Today, Reuters reports, via unnamed sources, that Amazon is thinking about charging between $5 to as much as $10 a month to access the AI-based version of Alexa.

Alexa first launched in 2014 alongside the company's first lineup of Echo smart speakers. However, this new AI-based version would be the first major revamp of the digital assistant since it went public a decade ago.

Reuters added that the new AI Alexa is being developed under the code name “Banyan" and Amazon has also referred to the new version as "Remarkable Alexa."

As far as when Amazon plans to launch the new AI Alexa, the story says that the company is pushing its development team to get it ready by a deadline of sometime in August. Amazon usually has a major press event to introduce new hardware and software items for consumers in September so that an August deadline would be in line for a product launch.

A spokesperson for Amazon did offer this vague statement to Reuters on these rumors:

We have already integrated generative AI into different components of Alexa, and are working hard on implementation at scale—in the over half a billion ambient, Alexa-enabled devices already in homes around the world—to enable even more proactive, personal, and trusted assistance for our customers,

By contrast, Microsoft charges $20 a month to access its Copilot Pro AI service, and Google charges $19.99 a month for Google One, which includes access to its Gemini Advanced AI model, among other features.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

The alleged Samsung 990 EVO Plus SSD could bridge the gap between the cheap 990 EVO model and the more expensive 990 PRO Series. A "Plus" variant of an SSD is not a new concept for Samsung. Previously, the company offered the Samsung 970 EVO Plus drive with slightly better specs than the non-plus variant. The upcoming 990 EVO Plus could provide faster speeds and higher capacities (the current 990 EVO tops at just 2TB) than its more affordable sibling, which currently costs $89.99 (1TB) and $139.99 (2TB).

Sadly, the spotted entries in KIPRIS do not contain any specific information, so one can only speculate about the details at this point.

While the 990 EVO Plus does not leave many question about what it is, the 9100 PRO is more of a mystery. Again, the KIPRIS database does not reveal much, but the 9100 PRO entry shares the same category with the 990 EVO Plus, which suggests that these two may be closely related.

With Samsung's PRO lineup now at 990, it makes sense for the company to show its next-generation PRO drive under a new name, and the 9100 PRO seems quite fitting. Also, Samsung might switch to a four-digit moniker for its first flagship PCIe Gen5 SSDs.

At this point, the only Gen 5 drive is the entry-level 990 EVO, which supports both PCIe 4.0 and PCIe 5.0. However, unlike heavy hitters like Crucial's T705 or MSI's Spatum M580 with their 14GBps+ speeds, the 990 EVO peaks at just 5GBps. The alleged Samsung 9100 PRO could finally fix that and offers Samsung SSD fans a model with up-to-date performance.

Via SamMobile

As an Amazon Associate, we earn from qualifying purchases.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

If you are a happy owner of either, beware of bricking the software of your fancy new Copilot+ PC. The thing is that Microsoft still has not published recovery images for its latest ARM-powered devices.

Microsoft provides official recovery images for each Surface computer to help users who are having trouble fixing their devices' software without sending the device to Microsoft. You can download a recovery image for your Surface from the official website, and it will come in handy in case you cannot boot your computer, it is malfunctioning, or experiencing another type of software problem.

From the official support page:

If your Surface won’t start—or if the recovery info has been removed—you can use your USB recovery drive to access recovery tools and solve problems. Your Surface comes with Windows recovery info that allows you to refresh it or reset it to its factory condition. If your Surface does turn on, see Restore or reset Surface.

For some unknown reason, two days after the sales began, official recovery images for the Surface Pro 11 and the Surface Laptop 7 are still unavailable. The most recent devices you can service using those images are the Surface Laptop 6 for Business and the Surface Pro 10 for Business. Therefore, you better not wreck your Copiltot+ PC's software; otherwise, you will be dead in the waters.

If you are curious to give Microsoft's new ARM computers a try, check out our Specs Appeal articles comparing the Surface Pro 11 with its predecessors and the Surface Laptop 7 with its predecessors.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

The city is found in the Mojave Desert of southern Nevada, and players flying over the area with the optional update will now be able to witness "The Entertainment Capital of the World" in much more detail than before. Everything from the iconic hotels and The Luxor Sky Beam (the brightest human-made light in the world), to the Last Vegas Sphere can all be seen up close with great detail. Check the trailer below for some shots of the sights in question:

"Microsoft Flight Simulator City Update VIII brings Las Vegas and its myriad facets alive like never before in the world," says the developer in the Flight Simulator blog. "With this latest offering, simmers can enjoy the sights of Las Vegas rendered in eye-popping resolution, with a fusion of the latest high-resolution aerial imagery, digital elevation model data, and TIN (triangulated irregular network) surface texturing to render the sights."

Microsoft Flight Simulator City Update 8: Las Vegas is now available as an optional download across PC (Steam and Microsoft Store), Xbox Series X|S, Xbox and PC Game Pass, as well as Xbox Cloud Gaming via Game Pass Ultimate. As it's not being offered as a mandatory update, it can be found in the in-game marketplace for free. The game version must be 1.37.19.0 or higher to install it, however.

Also, following the recent Xbox Games Showcase, the next entry in the franchise, Microsoft Flight Simulator 2024, now has a November 19 release date attached to it.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

Here is the official changelog:

The following update is available for Surface Laptop Studio devices running Windows 10 October 22 Update, Version 22H2, or greater.

 

Improvements and fixes:

 

• Addresses an issue that prevented DFCI removal via network.

• Resolves a problem that caused taskbar flickering and occasional device freezing when using Surface Slim Pen.

• Provides a seamless experience when connecting two devices via USB-C for direct peer-to-peer communication.

Here is the list of new drivers:

Here is extra information about the release:

You can update your Surface Laptop Studio to the latest firmware by heading to Windows Settings > Windows Update and clicking "Check for Updates" or downloading a manual installation package. We recommend backing up important data before installing any firmware update since they are non-uninstallable. In case something goes horribly wrong, download the Surface Laptop Studio recovery image that will help you bring the device back to life (you will need to flash it onto a USB drive).

The original Surface Laptop Studio will remain supported until October 5, 2027.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

Too good to be grue

Adventure’s impact on interactive fiction and game design has already been well-documented. So has the second big boom for interactive fiction: the glory days of Infocom. The company, composed of MIT staff and students, started in 1979, and they won many fans with their debut creation, Zork. That text adventure got a few sequels, and the company solidified its geek legacy by partnering with Douglas Adams for a retelling of A Hitchhiker's Guide to the Galaxy. For many of the key players of IF in the '80s and '90s, this was a golden age of irrevocable impact.

 

"I did play a few other kinds of fairly basic arcade-style computer games, but the text adventures felt much deeper and more engaging," said Emily Short, creative director at Failbetter Games and a pioneer in both IF and game narrative. "The legibility and purpose of the environments especially appealed to me. In the world of a text adventure, everything is there for a reason, to solve a puzzle or tell a piece of a story. In childhood, I felt like a lot of the real world was mysterious and inexplicable, and I enjoyed a game where the environment would in time be fully solved and explained."

