Researchers from Microsoft have introduced Kosmos-1, a multimodal model that can reportedly analyze images for content, solve visual puzzles, perform visual text recognition, pass visual IQ tests, and understand natural language instructions. From a report: The researchers believe multimodal AI -- which integrates different modes of input such as text, audio, images, and video -- is a key step to building artificial general intelligence (AGI) that can perform general tasks at the level of a human. "Being a basic part of intelligence, multimodal perception is a necessity to achieve artificial general intelligence, in terms of knowledge acquisition and grounding to the real world," the researchers write in their academic paper, Language Is Not All You Need: Aligning Perception with Language Models.

Visual examples from the Kosmos-1 paper show the model analyzing images and answering questions about them, reading text from an image, writing captions for images, and taking a visual IQ test with 22â"26 percent accuracy. [...] In this case, Kosmos-1 appears to be purely a Microsoft project, without OpenAI's involvement. The researchers call their creation a "multimodal large language model" (MLLM) because its roots lie in natural language processing, like a text-only LLM, such as ChatGPT. And it shows: For Kosmos-1 to accept image input, the researchers must first translate the image into a special series of tokens (basically text) that the LLM can understand.

San Diego Union-Tribune: Almost three years ago, the city of San Diego cut off access to its broad network of Smart Streetlights -- more than 3,000 devices perched atop light poles that could collect images and other data, some of which the Police Department used to solve criminal cases. The city removed that access, at least without a warrant, because of concerns from the public about surveillance and privacy issues. On Wednesday, the San Diego Police Department said it wants access to 500 of those devices to be restored -- and they want to add another crime-solving tool to the network: automated license plate readers.

The controversy surrounding the Smart Streetlights began in 2019 when it was revealed that the cameras had been installed without public input. Police started accessing the camera footage in 2018 for investigations. Direct access was cut off in 2020 as a result of public outcry. Because the Smart Streetlight cameras had not been well maintained over the years, the city would need to install new cameras. Adding the license plate reader technology would mark the first time the city of San Diego would have the readers in fixed locations. This is the first big push for surveillance technology in San Diego since the city approved ordinances last year specifically setting rules to govern this kind of technology in light of privacy concerns.

Reddit is fighting a film-industry attempt to identify users who discussed piracy, telling a federal court that the studios' request for users' real names should be rejected and that one of the studios' arguments is "nonsense." From a report: "Courts have long recognized that the First Amendment protects online anonymity and have established a stringent standard to use in precisely this scenario, where a litigant seeks to unmask users for the purpose of providing evidence in litigation that does not involve those users... Plaintiffs are far from meeting that strict standard here," Reddit said Tuesday in a filing in US District Court for the Northern District of California.

Reddit has no involvement in the lawsuit that triggered the request for users' identities -- the studios behind films such as Hellboy, Rambo V: Last Blood, Tesla, and The Hitman's Bodyguard sued cable broadband provider RCN in a different court, alleging that RCN failed to terminate Internet subscribers who illegally downloaded copyrighted movies. (RCN is now known as Astound Broadband after being combined with several other cable ISPs in the same ownership group.) In an attempt to prove that RCN turned a blind eye to users downloading copyrighted movies, the film studios subpoenaed Reddit seeking identifying information for specific users who commented in piracy-related threads. After Reddit provided information on only one user, calling the other requests a "fishing expedition," the studios filed a motion to compel Reddit to respond to the subpoena.

There may be a new ringtone in your life -- the urgent chime of a company-issued cell phone. From a report: In a throwback to the Blackberry era, telecom-service providers are seeing strong growth from companies handing out phones to employees. The phenomenon, which started during the pandemic, picked up recently thanks to new compliance policies around the use of WhatsApp and TikTok. It's provided a "tailwind" for subscriber gains at AT&T, Chief Financial Officer Pascal Desroches said at a conference this week. At the same event, T-Mobile US Inc. Chief Financial Officer Peter Osvaldik said his company's corporate customer count "grew every quarter in 2022."

