bad histories with security access to their networks.

LG smart TVs were found in 2013 to be uploading extensive data to their servers about all the activity happening on them, including watching files on USB sticks. At the time, LG admitted it was collecting this data, but it suggested the data was "not personal" and was used only for advertisement targeting or as part of software projects that were discontinued. LG is far from the only TV maker to participate in automated content recognition, but it's one of a select few that also makes a dishwasher.

More broadly, smart home (or Internet of Things, or IoT) devices are too often built with an "acquire, upload, whatever" mindset. Take the test models from iRobot/Roomba (up for potential acquisition by Amazon) that uploaded images of someone on the toilet to the cloud. Or any of the dozens of devices detailed in an Institute of Electrical and Electronics Engineers study, a Northeastern/Imperial College survey, or the Mozilla Foundation's "Privacy Not Included" list. The problems are so widespread and varied that the White House has called for universal IoT security labeling.

Appliance makers are eager for buyers to connect their smart devices, but at least some may think they've done the smart thing by letting them work offline.

Appliance makers sad that 50% of customers won’t connect smart appliances

ChatGPT is 'not particularly innovative,' and 'nothing revolutionary', says Meta's chief AI scientist

Tue, 24 Jan 2023 18:32:45 +0000

The public perceives OpenAI's ChatGPT as revolutionary, but the same techniques are being used and the same kind of work is going on at many research labs, says the deep learning pioneer.

Much ink has been spilled of late about the tremendous promise of OpenAI's ChatGPT program for generating natural-language utterances in response to human prompts.

The program strikes many people as so fresh and intriguing that ChatGPT must be unique in the universe.

Scholars of AI beg to differ.

"In terms of underlying techniques, ChatGPT is not particularly innovative," said Yann LeCun, Meta's chief AI scientist, in a small gathering of press and executives on Zoom last week.

"It's nothing revolutionary, although that's the way it's perceived in the public," said LeCun. "It's just that, you know, it's well put together, it's nicely done."

Such data-driven AI systems have been built in the past by many companies and research labs, said LeCun. The idea of OpenAI being alone in its type of work is inaccurate, he said.

"OpenAI is not particularly an advance compared to the other labs, at all," said LeCun.

"It's not only just Google and Meta, but there are half a dozen startups that basically have very similar technology to it," added LeCun. "I don't want to say it's not rocket science, but it's really shared, there's no secret behind it, if you will."

LeCun noted the many ways in which ChatGPT, and the program upon which it builds, OpenAI's GPT-3, is composed of multiple pieces of technology developed over many years by many parties.

"You have to realize, ChatGPT uses Transformer architectures that are pre-trained in this self-supervised manner," observed LeCun. "Self-supervised-learning is something I've been advocating for a long time, even before OpenAI existed," he said.

"Transformers is a Google invention," noted LeCun, referring to the language neural net unveiled by Google in 2017, which has become the basis for a vast array of language programs, including GPT-3.

The work on such language programs goes back decades, said LeCun.

"Large language models, the first neural net language model -- at the time, it was large, by today's standards, it's tiny -- was by Yoshua Bengio, about 20 years ago," said LeCun, referring to the head of Canada's MILA institute for AI. Bengio's work on the concept of attention was later picked up by Google for the Transformer and became a pivotal element in all language models.

OpenAI's program has, moreover, made extensive use of a technique called reinforcement learning through human feedback, which gets human agents to help rank output of the machine in order to improve it, much like Google's Page Rank for the web. That approach was pioneered not at OpenAI, but at Google's DeepMind unit, he said.

"So there's a whole history of this, and this did not come out of a vacuum," said LeCun, meaning, ChatGPT.

The ChatGPT program is a case less of scientific breakthroughs than it is an instance of decent engineering, said LeCun. He compared the program to IBM's Watson computer that competed in 2011 in the game show Jeopardy!, and entrepreneur Sebastian Thrun's self-driving vehicle that won DARPA's 2005 Grand Challenge. Thrun's award-winning tech "wasn't particularly innovative in terms of the underlying science," said LeCun, "it was just very well engineered."

"That's kind of what OpenAI has done," he said. "I'm not going to criticize them for that."

LeCun was an invited speaker for an hour and half talk hosted by the Collective[i] Forecast, an online, interactive discussion series that is organized by Collective[i], which bills itself as "an AI platform designed to optimize B2B sales."