 

Infocom occupied a unique space in the games industry of the early '80s, not just because of its genre of choice but also for its ability to connect with players. "They had a really funny newsletter, and they highlighted their game authors in a way that was unusual for the time," Daly recalled of the company. "It meant that IF always felt more personal to me than other kinds of games from that period."

 

Twine creator Chris Klimas had a similar story of falling in love before he ever booted up an Infocom game. He was wrapping up a copy of A Hitchhiker's Guide to the Galaxy as a holiday gift and became fascinated with the packaging. "It came in this folio, almost like a record, where there was a floppy disc that went into this sleeve and the instruction manual went into another pocket," he explained. "I was kinda curious about it, because I liked computer games anyway, so really stealthily, I slipped the manual out and read through it." He was excited by the manual even before falling in love with the actual game.

 

(As a side note, Klimas was the only person I spoke with who gave a nod to the Choose Your Own Adventure books as a source of inspiration, which I, too, remember fondly as my own first exposure to interactive fiction.)

 

But the Infocom heyday was shortlived. After the business was acquired by Activision in the mid-eighties and the brand was abandoned soon after, no other notable studio stepped up to carry the banner for text adventures. In spite of, or possibly because of, the short amount of time Infocom was around, their releases generated both a lot of fans and a lot of mystique. And since playing IF in those days often demanded at least some technical know-how and a lot of curiosity, Zork lovers around the world got enterprising. A few independent, unrelated efforts emerged to reverse-engineer how the company made its games.

 

One of them came from Graham Nelson while he was a student of mathematics at the University of Oxford. He created Inform, which allowed people to write their own games with the Z-machine virtual machine used by Infocom. Curses was his sprawling magnum opus to showcase Inform's potential, and many IF authors have cited it as both a watershed work and a personal inspiration. He told me via email about his motivations to build Inform:

 

I became fascinated with disassembling and deciphering the so-called story files of Infocom and wrote one of my own, with Inform simply being the tool I needed to make in order to do that. But the writing was really the point. I wrote a novel and some short stories and a play and a doctoral thesis about invariants of 3-manifolds with boundary…

 

I continue to think today that the act of concentration in writing a poem is quite like the mathematician's ideal of finding the cleanest line through a written proof. And I wrote an interactive fiction, Curses, in the same imitative spirit. To me, this was another genre of writing, not of computer science. There were game mechanics, of course, but 80s or 90s IF was more like writing poetry than programming an arcade game, even if a compiler was involved. The room descriptions alone were like sixty exercises in style, each to be done a little differently, each to consume no more than a few dozen words.

Some of the other essential developments for the early IF community were provided by Plotkin. He authored his own set of notable works, such as Spider and Web and Hadean Lands, but is equally famed in the scene for his technical contributions.

 

"The original reverse-engineered Infocom format was an excellent format, and people wrote a lot of games in it," he said. "But… it just wasn't that big. You could get in 128kB of text and code." Authors using Inform or TADS, the Text Adventure Development System, were starting to surpass the boundaries of those early formats, so Plotkin recognized a need to redesign Z-machine. The result was Glulx, a 32-bit virtual machine that resolved some of the limitations of Z-machine. He also separated the interface layer into its own application, allowing writers more flexibility and creativity in how they would present their works. "Even at that point in the '90s, I was imagining a web browser interface," he said. "I was imagining an interpreter which could be ported to Mac, Windows, Linux."

Subject: Seeking fellow Infocom fans

The place to share these tools and the games made with them was Usenet. rec.arts.int-fiction was the newsgroup for writers and craft talk, while rec.games.int-fiction was the place to discuss releases and get help as a player. It was a transformative time for IF, especially for those who were seeking fellow fans of the Infocom era.

 

"It was really fun and cool to be part of folks who remembered these games that seemed to us to have this huge cultural impact and then vanished entirely," said Interactive Fiction Technology Foundation co-founder Jason McIntosh. "So for a while, it felt like we were keeping this flame lit. And that felt very important."

 

"That period was tremendously formative for my own understanding of how to design and talk about games," Short reflected. "There was a terrific sense of creative community and shared inquiry. You might release a game and have it played only by a handful of people, but those people would often give really detailed and thoughtful feedback on everything from the implementation to the literary themes of the work. Sometimes they’d create their own pieces riffing on or responding to the ideas in yours."

Jon Ingold—narrative director and co-founder of Inkle Studios, which made 80 Days and Steve Jackson's Sorcery!—also recalled the experience of creating games and iterating ideas in a dialogue. "That ping-ponging between developers in a closed environment means that six games later, you get something really incredible," he said.

 

He also spoke about the special sense of immediacy in games during that era. "In the '90s, when a game came out, you had to play it because it might do something that would blow your mind," Ingold said. "It might completely pivot your idea of what's possible in a game."

 

While the collective memory of those Usenet days is mostly fond, the reality was never all sunshine and rainbows. Even while acknowledging the positive aspects of artistic development during that era, Short was clear about the problems. "It wasn’t a flawless environment. Usenet was unmoderated, and there was a certain amount of trolling and harassment there," she wrote. "I did have a few unpleasant experiences with that culture."

 

McIntosh didn't put so fine a point on it. "We were all assholes. We were just a bunch of young punks, and we would flame each other, and we were mean," he said.

Community via competition in IFComp

The outlook for commercial IF was sinking to a low point during the '90s as "adventure game" started to mean titles like King's Quest and Monkey Island rather than text-only. But people in the IF newsgroups were on the brink of two important events that changed the outlook for the better within their community.

 

In 1995, the writing IF newsgroup started talking about holding a competition for shorter games. From out of the rambling discussion, Kevin Wilson was the person who picked up the torch. “There had been some conversation on rec.arts.int-fiction about a competition of some sort for a while, but it hadn't really gone much of anywhere," Wilson told me via email. "I was fairly involved in the scene at the time, so I finally just kind of announced one, set some basic rules and a deadline, and off we went.”

 

The event was dubbed IFComp, and the response far surpassed Wilson's expectations. It became an annual fixture, and multiple people have since helmed the event, including Jason McIntosh. After McIntosh took over IFComp in 2014, he oversaw the event's transition to welcome hypertext entries as part of the IF canon officially. He and the event's recent leadership have emphasized that the competition be a snapshot of IF's current state. "We let the community drive what shape this wants to be," McIntosh explained. "[IFComp] is an annual community expression."

 

And it continues to play that role, entering its 29th year this summer.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

When OpenAI announced GPT-4, its latest large language model, last March, it sent shockwaves through the tech world. It was clearly more capable than anything seen before at chatting, coding, and solving all sorts of thorny problems—including school homework.

Anthropic, a rival to OpenAI, announced today that it has made its own AI advance that will upgrade chatbots and other use cases. But although the new model is the world’s best by some measures, it’s more of a step forward than a big leap.

Anthropic’s new model, called Claude 3.5 Sonnet, is an upgrade to its existing Claude 3 family of AI models. It is more adept at solving math, coding, and logic problems as measured by commonly used benchmarks. Anthropic says it is also a lot faster, better understands nuances in language, and even has a better sense of humor.