The phones are more than just a corporate perk, said Gartner analyst Lisa Pierce. "It's also about control" -- a means of restricting or blocking applications and keeping corporate data secure, she said. Businesses, especially those in finance, have grown concerned about the security of their data, and the Securities and Exchange Commission and the Commodity Futures Trading Commission have stepped up their scrutiny over unauthorized private communication on applications such as WhatsApp and through personal email. Late last year, Congress, along with several states, banned China-owned TikTok from government employees' devices over national security concerns. This puts organizations in the position of either requiring their workers to remove apps from personal phones, or offering a secure second device. That second device helps explain how wireless carriers keep racking up millions of new subscribers long after the time when the mobile market passed saturation, with nearly every adult in the US owning at least one phone.

Tesla announced plans to open a gigafactory in Mexico as it pushes to expand its global output. Reuters reports: Mexico on Tuesday said Tesla had chosen the northern border state of Nuevo Leon for a new factory worth more than $5 billion, calling it the "the biggest electric vehicle plant in the world." Tesla will ramp up output at all of its gigafactories, Musk said at an Investor Day event Austin, Texas. The company laid out plans to cut assembly costs by half in future generations of cars and discussed its innovation in managing operations from manufacturing to service.

The plant near the city of Monterrey "will be supplemental to the output of all the other factories," Musk said at the end of the more than 3-hour long presentation, calling it "probably the most significant announcement of the day." Tesla's global capacity was 2 million vehicles a year, Tom Zhu, the new global production chief, said at the event. Musk did not provide details of how many vehicles the factory would produce in a year and Tesla's investment in it.

An anonymous reader quotes a report from ZDNet: Now, leaders from the GNOME Foundation and KDE Foundation, along with Debian Linux, are considering [...] building an app store on top of Flatpak, a universal Linux software deployment and package management program. This idea of replacing traditional but not very friendly ways of delivering Linux desktop apps, such as DEB and RPM package management systems, has been around for a while. Besides being easier to use, Flatpak and its rivals, such as Appimage and Snaps, can also run on any Linux distribution. All the programs do this by containerizing applications with all their necessary libraries and associated files.

Now, as laid out in former Google chairman Eric Schmidt's Plaintext Group, the proposal is to "Promote diversity and sustainability in the Linux desktop community by adding payments, donations, and subscriptions to the Flathub app store." Behind this idea are several Linux desktop leaders, such as GNOME president Robert McQueen; former GNOME executive director and Debian project leader Neil McGovern; and KDE president Aleix Pol. Flatpak, unlike the earlier store attempts, works on essentially all Linux distros. This makes it much more interesting.

Why Flakpak, instead of its chief rival, Snaps? They explained, "Flathub is a vendor-neutral service for Linux application developers to build and publish their applications directly to their end users. A healthy application ecosystem is essential for the success of the open-source software desktop, so end-users can trust and control their data and development platforms on the device in front of them." [...] Be that as it may, while the proposal for a paid Flathub app store remains just an idea, it's still one that may garner support. If this plan can generate enough support, and then the revenue, to cover its costs, it may create the first popular universal Linux app store. Then, who knows, maybe the Linux desktop will finally become broadly popular. Stranger things have happened.

The Biden administration is promising to hold software developers and critical infrastructure to tougher security standards and apply more pressure on ransomware gangs as part of its first national cybersecurity strategy, released Thursday. From a report: The nearly 40-page document provides a roadmap for new laws and regulations over the next few years aimed at helping the United States prepare for and fight emerging cyber threats. The strategy -- which was crafted by the two-year-old Office of the National Cyber Director (ONCD) -- has five "pillars": defend critical infrastructure; disrupt and dismantle threat actors; shape market forces to drive security and resilience; invest in a resilient future; and forge international partnerships.

The strategy includes a wide range of tasks, from modernizing federal systems' cybersecurity defenses to increasing offensive hacking capabilities in the intelligence community. The administration will start working with Congress and the private sector on legislation that would hold software makers liable for security flaws if they fail to follow security best practices, like those developed by the National Institute of Standards and Technology.