LeCun made his remarks about OpenAI in response to a question during the colloquium posed by New York Times journalist Cade Metz. Metz asked if Meta's AI team, FAIR, which LeCun built, will ever be identified in the public mind with breakthroughs the way that OpenAI is.

"Are we going to see this from Meta? Yeah, we're going to see this," replied LeCun. "And not just text generation, but also creation aids," he said, including "generative art," which, he said, "I think is going to be a big thing."

Meta will be able to help small businesses promote themselves by automatically producing media that promote a brand, he offered.

"There's something like 12 million shops that advertise on Facebook, and most of them are mom and pop shops, and they just don't have the resources to design a new, nicely designed ad," observed LeCun. "So for them, generative art could help a lot."

At another point in the talk, LeCun observed, "You might ask the question, Why aren't there similar systems from, say, Google and Meta," referring again to ChatGPT.

"And, the answer is, Google and Meta both have a lot to lose by putting out systems that make stuff up," said LeCun with a laugh.

LeCun is the winner of the 2019 Turing Award for contributions to computer science, the equivalent of computing's Nobel Prize, along with MILA's Bengio and University of Toronto professor, and Google fellow, Geoffrey Hinton. The three helped pioneer today's deep learning era of AI.

OpenAI is funded by Microsoft, which has exclusive access to the code being produced by the startup, and which is gradually incorporating the programs into its various software offerings, including its Azure cloud service.

Source

ChatGPT talks its way through Wharton MBA, medical exams

Tue, 24 Jan 2023 17:05:55 +0000

This perhaps says more about the tests than the artificial intelligence on display

OpenAI's chat software ChatGPT, if let loose on the world, would score between a B and a B- on Wharton business school's Operations Management exam, and would approach or exceed the score needed to pass the US Medical Licensing Exam (USMLE).

While this may say more about the static, document-centric nature of testing material than the intellectual prowess of software, it's nonetheless a matter of concern and interest for educators, and just about everyone else living in the age of automation.

Academics have been fretting that assistive systems like ChatGPT and GitHub's Copilot (based on an OpenAI model called Codex) will require teachers to reevaluate how they teach and mark exams because assistive technology based on machine learning has become so capable.

In educational settings, AI advice is becoming commonplace: The Stanford Daily just reported, "a large number of students have already used ChatGPT on their final exams." An estimated 17 percent of students, based on an anonymous poll of 4,497 respondents, said they had used ChatGPT to assist in fall quarter assignments and exams, with 5 percent saying they had submitted material directly from ChatGPT with little or no editing – which is presumably an honor code violation.

Separately, Christian Terwiesch, a professor at the Wharton School of the University of Pennsylvania, and a group of medical researchers mostly affiliated with Ansible Health, decided to put ChatGPT, an arguably amoral automated advisor and factually-challenged expert system, to the test.

Both Terwiesch and the Ansible Health boffins made clear that ChatGPT has limitations and gets things wrong. Overall, they gave it middling marks but they made it clear that they expect AI assistive systems will find a place in teaching and in other sectors.

The model has, after all, been trained on countless pieces of human-made writing, and so its ability to guesstimate a satisfactory answer to a question from all that inhaled knowledge and factoids isn't unexpected.

"First, it does an amazing job at basic operations management and process analysis questions including those that are based on case studies," said Terwiesch in his paper. "Not only are the answers correct, but the explanations are excellent."

That said, he observed that ChatGPT makes simple math mistakes and fumbles advanced process analysis questions. However, the AI model is responsive to hints from people about how to improve – it can successfully correct itself when given hints from a human expert.

Human guidance has also served as a source of malicious input, as demonstrated by Microsoft's Tay chatbot and by subsequent research.
Doctor, doctor

The medical research group that wrote "Performance of ChatGPT on USMLE: Potential for AI-Assisted Medical Education Using Large Language Models" includes "ChatGPT" as a co-author.

"ChatGPT contributed to the writing of several sections of this manuscript," the biological authors state in their paper.

Other organizational affiliations of the authors include: Massachusetts General Hospital, Harvard School of Medicine, in Boston, Mass; Warren Alpert Medical School, Brown University, in Providence, Rhode Island; and Department of Medical Education at UWorld, LLC, a health e-learning firm based in Dallas, Texas.

The authors – Tiffany Kung, Morgan Cheatham, ChatGPT, Arielle Medenilla, Czarina Sillos, Lorie De Leon, Camille Elepaño, Maria Madriaga, Rimel Aggabao, Giezel Diaz-Candido, James Maningo, and Victor Tseng – came to a similar conclusion as Wharton's Terwiesch. Specifically, they found that ChatGPT performed passably – above the variable passing threshold of about 60 percent – on the USMLE exam, if given the benefit of indeterminate answers. And they expect large language models (LLMs) will play a growing role in medical education and in clinical decision making.