That’s no doubt useful to people trying to build apps and services on top of Anthropic’s AI models. But the company’s news is also a reminder that the world is still waiting for another AI leap forward in AI akin to that delivered by GPT-4.

Expectation has been building for OpenAI to release a sequel called GPT-5 for more than a year now, and the company’s CEO, Sam Altman, has encouraged speculation that it will deliver another revolution in AI capabilities. GPT-4 cost more than $100 million to train, and GPT-5 is widely expected to be much larger and more expensive.

Although OpenAI, Google, and other AI developers have released new models that out-do GPT-4, the world is still waiting for that next big leap. Progress in AI has lately become more incremental and more reliant on innovations in model design and training rather than brute-force scaling of model size and computation, as GPT-4 did.

Michael Gerstenhaber, head of product at Anthropic, says the company’s new Claude 3.5 Sonnet model is larger than its predecessor but draws much of its new competence from innovations in training. For example, the model was given feedback designed to improve its logical reasoning skills.

Anthropic says that Claude 3.5 Sonnet outscores the best models from OpenAI, Google, and Facebook in popular AI benchmarks including GPQA, a graduate-level test of expertise in biology, physics, and chemistry; MMLU, a test covering computer science, history, and other topics; and HumanEval, a measure of coding proficiency. The improvements are a matter of a few percentage points though.

This latest progress in AI might not be revolutionary but it is fast-paced: Anthropic only announced its previous generation of models three months ago. “If you look at the rate of change in intelligence you’ll appreciate how fast we’re moving,” Gerstenhaber says.

More than a year after GPT-4 spurred a frenzy of new investment in AI, it may be turning out to be more difficult to produce big new leaps in machine intelligence. With GPT-4 and similar models trained on huge swathes of online text, imagery, and video, it is getting more difficult to find new sources of data to feed to machine-learning algorithms. Making models substantially larger, so they have more capacity to learn, is expected to cost billions of dollars. When OpenAI announced its own recent upgrade last month, with a model that has voice and visual capabilities called GPT-4o, the focus was on a more natural and humanlike interface rather than on substantially more clever problem-solving abilities.

Gauging the rate of progress in AI using conventional benchmarks like those touted by Anthropic for Claude can be misleading. AI developers are strongly incentivized to design their creations to score highly in these benchmarks, and the data used for these standardized tests can be swept into their training data. “Benchmarks within the research community are riddled with data contamination, inconsistent rubrics and reporting, and unverified annotator expertise,” says Summer Yue, research director at Scale AI, a company that helps many AI firms train their models.

Scale is developing new ways of measuring AI smarts through its Safety, Evaluations and Alignment Lab. This involves developing tests based on data that is kept secret and vetting the expertise of those who provide feedback on a model’s capabilities.

Yue is hopeful that companies will increasingly seek to demonstrate their model’s intelligence in more meaningful ways. She says those may include “by showcasing real-world applications with measurable business impact, providing transparent performance metrics, case studies, and customer testimonials.”

Anthropic is touting such impacts for Claude 3.5 Sonnet. Gerstenhaber says that companies using the latest version have found its newfound responsiveness and problem-solving abilities beneficial. Customers include the investment firm Bridgewater Associates, which is using Claude to help with coding tasks. Some other financial firms, which Gerstenhaber declines to disclose, are using the model to provide investment advice. “The response during the early access period has been enormously positive,” he says.

It’s unclear how long the world must wait for that next big leap in AI. OpenAI has said it has started training its next big model. In the meantime, we will need to figure out new ways to measure how useful the technology really is.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

When a Finn talks to an AI helper like ChatGPT, they often get the sense that something is subtly wrong. “You really feel that this conversation is not the way that you would have a discussion in Finland,” says Peter Sarlin. For a start, Finnish people are known for a blunt approach to dialog and chatbots are usually tuned to be overly courteous. But there’s also the fact that most leading chatbots and the large language models behind them are developed in the US and trained on mostly US data. Cutting-edge AI products often come with a tonality that is essentially American.

Sarlin, founder and CEO of Helsinki-based Silo AI, one of Europe’s largest independent artificial intelligence labs, worries that in the age of ChatGPT, regional social nuances across Europe will start to disappear. As chatbots and large language models mostly derived from North American data become ubiquitous, the understanding of what normal conversation looks like “converges toward one,” he says.

That cultural flattening could have huge consequences as large language models start to power not just chatbots, but also many other digital services and products. American models might be able to output text in Finnish, but they can’t think—or appear to think—in Finnish. “These models usually have what you call a reasoning capability, and that reasoning capability should descend from data representing this region,” Sarlin says. “But if you look at most of the models out there, they are dominated by English, and dominated by English that represents North America.”

The concern isn’t just cultural, but economic. If closed source models owned by American companies come to dominate across Europe, the economic value will flow to them, Sarlin says.

The dominance of American models is driving many in Europe to talk about the concept of “AI sovereignty”: making sure that the core “digital infrastructure” behind the AI boom isn’t entirely controlled by private companies outside of the continent. Europe is investing heavily in supercomputers and AI research to try to catch up with the US and create domestic champions. But Europe’s AI challengers are starting from a long way behind. The continent lags a long way behind the US and China in the availability of capital and computing power. And it lacks big homegrown tech companies—the Microsofts, Googles, and Metas—which are vital conduits linking AI products to users.

“What is sovereignty when you don't have any champions?” says Raluca Csernatoni, a research fellow specializing in emerging technologies at Carnegie Europe, a think tank.

European preoccupations with the power of American tech aren’t new. Generation after generation of technology has been dominated by big US companies, whose products have become embedded into Europe’s social and economic infrastructure. Europe’s businesses run on Microsoft Office and Amazon Web Services, and its mobile devices rely on Apple and Google, which also run the app stores. European politics happens on WhatsApp, and its news media happens on Facebook, Instagram, and Twitter. Even the French watch Netflix.

And US tech companies operate on a different scale. Only two of the 10 most valuable public European corporations are in tech: German business software provider SAP and Dutch semiconductor equipment maker ASML. Six of the world’s 10 most valuable public corporations are US tech companies. Nvidia and Microsoft, the two largest, are each worth more than 15 times SAP.

That concentration of power is uncomfortable for European governments. It makes European companies downstream customers of the future, importing the latest services and technology in exchange for money and data sent westward across the Atlantic. And these concerns have taken on a new urgency—partly because some in Brussels perceive a growing gap in values and beliefs between Silicon Valley and the median EU citizen and their elected representatives; and partly because AI looms large in the collective imagination as the engine of the next technological revolution.

European fears of lagging in AI predate ChatGPT. In 2018, the European Commission issued an AI plan calling for “AI made in Europe” that could compete with the US and China. But beyond a desire for some kind of control over the shape of technology, the operational definition of AI sovereignty has become pretty fuzzy. “For some people, it means we need to get our act together to fight back against Big Tech,” Daniel Mügge, professor of political arithmetic at the University of Amsterdam, who studies technology policy in the EU, says. “To others, it means there’s nothing wrong with Big Tech, as long as it’s European, so let’s get cracking and make it happen.”