If an artificial intelligence machine can be named as an inventor for a patent, pet cats could be next, lawyers said at the UK's top court arguing only humans can be inventors in law. From a report: The UK's Supreme Court will decide whether an AI machine can be named as an inventor and who may own the patent. Imagination Engines founder, Stephen Thaler, challenged the rejection of his patent applications naming his AI machine as inventor for a beverage container and a flashing light. Allowing an AI machine to be named as the inventor can open doors to "plainly ridiculous assertion," Stuart Baran, a lawyer for the patent office, said in documents prepared for the case. Should the judges rule in favor of Thaler inventors could include "my cat Felix" or "cosmic forces," he said. Thaler tried registering the patent naming his system, DABUS, as inventor in several countries but was successful only in Australia and South Africa, according to the court documents.

China has a "stunning lead" in 37 out of 44 critical and emerging technologies as Western democracies lose a global competition for research output, a security think tank said on Thursday after tracking defence, space, energy and biotechnology. From a report: The Australian Strategic Policy Institute (ASPI) said its study showed that, in some fields, all of the world's top 10 research institutions are based in China. The study, funded by the United States State Department, found the United States was often second-ranked, although it led global research in high-performance computing, quantum computing, small satellites and vaccines. "Western democracies are losing the global technological competition, including the race for scientific and research breakthroughs," the report said, urging greater research investment by governments.

China had established a "stunning lead in high-impact research" under government programs. The report called for democratic nations to collaborate more often to create secure supply chains and "rapidly pursue a strategic critical technology step-up." ASPI tracked the most-cited scientific papers, which it said are the most likely to result in patents. China's surprise breakthrough in hypersonic missiles in 2021 would have been identified earlier if China's strong research had been detected, it said. "Over the past five years, China generated 48.49% of the world's high-impact research papers into advanced aircraft engines, including hypersonics, and it hosts seven of the world's top 10 research institutions," it said.

An anonymous reader quotes a report from CBS News: The U.S. Postal Service's plans for a nationwide fleet of electric vehicles are getting closer to being realized. The service awarded contracts on Tuesday for 9,250 battery electric vehicles and for more than 14,000 charging stations. The vehicles purchased are Ford E-Transit Battery Electric Vehicles (BEVs), which according to USPS are "100 percent electric." It's part of the agency's plans, announced in December, to make 75% of its newly acquired vehicles, known as Next Generation Delivery Vehicles, over the next five years electric. After 2026, NGDV purchases will be 100% electric, the agency said. The goal is to have a fleet of 66,000 electric vehicles deployed by 2028.

Three suppliers were awarded contracts for more than 14,000 charging stations, as well, USPS said, to kick off its Electric Vehicle Supply Equipment (EVSE) inventory. However, the agency did note that the contracts they just awarded will not provide an immediate supply. The vehicles won't be delivered until December, assuming that the agency successfully finishes its Supplemental Environmental Impact Statement and other necessary tasks. It also remains unclear where the vehicles and charging stations will be placed, as those details have yet to be finalized, the agency said.

As such, a contract has also been awarded for the agency to acquire 9,250 commercial-off-the-shelf internal combustion engine vehicles "to fill the urgent need for vehicles." In December, the agency said that 21,000 COTS vehicles will be purchased and are "expected to be battery electric," but said that depends "on market availability and operational feasibility." In this case, the internal combustion engine vehicles will be gas-powered and made by Fiat Chrysler Automobile, a spokesperson for USPS told CBS News. They added that, unlike older USPS vehicles, these will "feature air conditioning and advanced safety technology and are more suited to modern day operational requirements." "We have an urgent need to replace some of our vehicles as soon as possible, and in those instances we will look to obtain vehicles that can be provided to us expeditiously, recognizing that there are a limited amount of BEV options currently available and that the charging infrastructure buildout will also take some time," they said in an email.