"ChatGPT yields moderate accuracy approaching passing performance on USMLE," the authors state in their paper. "Exam items were first encoded as open-ended questions with variable lead-in prompts. This input format simulates a free natural user query pattern. With indeterminate responses censored/included, ChatGPT accuracy for USMLE Steps 1, 2CK, and 3 was 68.0 percent/42.9 percent, 58.3 percent/51.4 percent, and 62.4 percent/55.7 percent, respectively."

Describing ChatGPT's performance as "approaching passing" is a generous way of phrasing it, particularly with the AI being given credit for indeterminate answers. Arriving in a physician's office and seeing a diploma advertising a grade of D might provoke a bit more concern among patients.

But the researchers maintain that the things ChatGPT did get right conformed closely with accepted answers and that the AI model has improved remarkably, having months earlier achieved a success rate of only about 36.7 percent.

Interestingly, they observed that ChatGPT performed better than PubMedGPT, an LLM based solely on biomedical data that managed accuracy of only about 50.8 percent (based on unpublished data).

"We speculate that domain-specific training may have created greater ambivalence in the PubMedGPT model, as it absorbs real-world text from ongoing academic discourse that tends to be inconclusive, contradictory, or highly conservative or noncommittal in its language," the authors state.

Essentially, the less scientific, more opinionated material that went into ChatGPT's training, like patient-facing disease explanation pamphlets, appears to have made ChatGPT more opinionated.

"As AI becomes increasingly proficient, it will soon become ubiquitous, transforming clinical medicine across all healthcare sectors," the authors conclude, adding that the clinicians associated with AnsibleHealth have been using ChatGPT in their workflows and have reported a 33 percent reduction in the time required to complete documentation and indirect patient care tasks.

This perhaps explains Microsoft's decision to funnel billions into OpenAI for its future software.

The utility of ChatGPT in an education setting – despite the fact that it's often wrong – was underscored in a blog post published Sunday by Thomas Rid, professor of strategic studies and the founder director of the Alperovich Institute for Cybersecurity Studies.

Rid describes a recent five-day Malware Analysis and Reverse Engineering course taught by Juan Andres Guerrero-Saade.

"Five days later I no longer had any doubt: this thing will transform higher education," said Rid. "I was one of the students. And I was blown away by what machine learning was able to do for us, in real time. And I say this as somebody who had been a hardened skeptic of the artificial intelligence hype for many years. Note that I didn’t say 'likely' transform. It will transform higher education."

Guerrero-Saade, in a Twitter thread, acknowledges that ChatGPT got things wrong but insists the tool helped students come up with better answers. He suggests that it functions like a personal teaching assistant for each student.

"Fearmongering around AI (or outsized expectations of perfect outputs) cloud the recognition of this LLMs staggering utility: as an assistant able to quickly coalesce information (right or wrong) with extreme relevance for a more discerning intelligence (the user) to work with," he wrote.

Rid argues that while concerns about AI as a mechanism for plagiarism and cheating in education need to be addressed, the more important conversation has to do with how AI tools can improve educational outcomes. ®

Source

Microsoft can't stop itself blowing billions on OpenAI

Mon, 23 Jan 2023 21:55:04 +0000

Embrace, extend ... extinguish, but whom or what this time?

Microsoft announced its latest "multi-year, multi-billion dollar" investment into OpenAI on Monday, cementing its position as the startup's exclusive cloud provider.

As far as some are concerned, the Windows giant is the de facto owner of OpenAI, having poured tons of resources into the precocious machine-learning lab so far.

"We formed our partnership with OpenAI around a shared ambition to responsibly advance cutting-edge AI research and democratize AI as a new technology platform," Microsoft boss Satya Nadella said in a canned statement.

"In this next phase of our partnership, developers and organizations across industries will have access to the best AI infrastructure, models, and toolchain with Azure to build and run their applications."

The financial and technical details of this latest collaboration were not officially disclosed, but has been rumored to be worth as much as $10 billion. The injection of support, led by Microsoft, is backed by other investors, and marks the third time Redmond has bought into OpenAI.

Microsoft began its partnership with the San-Francisco-based upstart in 2019 when the IT giant pledged $1 billion, and splashed out on a second series of funding in 2021. One year prior to this, Redmond had secured an exclusive license to incorporate OpenAI's text-generation GPT-3 system into its own tech and applications.