Those competing priorities have begun to complicate EU regulation. The bloc’s AI Act, which passed the European Parliament in March and is likely to become law this summer, has a heavy focus on regulating potential harms and privacy concerns around the technology. However, some member states, notably France, made clear during negotiations over the law that they fear regulation could shackle their emerging AI companies, which they hope will become European alternatives to OpenAI.

Speaking before last November’s UK summit on AI safety, French finance minister Bruno Le Maire said that Europe needed to “innovate before it regulates” and that the continent needed “European actors mastering AI.” The AI Act’s final text includes a commitment to making the EU “a leader in the uptake of trustworthy AI.”

“The Italians and the Germans and the French at the last minute thought: ‘Well, we need to cut European companies some slack on foundation models,’” Mügge says. “That is wrapped up in this idea that Europe needs European AI. Since then, I feel that people have realized that this is a little bit more difficult than they would like.”

Sarlin, who has been on a tour of European capitals recently, including meeting with policymakers in Brussels, says that Europe does have some of the elements it needs to compete. To be a player in AI, you have to have data, computing power, talent, and capital, he says.

Data is fairly widely available, Sarlin adds, and Europe has AI talent, although it sometimes struggles to retain it.

To marshal more computing power, the EU is investing in high-performance computing resources, building a pan-European network of high-performance computing facilities, and offering startups access to supercomputers via its “AI Factories” initiative.

Accessing the capital needed to build big AI projects and companies is also challenging, with a wide gulf between the US and everyone else. According to Stanford University’s AI Index report, private investment in US AI companies topped $67 billion in 2023, more than 35 times the amount invested in Germany or France. Research from Accel Partners shows that in 2023, the seven largest private investment rounds by US generative AI companies totaled $14 billion. The top seven in Europe totaled less than $1 billion.

Sarlin sees an even bigger gap looming for European companies once they have an AI product ready to launch. “We have a very good basis in Europe in terms of data, compute, and talent,” he says. “But that’s not enough, you need distribution. And that happens through software products, usually software platforms.”

OpenAI’s partnership with Microsoft gives its multiple distribution points, through code hosting platform GitHub, Copilot assistants built into productivity tools, Windows, and Bing. It has a new deal with Apple that provides still more ways for people to interact with its AI services. “Every single interaction gives you a better model, gives you a better product, which affects more users, which again, gives you a better model,” Sarlin says.

It’s here that the limits of Europe’s sovereignty in AI start to show. French government officials, up to and including President Emmanuel Macron, have praised the Paris-based startup Mistral AI in interviews and speeches, and hailed it as a domestic alternative to American companies. Macron, who hosted Mistral’s CEO for dinner at the Élysée Palace, has called the startup an example of “French genius.” The company was founded in April 2023 and is already nominally valued at around €5 billion ($5.4 billion). But in February, Mistral announced a partnership with Microsoft. The US company invested €16 million into the startup and will make its models available to its own customers. Mistral will train its large language models on Microsoft servers. The response in Brussels was one of dismay, as it seemed to undermine the arguments that had been made—including by Mistral itself—that avoiding stronger regulation of foundation models would allow European AI champions to flourish. Competition authorities are already investigating the deal.

Carnegie’s Csernatoni says deals like Mistral’s with Microsoft demonstrate that the kind of “pure” AI sovereignty imagined by many in Europe simply isn’t practical. “In the negotiations on the AI Act, Mistral AI was very much championed as a European technological sovereignty champion, but, of course, the discussions were much more nuanced than that, because you cannot have autarky,” she says.

“You need to balance collaborations and partnerships with [companies like] Microsoft to be profitable.”

The EU simply doesn’t have the platform companies with the scale and reach of a Microsoft or a Google. To compete in AI would mean going back a couple of generations to solve an older problem. “Marc Andreessen said in 2011, software's eating the world, software's eating traditional industries … AI is accelerating that,” Sarlin says. “Unless we ensure that we have these Big Tech, sizable software product companies in Europe, we're not going to be able to create value in Europe with AI.”

Source

However, closer to the end of 2023, SSD prices started to stabilize and even go back up, although not quite to their previous highs. That was due to the fact that there was now a glut of SSDs available, and the major manufacturers of NAND flash memory decided to cut back on their production.

Now, a report from a South Korean news outlet may indicate prices of SSDs could be going back down in the near future. According to The Chosun Daily (via PC Gamer), the four biggest NAND flash memory makers (Samsung, SK Hynix, Kioxia, and Western Digital) have started to ramp up production in their factories.

The report says that back in 2023, Samsung and SK Hynix reduced their NAND memory production to as low as 20 percent. Western Digital also cut down its NAND memory factory output to below 50 percent of its capacity last year.

In 2024, the demand for these kinds of products has now increased, especially for high-end NAND memory that can be used in AI-themed data centers. So, the major NAND memory makers are increasing production again to meet that demand.

The story states:

Samsung’s NAND flash production capacity has climbed to around 70%, while SK Hynix is ramping up production of high-capacity NAND products, such as high-capacity eSSDs. Western Digital, the third-largest NAND flash producer, increased its production utilization rate to around 90%.

All of that increased production could mean that SSD prices may be going back down to the level we say in much of 2023 very soon, perhaps as soon as the holiday shopping season. That's good news for PC consumers who will likely want to buy internal and external SSDs as gifts for themselves and others.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

There's lots and lots of game trailers out there, but many won't stick with me. That's not the fault of anyone making them. Many concepts can be difficult to convey in a mere 90 seconds to four minutes, and a game doesn't always have the luxury of relying on established information to avoid retreading itself. It's something I've come to expect in coverage, and I always caution waiting for more information before getting upset over some ill-understand nuance. Every so often however, there's a trailer that burns itself in my brain. 

That is the case with the recent reveal trailer for DOOM: The Dark Ages, which was unveiled by id Software and Bethesda Softworks during the Xbox Games Showcase 2024. If you somehow haven't watched the trailer yet, you can find it below. 

Here is what is new in the first firmware update for the Surface Pro 11:

The first update is available for Surface Pro (11th Edition) running Windows 11, Version 24H2 or greater.

 

Improvements and fixes:

 

• Improves display experience with new Windows features.

Here is the list of new drivers:

If you purchased the Surface Laptop 7, you can download a similar update:

The first update is available for Surface Laptop (7th Edition) running Windows 11, Version 24H2 or greater.

 

Improvements and fixes:

 

• Improves display experience with new Windows features.

The list of new drivers includes only one entry:

Unlike Intel and AMD-powered Surface devices, firmware updates for Surface computers with ARM processors are only available through Windows Update (recovery images for the new Surface models are not available yet). To get the latest updates, go to Settings > Windows Update > Check for Updates. Keep in mind that firmware updates for Surface devices are non-uninstallable, so back up important data before proceeding.

Firmware updates for the Surface Pro 11 and the Surface Laptop 7 are available on Windows 11 version 24H2, which was recently released for Copilot+ PCs with ARM processors. Other devices will get Windows 11 version 24H2 later this year.

If you plan to purchase a Copilot+ PC, check out the list of exclusive features you will get in our dedicated coverage.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

Nvidia is a name which most people in the PC world will know as a consumer graphics card maker. Sure enough, a company founded about 31 years ago, was mostly and is still mostly known for making GPUs to be used in the Nvidia branded graphics cards.