"...Today there remain routes and applications which do not support BEVs. As BEV technology matures and capabilities increase, the Postal Service will continue to review its ability to utilize and expand BEV usage."

dcblogs writes: Despite the demand of employers like Apple, Amazon, Microsoft, AT&T and others, nearly 40% of software engineers preferred only remote roles, and if their employers mandated a return to the office, 21% indicated they would quit immediately, while another 49% said they would start looking for another job, according to Hired's 2023 State of Software Engineers. This report gathered its data from 68,500 software engineering candidates and a survey of more than 1,300 software engineers and 120 talent professionals. Employers open to remote workers "are able to get better-quality talent that's a better fit for the organization," said Josh Brenner, CEO of Hired, a job-matching platform for technology jobs.

TikTok has announced a batch of new features intended to reduce screen time and improve the well-being of its younger users. The Verge reports: In the coming weeks, a daily screen time limit of 60 minutes will be automatically applied to every TikTok user under 18 years old. Teens that hit this limit will be asked to enter a passcode to continue watching. They can disable the feature entirely, but if they do so and spend more than 100 minutes on TikTok a day, they'll be prompted to set a new limit.

TikTok claims these prompts increased the use of its screen time management tools by 234 percent during the feature's first month of testing. Teens will also be sent an inbox notification each week that recaps their screen time, allowing younger users to be aware of how much time they spend on the app and requiring that they make active decisions to extend the recommended screen time. These weekly updates are available now, alongside prompts to encourage teens to use screen time tools.

TikTok says it consulted current academic research and experts from the Digital Wellness Lab at Boston Children's Hospital when deciding how long the time restriction should be. "While there's no collectively-endorsed position on how much screen time is 'too much', or even the impact of screen time more broadly, we recognize that teens typically require extra support as they start to explore the online world independently," said Cormac Keenan, Head of Trust and Safety at TikTok, in a statement.

OpenAI is changing the terms of its API developer policy, aiming to address developer -- and user -- criticism. From a report: Starting today, OpenAI says that it won't use any data submitted through its API for "service improvements," including AI model training, unless a customer or organization opts in. In addition, the company is implementing a 30-day data retention policy for API users with options for stricter retention "depending on user needs," and simplifying its terms and data ownership to make it clear that users own the input and output of the models. Greg Brockman, the president and chairman of OpenAI, asserts that some of these changes aren't changes necessarily -- it's always been the case that OpenAI API users own input and output data, whether text, images or otherwise. But the emerging legal challenges around generative AI and customer feedback prompted a rewriting of the terms of service, he says.

What happens when a newspaper dies? Apparently, in some cases, its digital ghost lives on in mysterious, unrecognizable forms. From a report: Minneapolis neighborhood newspaper the Southwest Journal shuttered at the end of 2020, but its web domain continues to post fresh content under the auspices of a Delaware "SEO company" whose leader lives in Serbia. Though the site still includes a few legacy Journal articles now under fictitious bylines, all of the most recent posts are more or less junk content evidently designed to manipulate search engines. There's a Feb. 10 article about handling raw chicken. Another article highlights the "10 most popular bitcoin casino games."

While there is a recent article on creating "a breathtaking rock garden" written from the perspective of someone purportedly living in the East Harriet neighborhood, the site's content, generally speaking, is no longer in line with the Journal's longstanding coverage of South Minneapolis neighborhoods. The "Contact Us" link at the bottom of the site pointed to an email address connected to an entity known as Shantel LLC. According to its own website, Shantel LLC is an "SEO company" from Delaware, and, as of Feb. 17, its homepage read, "Let's make the internet a great again!" The company said it specializes in "writing services, SEO optimization services, and similar SEO-related services." (Shantel LLC's website was utterly emptied of content around the time this article published, but archived versions of the site include that same company description.)