Now, Microsoft has negotiated a fresh agreement in the wake of various machine-learning models crafted by OpenAI, including the brains behind the automated pair-programmer tool GitHub Copilot; text-to-image model DALL·E 2; and its latest text-generation bot ChatGPT. Microsoft will also expand its artificially intelligent offerings on its Azure cloud platform using OpenAI's know-how; the duo's Azure OpenAI service was made generally available last week.

As OpenAI's exclusive cloud provider, Microosft will host and deploy the startup's models for R&D, as well as for products and API services. The latest investment is a win-win for both companies.

Redmond will gain access to cutting-edge AI technology to create products and revamp old ones, while attracting customers who want to use OpenAI's models to Azure. Meanwhile, OpenAI will get more funding and/or compute to carry out AI research and develop more products. We all know that today's state-of-the-art models require incredible amounts of data, storage, processing, and more to build, test, and deploy. OpenAI's latest beast, GPT-4, is supposed to be coming soon.

Microsoft is reportedly exploring how ChatGPT-esque responses to queries can be used to improve its web search engine Bing, and its Microsoft Office programs, such as Outlook, Powerpoint, and Word.

OpenAI CEO Sam Altman said: "Microsoft shares our values and we are excited to continue our independent research and work toward creating advanced AI that benefits everyone."

Rumor has it Microsoft will nab 75 per cent of OpenAI's profits until it recoups its portion of the investment, after which it will go on to take a 49 per cent stake into the startup. We'll see, but Microsoft will always have its due, one way or another. ®

Source

X-Plane is the fearsome future of unmanned stealth fighters

Mon, 23 Jan 2023 20:35:15 +0000

New US stealth aircraft concept uses active flow control technology rather than moving control surfaces to maneuver

The US is moving to develop unmanned aircraft with no moving control surfaces, a technological advance that promises to improve significantly the performance of both stealth aircraft and drones.

The Warzone reported that the US Defense Advanced Research Projects Agency (DARPA) and Aurora Flight Sciences have moved on to the next stage of the Control of Revolutionary Aircraft with Novel Effector (CRANE) X-Plane program.

The report notes that the project features a technology known as Active Flow Control (AFC), which maneuvers the unmanned aircraft using bursts of high-pressure air from actuators embedded in the aircraft’s body instead of traditional ailerons and rudders.

The Warzone report cites a January DARPA press release that states that it has selected Aurora Flight Sciences for Phase 2 of the CRANE program. The press release says that this follows Phase 1 of the project, which resulted in an innovative unmanned testbed aircraft using AFC to generate control forces during wind tunnel testing.

The press release said Phase 2 “will focus on detailed design and development of flight software and controls, culminating in a critical design review of an X-plane demonstrator that can fly without traditional moving flight controls on the exterior of the wings and tail.”

The press release also states that the project includes a Phase 3 option wherein DARPA intends to fly a 3,175-kilogram prototype aircraft that addresses the problems of incorporating AFC into a full-scale aircraft and reliance on the technology for controlled flight.

The Defense Post notes in an article this week that the prototype’s interchangeable wings and effectors allow it to be tested for designs from Aurora and other aerospace firms. It also says that a flight demonstration is expected in 2025, with a demonstration low subsonic flight speed of Mach 0.7.

The CRANE program is in line with the evolution of stealth aircraft. As noted by a 2002 Frontiers of Engineering Report, as combat aircraft evolve, they must maintain and improve current performance levels and become more affordable and stealthier.

DARPA’s X-Plane stealth concept is gaining altitude. Image: DARPA

The report notes that such improvements will require innovations in inlets and exhausts, which must be shorter, more compact, simpler and lighter to conform to flying wing designs of future stealth aircraft.

However, the same report notes that many of the necessary design characteristics for future stealth aircraft were impossible with the aerodynamic technology at that time two decades ago.

The defense website International Defense, Security, and Technology (IDST) notes in a March 2022 article that advances in systems integration, miniaturization, actuators, sensors and computing power have now made AFC technology feasible for military aircraft.

Moreover, the article notes that AFC is more desirable than passive measures such as traditional hinged flight control surfaces as they can be turned on and off as required – although the technology entails higher costs.

It notes that actuators, characterized as lightweight, low profile, energy efficient, durable, easy to use, scalable, and rapidly responsive, are at the heart of AFC technology. But given their importance, it also says that these parts could also represent the weak link in the technology.