But despite having a huge market share in graphics cards worldwide, Nvidia never really made is so big that it could become among the top richest companies in the world. Until recently, that is.

Two years ago, Nvidia officially decided to enter the AI race, with its announcement of Hopper series of datacenter GPUs, with Nvidia H100 being the most famous of them. These GPUs offered far better AI performance than its own previous products and also products from its rivals.

Since then, Nvidia has announced many upgrades and updates to its originally announced AI graphics cards. Nvidia GH100 and Nvidia GB100 being some of them.

These AI GPUs, as one might call them, are selling like crazy worldwide. They are so overbooked that companies are expected to wait for years to come before they can get their hands on them.

So much is the demand that TSMC – which produces Nvidia GPUs, is busy setting more manufacturing units in multiple places all around Taiwan to keep up with the unprecedented demand. As the current manufacturing capacity is just not enough.

This unprecedented sales of Nvidia AI graphics cards has ensured that Nvidia has become the top company worldwide in terms of market cap.

Nvidia Becomes Richest Company In Terms Of Market Cap

As per CompaniesMarketCap, Nvidia has now become the world’s most valuable company in terms of market cap (thanks to @ghost_motley for the tip).

At the time of writing this article, Nvidia’s market cap stands at $3.343 trillion, which is slightly above the $3.314 trillion market cap of Microsoft, which was first on the list before Nvidia overtook it.

By doing this, Nvidia has left many other big brands like Google, Amazon and others behind in terms of market cap.

Just to explain how big its rise is. Four years ago, Nvidia’s market cap was $323.24 billion. Since then, Nvidia has grown 10x times to become $3.343 trillion in just four years. This growth is possibly unheard before and unprecedented.

What’s interesting is that while its rise is quick, it hasn’t happened overnight. Just 15 days ago, Bloomberg reported how Nvidia had crossed $3T in market cap.

Not to forget, it was overtaking other companies one after another recently in an apparent manner.

Another thing worth mentioning is that quarter after quarter, Nvidia is announcing record profits in recent times. So it’s not only its shares alone which are rising. Its overall profits are rising too thanks to high margins of these AI graphics cards.

Conclusion

Nvidia is all craze worldwide thanks to the AI boom. Its founder and chief Jensen Huang is followed everywhere and anywhere he goes, almost like a superhero. He gets more attention and fills up stadiums faster than many of the big superstars in the music industry.

But some commentators suggest that such a quick rise in its shares is a bubble which isn’t going to stay long. There are some who also say that even AI boom is a bubble, which may not be true, but that’s an entirely different story.

The unfortunate victims in this are the common graphics cards consumers though. In recent times, Nvidia seems to have diverted all its resources towards its money making datacenter GPUs. Leaving consumer graphics cards behind.

The outcome of this is that the consumer RTX graphics cards have become very expensive and come with cut-down VRAM and memory bus. Not to forget, cut-down versions of GPUs are being used by them in their graphics card models, making people buy expensive graphics while giving them a cheaper GPU inside them. This is especially the case when compared to earlier series of graphics cards from Nvidia.

Additionally, how big of an improvement will the future consumer RTX 5000 graphics cards offer and how much will they cost, it’s to be seen. Provided that Nvidia has any manufacturing space left to make the GPU for them.

Either way, this news of Nvidia becoming the most valuable company should make Nvidia, its shareholders and Jensen happy. They have all the reasons to be happy. After all, they have become the most valuable company in the whole world.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

What you need to know

• Western Digital is a well-known computer hardware company that produces SSDs, HDDs, and other devices. 
• Today, Western Digital revealed a next-gen WD Blue SN5000 NVMe SSD that the company says is intended for creators and professionals.
• This WD Blue SN5000 features Western Digital nCache 4.0 Technology to help the SSD copy data and files extremely fast with the help of AI.
• The SSD is available to order now and comes in 500GB, 1TB, 2TB, and 4TB capacities. 

For many, Western Digital is the go-to tech company for computer storage devices and that's not surprising given the quality and variety of its offerings. Today, Western Digital announced a brand new next-generation line of NVMe SSDs known as WD Blue SN5000. These come in 500GB, 1TB, 2TB, or 4TB capacities. This is 2x more storage capacity over the previous SSD generation. These SSDs are available for purchase right now, with a starting price of $79.99 at Western Digital.

New SSDs from a quality manufacturer

You will need an M.2 2280 slot on your computer in order to use the new WD Blue SN5000 NVMe SSD.  

(Image credit: Western Digital)

Western Digital produces hard drives and solid-state drives for various devices including laptops, desktops, and gaming handhelds (see our ROG Ally SSD upgrade guide for more info). As time goes on, we keep seeing the company produce larger capacity options to better suit our growing storage needs. But these SSDs don't just provide plenty of storage space, they're also fast and reliable, too. That's why Western Digital is found on our list of the best SSDs.

In the case of the new WD Blue SN5000 NVME PCIe Gen 4 SSD, you will need to have an M.2 2280 slot on your computer in order to accommodate it. The 4TB model can reach read speeds up to 5,500MB/s, which is a solid pace that should serve most computer users well. You should even be able to multitask with it at a smooth rate. In case you didn't know, the faster the read speed, the faster your computer will load and the more responsive it will be. This translates to a smoother user experience.

The WD Blue SN5000 also features Western Digital nCache 4.0 Technology, which allows the SSD to copy data and files extremely fast with the help of AI. The 4TB model also has 1,200TBW (Terabytes Written) endurance, this basically tells you that it will last a very long time, roughly to the point of writing 1,200TB worth of data over its life time before needing to be replaced — so that's a pretty long estimated lifespan. As creators and busy professionals know, several terabytes of data containing your work often need to be written onto an SSD as time goes on, so having a more durable device with a long lifespan is especially helpful. 

What's more, Western Digital offers a 5-year limited warranty and you can use the Western Digital Dashboard to monitor the SSDs performance on your computer. Whether you're needing to upgrade your current computer, are looking to upgrade your PC, or have another use for an SSD, the new WD Blue SN5000 is a fantastic option to go with. 

If you plan to buy a Copilot+ PC, here is everything you will get that is not available to peasants with Intel and AMD-based non-Copilot+ computers. Those two companies are also preparing their own Copilpt+ PC-ready chips in the form of Lunar Lake and Ryzen AI 300 series, but for now, all the features below are only available on the new Windows on ARM devices.

Cocreator in Paint

This feature makes boring AI image generation slightly more personal. You can start doodling on the canvas in Paint and Cocreator will harness the power of AI to complement your drawing with more details. It also applies your edits, adjusting the output in real-time.

Note that Cocreator requires signing in with a Microsoft Account.

Image Restyle and Creator

This part of the Microsoft Photos app applies AI-based filters to your photos. You can pick one of the pre-built filters or describe your own using text prompts.

Like Cocreator in Paint, Image Restyle requires a Microsoft Account.

Live Captions and Translate

Copilot+ PCs can use their Neural Processing Units to process any audio going through your computer and generate live captions or translate them from 44 languages to English (other languages are not yet supported). You can try Live Captions by enabling the feature from the Start menu and listening to any foreign-language audio.