DARPA mentions that AFC “enables multiple opportunities for aircraft performance improvements, such as elimination of moving control surfaces, drag reduction and high angle of attack flight, thicker wings for structural efficiency and increased fuel capacity, and simplified high-lift systems.”

Furthermore, as noted by David Cenciotti in a January article for The Aviationist, AFC technology is considered the way ahead for designing future stealth combat aircraft.

However, Cenciotti notes that moving control surfaces impacts aircraft radar cross section (RCS), which can impact overall stealth shaping by influencing how radar returns are generated, potentially compromising stealth.

Given that, Cenciotti mentions that future stealth aircraft, such as the 6th generation US Next Generation Air Dominance (NGAD), will probably feature a combination of AFC and thrust vectoring to increase maneuverability.

However, China may already have a leg up on the US. In August 2022, South China Morning Post (SCMP) reported that China had developed plasma AFC technology that China could integrate into its upcoming H-20 stealth bomber to increase its aerodynamic performance.

The report notes that the technology works by having a thin membrane on the front of a flying wing aircraft, which can detect when the speed of air flowing on top of the aircraft decreases to the point of inducing a stall.

Once it detects a potential stall, the membrane ionizes air molecules with high-voltage electricity, generating a shower of plasma that can accelerate airflow, increasing the lift coefficient of the aircraft by a third and preventing a stall even if the aircraft drops to unusually low speeds of up to 108 kilometers per hour.

As with many emerging military technologies, information is scant on AFC-related developments. Judging by NGAD and H-20 development timelines, it is plausible that AFC technology will reach maturity in the 2030s.

China’s secretive new H-20 Stealth Bomber. Credit: Handout.

In a September 2022 article in Defense News, Stephen Losey notes that NGAD is still in the design process and has not entered the engineering, manufacturing and development (EMD) stage, with the operational version of the aircraft not being ready until at least 2030.

Likewise, Brent Eastwood notes in a July 2022 article in 1945 that China’s H-20 is not likely to be ready until 2030. Eastwood mentions that the H-20 may be a copy of the US B-21 Raider, which was unveiled last December.

Given that, he says the H-20 may be on a slightly delayed timeline as the B-21, as China embarks on efforts to copy or steal US military technology to match it step-by-step.

Source

CNET Halts Publishing AI-Generated Stories Following Disclosure Controversy

Mon, 23 Jan 2023 19:59:43 +0000

CNET is a media company that has been around for more than two decades. It was founded in 1994 and is based in San Francisco, California. They have a website called CNET.com, which covers a wide range of technology topics and more. They also run a YouTube channel where they post tutorials and product reviews and also have a podcast network that focuses on technology-related topics. Recently, I discovered that CNET has put a pause on all its publications because their stories were being generated using an artificial intelligence (AI) tool. CNET only formally announced the use of AI when readers noticed a tiny disclosure. Their Editor in Chief claimed the use of the AI was not secret but quiet.

Readers noticed that original publications of the stories contained errors such as confusing abbreviated terms. You would find terms such as APR and APY and some incorrect calculations on some figures. Apparently, CNET is amongst several websites that have been publishing and writing using AI. An example of other sites is Creditcards.com and Bankrate, which have also paused any AI stories.

The unnamed AI tool was built by Red Ventures, and editors can choose domain-level sections and domains from which to get data to generate their stories. The Editors could also use AI-generated text as well as their own reporting and writing. No information was received on the data set that was used to train the AI, and no information was received concerning plagiarism. This was dismissed by saying that more information would be made available at a later stage.

The leadership of the company went on to differentiate between the unnamed AI tool and other automated technology that Red Ventures uses. These automated tools are used to insert numbers for refinance rate and mortgage rate stories. These tools have been in use for longer than the company has disclosed it. CNET is set to start disclosing its own stories about AI.

According to The Verge, the staff were clueless about what was going on in terms of the use of AI tools. Publishing content on banking and finance is something media sites love because it draws a lot of attention through search engines, and this can be converted into affiliate link profits. When a media company creates content, it's standard practice that they want the information optimized for search engines. The problem comes when a bot is used to identify and create these stories, and this totally defies the ethical editorial practice. CNET prioritized making money rather than staying timely and relevant with its news.

In response to all these accusations, CNET published a post called ‘Opens in a new window', explaining that the AI was solely intended to test if it could assist their busy staff of editors and reporters by covering topics from a wider perspective. The justification is that the technology gave them time and energy to focus on deeper reporting and analysis.

Source