Windows Studio Effects

These effects can help you look best during video calls or recording videos. You can apply background blur, reduce surrounding noise, use various filters, adjust lights, correct eye contact (good when reading from a teleprompter), and more. You can preview Windows Studio Effects from the Quick Settings menu in the notification area.

Recall

Recall is the only feature that is not available today. It was supposed to be the biggest thing, but its misfired launch and privacy issues forced Microsoft to delay Recall and get back to the drawing board. In a nutshell, Recall is a feature that gives your computer photographic memory - it literally snapshots everything you do on your computer so you can rewind at any time and search for things using natural language.

The first Copilot+ PCs include models from Microsoft itself (the Surface Pro 11 and Surface Laptop 7), Samsung, ASUS, Lenovo, and more.

You can learn more about the exclusive Copilot+ PC features in a post on the official Windows Blog.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

While Howard acknowledged Fallout's popularity has never been higher thanks to the Amazon Prime Video's TV show, he said Bethesda is not rushing to put out another game. He feels they don't "need to get stuff out that is different than the work we're doing in Fallout 76."

Howard understands the demand for a "new kind of 'mainline' single-player game." However, he explained such games take extensive time and care to develop. Rather than rushing production, Bethesda wants to get it right and make each new entry feels like a "meaningful moment."

We don't feel like we need to rush any of that. Right now, the Fallout TV show fills a certain niche in terms of the franchise and story-telling.

 

I totally get the desire for a new kind of 'mainline' single-player game, but those things take time. I don't think it's bad for people to miss things, as well. We just want to get it right and make sure that everything we're doing in a franchise, whether it's Elder Scrolls or Fallout or now Starfield, that those become meaningful moments for everybody who loves these franchises as much as we do.

The last mainline single-player game in the franchise was 2015's Fallout 4. While it's been nearly a decade, Howard believes it's better not to rush future installments. He said "I don't think it's bad for people to miss things."

Bethesda is currently busy updating Fallout 4 and Starfield. An expansion pack called Shattered Space for Starfield is due out later this year, while next-gen update for Fallout 4 is released in April.

Bethesda temporarily cut prices on Fallout franchise last month to help promote the TV show. While the games are selling well, they also saw a large increase in concurrent players on Steam. In the meantime, the Prime Video's TV show will return for season two.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

The new four-week series begins this Thursday. As usual, it will let gamers earn exclusive and new-to-Forza cars by completing weekly challenges, championships, and multiplayer events. Earning points will grant you the following vehicles:

• 2024 Ford Mustang Dark Horse - starting June 20
• 2023 Hyundai IONIQ 5 N - starting June 27
• 2023 Kia EV6 GT - starting June 27
• 2021 Toyota GR Yaris - starting July 4
• 2023 Porsche Taycan Cross Turismo Turbo S - starting July 11

With the latest update, Event Lab creators can make use of 17 new props for their tracks and maps:

• Highway Pillars

• Highway Short Curves

• Lane Split

• Junction

• Lane Merge

• Incline (medium & long)

• Highway Roads (4 lanes & 2 lanes)

• Short Highway Roads (4 lanes & 2 lanes)

Players are also offered a new premium car pack, which consists of five legendary vehicles: all three versions of the Time Machine from the Back to the Future trilogy, the Jurassic Park 1992 Jeep Wrangler Sagara, and the K.I.T.T. from Knight Rider. The Universal Icons pack will be available for $10 on Steam, Microsoft Store, and Xbox.

Finally, the latest Forza Horizon 5 update fixes performance and stability issues, plus patches the bug preventing tuning the gear ratio in the 2021 Volkswagen ID.4 in the stock configuration. You can learn more about the update in a blog post on the official Forza website.

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

According to Polish financial media outlet Bankier, CI Games will self-publish Lords of the Fallen 2 on consoles. However, Epic Games has secured exclusive PC distribution rights for the "entire product lifecycle" through its own distribution service, Epic Games Store.

CI Games trademarked "Death of the Fallen" last January. It is unclear whether Lords of the Fallen 2 is directly linked to this trademark. More details about the game's story and setting are expected to be revealed in the future.

The original Lords of the Fallen was a commercial success for CI Games, selling over 1 million copies within its first 10 days of release across PlayStation 5, Xbox Series X/S, and PC. The game has also been available on Xbox Game Pass since May.

But the PC port was plagued by crash and lag issues when it was first released. Due to a rocky launch, it has received mixed reviews on Steam. Then developer Hexworks has rolled out the first post-launch patch for Lords of the Fallen on PC.

Here's a short summary of the original game:

Taking place more than a thousand years after the events of the first game, Lords of the Fallen introduces an all-new adventure in a vast, interconnected world, more than five times larger than the original game. An expansive RPG experience – filled with NPC quests, compelling characters, and rich narrative – players will need to create their own hero before tackling the immersive single player campaign. They will also have the option to invite a second player to join their adventure in uninterrupted online co-op – a new feature for the franchise.

Let us know what you think about Lords of the Fallen in the comments below!

Source

Hope you enjoyed this news post.

Thank you for appreciating my time and effort posting news every single day for many years.

2023: Over 5,800 news posts | 2024 (till end of May): Nearly 2,400 news posts

The study conducted by researchers at the Icahn School of Medicine at Mount Sinai (New York, NY, USA;) aimed to assess whether rapid AI and machine learning-generated alerts, trained on diverse patient data, could reduce the need for intensive care and mortality rates. This prospective, non-randomized study involved 2,740 adult patients across four medical-surgical units at Mount Sinai Hospital, divided into two groups: one received real-time alerts on potential deterioration directly to their care teams, and the other had alerts generated but not delivered.

In the units where alerts were not delivered, patients meeting standard deterioration criteria received immediate intervention from a rapid response team. Further results from the intervention group showed that these patients were more likely to receive cardiovascular support medications, suggesting proactive measures by physicians; they also exhibited a reduced mortality rate within 30 days. The algorithm has since been implemented across all stepdown units at Mount Sinai Hospital, with a streamlined workflow. A team of intensive care physicians reviews the 15 highest-scoring patients daily, providing treatment recommendations to the attending doctors and nurses. As the algorithm is continuously retrained with data from an increasing number of patients, assessments by the intensive care team act as the benchmark for accuracy, further enhancing the algorithm's precision through reinforcement learning.

"Our research shows that real-time alerts using machine learning can substantially improve patient outcomes," said senior study author David L. Reich, MD, President of The Mount Sinai Hospital and Mount Sinai Queens, the Horace W. Goldsmith Professor of Anesthesiology, and Professor of Artificial Intelligence and Human Health at Icahn Mount Sinai. "These models are accurate and timely aids to clinical decision-making that help us bring the right team to the right patient at the right time. We think of these as ‘augmented intelligence’ tools that speed in-person clinical evaluations by our physicians and nurses and prompt the treatments that keep our patients safer. These are key steps toward the goal of becoming a learning health system."

Source

Google image search is serving users AI-generated images of celebrities in swimsuits and not indicating that the images are AI-generated. In a few instances, even when the search terms do not explicitly ask for it, Google image search is serving AI-generated images of celebrities in swimsuits, but the celebrities are made to look like underage children. If users click on these images, they are taken to AI image generation sites, and in a couple of cases the recommendation engines on these sites leads users to AI-generated nonconsensual nude images and AI-generated nude images of celebrities made to look like children.

The news is yet another example of how the tools people have used to navigate the internet for decades are overwhelmed by the flood of AI-generated content even when they are not asking for it and which almost exclusively use people’s work or likeness without consent. At times, the deluge of AI content makes it difficult for users to differentiate between what is real and what is AI-generated.

I discovered that Google image search was doing this completely by accident. Jason was reporting his story about how fan pages of Taylor Swift and other celebrities on Facebook have been taken over by images bestiality and scams and we were trying to check whether an image of Taylor Swift sticking her tongue out was AI-generated or edited. To find it, I typed the search term “taylor swift beach sunglasses tongue out,” which pulled up a couple of AI-generated images on the Google image search tab. One, pulled from a Japanese Pinterest page which included a number of other AI-generated images of Swift, looked fairly realistic. The other, pulled from an AI-image generation site called Neural Love, included a bunch of typical AI-generated deformities and was clearly fake.

Since both results were on the first page (the AI-generated image of Swift in the car is currently the third result), I wanted to check if Google image searching for celebrity names plus the term “swimsuit” or “bikini” would consistently turn up AI-generated images. Overall, I tested this method with 13 female celebrity names, including some of the most famous people on the planet and some lesser known YouTubers and Twitch Streamers, and every single Google Image search indexed AI-generated images of them even though the search term did not include “AI.”

A search for two celebrities also turned up AI-generated images of them in swimsuits, but as children, even though the search didn’t include terms related to age.

“Our automated systems aim to show the highest quality and most relevant images for the billions of queries we receive every day,” a Google spokesperson told me in an email. “Given the scale of the open web, there are cases when our ranking systems might surface relevant web content that unfortunately falls short of our quality standards. We use these instances to inform overall improvements, as we continue to prioritize efforts to prevent low-quality content – including low-quality AI-generated content – from surfacing highly in Search.”

The spokesperson said that Google Search indexes hundreds of billions of pages and images across the web, and while tools to reliably identify AI-generated images at this scale exist, they are in a nascent stage and are not infallible. The spokesperson indicated that Google could do a better job labeling AI images created with its own generator, as detailed in this company blog.

Google did not respond to specific questions about why it was surfacing AI-generated content in search results, and we can’t say for certain what is happening, but it’s easy to imagine why “automated systems” would surface these images. The images are always tagged with the celebrities’ names, and additional terms like “bikini,” “bathing suit,” or “swimsuit.” Because these images are coming from AI-image generation sites, they often also include the text prompts that created them, providing a kind of detailed annotation for what’s in the image. Finally, these AI-image generation sites are popular, and it’s easy to imagine people click on never-before-seen photographs of Taylor Swift in a bathing suit (because it’s not real), possibly driving it up higher in search results.

Clicking one of these images in Google image search results takes users to these AI-image generation sites, which presents them with dozens of similar AI-generated images. For example, when I clicked on one AI-generated image of a celebrity in a swimsuit made to look like a child, I was taken to Playground.com, an AI image generating site, which presented me with dozens more images of the same celebrity in child form, in a bathing suits, as well as dozens of other AI-generated images of children in bathing suits. A few of the images were titled “[name of celebrity] on Epstein Island in bikini,” referring to Little Saint James, a private island owned by Jeffery Epstein where he was accused of trafficking and sexually assaulting children.

Clicking on another AI-generated image of a celebrity made to look like a child in a swimsuit leads users to a site called sdxlturbo.ai. Under that image, are a number of other images of child celebrities in swimsuits, and a few AI-generated images of topless celebrities made to look like children. The prompts that generated one of these images is: “[name of celebrity] at 8 years old, very muscular abs, showing her belly, bathing.”

Given how the recommendation engine works on these AI image generation sites, clicking on one of these images sends users down a rabbit hole with more images of the same kind. The results are also another reminder that AI-generated CSAM isn’t produced and shared just on the dark web or other hard to access corners of the internet, but easily accessible on AI tools and platforms, despite these platforms’ policies, and are just a couple of clicks away from a Google search.

The AI-generated images of celebrities that appear in Google search results tend to come from a handful of AI image generation sites: Lexica, Neural Love, Prompt Hunt, Night Cafe, and also Deviantart, which doesn’t have an image generation tool but where people host AI-generated images. The content policies for these sites, in the instances I could find them, vary, but generally prohibit nonconsensual pornography and CSAM. That doesn’t mean the policies are enforced perfectly.

Playground, for example, which currently hosts several AI-generated images of celebrities as children in bathing suit on “Epstein island,” has a content policy that lists “content depicting children in a way that encourages or promotes attraction or sexualization of them” as “unacceptable use.”

“If you have any doubts about content in this category, you should refrain,” Playground’s content policy page says. "A perfect brightline does not exist so for anything in the gray area, we will moderate conservatively. As some guidance: young girl/boy/teen in their bikini/bra/underwear is problematic (it’s even worse if you negative prompt the clothing); anything ‘bulging’ is unlikely to pass muster; and prompting ‘eighteen’ does not necessarily make it acceptable.”

I could not find a content policy page sdxlturbo.ai, which currently hosts AI-generated images nude celebrities made to look like children, or Lexica, where a quick search turned up an AI-generated nonconsensual nude of a celebrity.  

Overall, I saw Google pull AI-generated image of celebrities from 11 different AI-image generation sites. Lexica and sdxlturbo.ai did not respond to a request for comment.

The Google spokesperson said that in addition to its legal obligations, Google has product policies that dictate when it will remove content from search results. Google says it blocks search results that lead to child sexual abuse imagery or material that appears to victimize, endanger, or otherwise exploit children. The Google spokesperson said that the company isn’t able to comment on any actions taken against specific sites, but I have not noticed any of the sites mentioned in this story being blocked from search results after I flagged them to Google.

404 Media has spent the last year reporting on the way AI, and specifically AI-generated images, have flooded the internet. Jason has written multiple articles on how AI-generated slop has zombified Facebook, and I have focused on a few image generation sites like Mage.Space, Tensor.Art (which also shows up in Google image search results), and most importantly Civitai, which is a crucial resource for the AI image generation scene because it also hosts custom image generation models. However, these are just a fraction of a rapidly growing industry of AI-image generation sites that are flooding and overwhelming the internet. We’ve seen how these AI-generated image have completely changed Facebook, and because adult content is often a good early indicator for where the internet is going, it’s easy to imagine how AI-generated images can change Google image search as we know it as well.

Source

Scientists have developed an imaging chip that could equip future smartphones with "Superman-inspired" X-ray vision — albeit operating within a much more limited range than the caped Kryptonian superhero.

The experimental chip consists of an array of three sensor pixels that emit and receive high-frequency radio signals in the millimeter-wave (mmWave) band of the electromagnetic spectrum. Signals reflected back from the target object are then amplified and mixed by onboard components, enabling outlines of the object to be viewed on a display.

In tests, the chip could detect an object behind cardboard at a distance of around 0.39 inches (1 centimeter). The researchers published their findings Jan. 5 in the journal IEEE Transactions on Terahertz Science and Technology.

It took 15 years of work — and an improvement in pixel performance by 100 million times — to make the chip small enough to fit in a mobile device, the researchers said in a statement. In the future, smartphones equipped with the chip may be capable of detecting the contents of envelopes or packages, or could be used to find studs, wires or cracked pipes behind walls.

Related: New display tech paves the way for 'most realistic' holograms in regular eyeglasses

"We designed the chip without lenses or optics so that it could fit into a mobile device. The pixels, which create images by detecting signals reflected from a target object, have the shape of a 0.5-mm square, about the size of a grain of sand," said co-author of the paper Wooyeol Choi, assistant professor of electrical engineering at Seoul National University, in the statement.

Beyond making it possible to peer through walls and inside envelopes, the new imaging tech may find applications in medicine and healthcare, the researchers said. They likened the technology to that already used in passenger scanners found at airports — though they noted that their imager chip does not use microwaves.

Instead, the imager chip uses 300 GHz signals in the millimeter-wave (mmWave) band of the electromagnetic spectrum. These sit between the microwave and infrared bands and are considered safe for humans.

"This technology is like Superman’s X-ray vision. Of course, we use signals at 200 gigahertz to 400 gigahertz instead of X-rays, which can be harmful,” said Kenneth O, professor of electrical engineering at the University of Texas at Dallas, in the statement

Also unlike Superman’s X-ray vision, the imaging tech used in the chip can only be used at very close range — approximately 1 inch (2.5 cm) away from the object being scanned. This means that thieves wouldn’t be able to covertly scan the contents of your bag or pockets without your knowledge, the researchers said.

The next iteration of the chip will be designed to scan objects from a little further away — up to 5 inches (12.7 cm) — making it better at capturing small objects.

Source

Aaron Damigos’ inbox was hit with a dreaded, ubiquitous business-update calendar invite on June 3. The meeting included someone from HR, his manager, and upper management—and ultimately resulted in the sudden end to his job as a web support associate with Microsoft.

Microsoft reportedly laid off some 1,000 people in early June, pulling from its mixed reality and Azure cloud departments, and also Damigos’ consumer sales division. An email to employees from Jason Zander, executive vice president of strategic missions and technologies at Microsoft, leaked to Business Insider, blamed a pivot to invest in artificial intelligence: “Our clear focus as a company is to define the AI wave and empower all our customers to succeed in the adoption of this transformative technology. Along the way, we make decisions that align with our long-term vision and strategy while ensuring the sustainability and growth of Microsoft.”

Damigos, who lives in Tacoma, Washington, says he wasn’t told that a push for AI directly led to the end of his job specifically, which involved helping customers understand how to use Microsoft products. But it’s clear that Microsoft, the largest backer of OpenAI, is all in on the tech. “I think the shift toward AI work has unfortunately led to the deprioritization of some essential customer-facing roles,” says Damigos, who has been chronicling his layoff journey and showcasing his skills on TikTok. “I helped people understand how to effectively use and understand the products.” He adds that he felt his team had “a lot of potential” to make a better customer experience for Microsoft—but ultimately, the company moved to make investments elsewhere.

Microsoft did not confirm the authenticity of Zander’s emails. “Organizational and workforce adjustments are a necessary and regular part of managing our business,” says Craig Cincotta, a Microsoft spokesperson. “We will continue to prioritize and invest in strategic growth areas for our future and in support of our customers and partners.”

No one knows yet quite how AI will impact work in the long term, but many experts largely agree that AI will not replace most workers anytime soon. “AI will reshape the labor market,” says Nick Bunker, director of economic research at the job board Indeed. “It’s just not clear how it will reshape it.” Some predict that it will create more jobs—but some workers are currently training their own AI replacements. But the layoffs happening now show that AI hype, not futuristic AI colleagues, can cause thousands of people to lose their jobs.

Microsoft isn’t alone. Dropbox announced 500 layoffs in April 2023, and CEO Drew Houston acknowledged that AI had played a role. “In an ideal world, we’d simply shift people from one team to another. And we’ve done that wherever possible,” Houston’s statement said.

“However, our next stage of growth requires a different mix of skill sets, particularly in AI and early-stage product development.” Meta’s Mark Zuckerberg made similar statements about cutting jobs to invest in AI earlier this year, saying in a post in February: “A major goal will be building the most popular and most advanced AI products and services,” as Meta left its “year of efficiency,” which resulted in downsizing the company. Google, too, has funneled money into its Anthropic AI developments, and its CEO, Sindar Pichai, warned of continuous cuts throughout 2024, which began in January. That comes despite Google reporting strong growth. “We’re responsibly investing in our company's biggest priorities and the significant opportunities ahead,” says Bailey Tomson, a Google spokesperson. In 2023 and 2024, several Google teams “made changes to become more efficient and work better,” Tomson says. “Through this, we’re simplifying our structures to give employees more opportunity to work on our most innovative and important advances and our biggest company priorities, while reducing bureaucracy and layers.”

This pivot-to-AI narrative echoes past moves by tech companies, like outsourcing workers, which led to poor working conditions for some contracted workers in other countries. “It feels less like there’s a real connection between investment in AI and trade-offs having to be made in other parts of the workforce, but really that this is a narrative shift being used to package a shift that predates the move to AI,” says Parul Koul, president of the Alphabet Workers Union-CWA, which represents some employees from the companies affected by recent layoffs. But because workers don’t receive a lot of transparency about whether their layoffs are tied to AI, it’s still hard to tie some job cuts directly to the tech, Koul says.

The next step, naturally, would be to see the AI that these companies invest in further disrupt their own workplaces. But for now, that doesn’t seem to be happening. AI-fueled layoffs are making up a small portion of job cuts across industries. More than 5,000 jobs were cut between May 2023 and April 2024 where companies cited AI as the reason—but this was either due to companies shifting focus to developing AI tech or because they used AI tools to take over tasks and roles, according to a report from outplacement services firm Challenger, Gray, and Christmas.

In the tech world alone, there have been nearly 100,000 layoffs in 2024, according to Layoffs.fyi, a site that tracks job cuts in the tech industry. Still, specific types of jobs are beginning to bounce back. Openings for AI roles or those that require AI skills made up 12 percent of all tech job offerings in May—the largest percentage in six years—according to CompTIA, a nonprofit trade association for the US IT industry. But AI doesn’t exist in a silo, says Tim Herbert, CompTIA’s chief research officer, and its adoption will likely create adjacent jobs needed to support the new technology. “AI will probably spur investment in other areas,” he says.

The AI reshuffle may not be the great AI takeover, but if AI is the next big opportunity for companies like Alphabet, the lack of efforts to up-skill and train employees to work in those divisions is troubling, Koul says. “There are ways in which the existing workforce can be kept whole or treated with dignity and respect through this process,” Koul says. “A lot of my coworkers, a lot of our union members, work here because they are mission-driven, they believe in the utility of the products they are working on. More opportunities for retraining and moving people to other divisions would be very welcome.”

Source