Shantel's apparent CEO and founder is Nebojsa Vujinovic, a businessman living in Belgrade, Serbia, per his LinkedIn profile. When I reached out to Vujinovic via LinkedIn on Feb. 10, he said he had only owned the Journal's domain for a matter of days. He confirmed that he uses a mix of artificial intelligence and human writers to create new content on the sites he owns. As he puts it: "AI + human correction." [...] The Southwest Journal isn't the only site under Vujinovic's ownership. Several other former news sites have begun listing a Shantel LLC email address as a primary contact. That includes the Missoula Independent, which was at one time the largest weekly paper in Montana, according to archived versions of the website. News conglomerate and former owner Lee Enterprises shut down the Independent in 2018. Like the Southwest Journal's website, the Independent's site now includes a few legacy articles on local politics and culture, but all the articles posted after June 2022 have taken a strange turn.

An anonymous reader quotes a report from The Verge's Alex Heath: Meta plans to release its first pair of smart glasses with a display in 2025 alongside a neural interface smartwatch designed to control them, The Verge has learned. Meanwhile, its first pair of full-fledged AR glasses, which CEO Mark Zuckerberg has predicted will eventually be as widely used as mobile phones, is planned for 2027. The details were shared with thousands of employees in Meta's Reality Labs division on Tuesday during a roadmap presentation of its AR and VR efforts that was shared with The Verge. Altogether, they show how Meta is planning to keep investing in consumer hardware after a series of setbacks and broader cost cutting across the company.

With regards to the VR roadmap, employees were told that Meta's flagship Quest 3 headset coming later this year will be two times thinner, at least twice as powerful, and cost slightly more than the $400 Quest 2. Like the recently announced Quest Pro, it will prominently feature mixed reality experiences that don't fully immerse the wearer, thanks to front-facing cameras that pass through video of the real world. [...] There will be 41 new apps and games shipping for the Quest 3, including new mixed reality experiences to take advantage of the updated hardware, Rabkin said. In 2024, he said that Meta plans to ship a more "accessible" headset codenamed Ventura. "The goal for this headset is very simple: pack the biggest punch we can at the most attractive price point in the VR consumer market."

During Tuesday's roadmap presentation, Alex Himel, the company's vice president for AR, laid out the plan for a bevy of devices through 2027. The first launch will come this fall with the second generation of Meta's camera-equipped smart glasses it released in 2021 with Luxottica, the parent company of Ray-Ban. In 2025, Himel said the third generation of the smart glasses will ship with a display that he called a "viewfinder" for viewing incoming text messages, scanning QR codes, and translating text from another language in real time. The glasses will come with a "neural interface" band that allows the wearer to control the glasses through hand movements, such as swiping fingers on an imaginary D-pad. Eventually, he said the band will let the wearer use a virtual keyboard and type at the same words per minute as what mobile phones allow. While Meta halted development of its smartwatch with dual cameras,Himel said that the company is still working on another smartwatch to accompany its 2025 glasses. "We don't want people to have to choose between an input device on their wrist and smartwatch functionality that they've come to love," he said. "So we are building a neural interfaces watch. Number one, this device will do input: input to control your glasses, input to control the functionality on your wrist, and input to control the world around you."

The Verge's Alex Heath adds: "Meta's first true pair of AR glasses, which the company has been internally developing for 8 years under the codename Orion, are more technically advanced, expensive, and designed to project high-quality holograms of avatars onto the real world." These glasses will "won't be released to the public until 2027," but an "internal launch" for employees will begin in 2024.

As for advertising, Meta is planning to utilize its existing business model for these future devices. "We should be able to run a very good ads business," he said. "I think it's easy to imagine how ads would show up in space when you have AR glasses on. Our ability to track conversions, which is where there has been a lot of focus as a company, should also be close to 100 percent."

"If we're hitting anything near projections, it will be a tremendous business," he said. "A business unlike anything we've seen on mobile phones before."

Bluesky, the Twitter alternative backed by Twitter co-founder and CEO Jack Dorsey, has hit the App Store and more testers are gaining access. Though the app is still only available as an invite-only beta, its App Store arrival signals that a public launch could be nearing. TechCrunch reports: We haven't heard much from Bluesky since October 2022, when the team behind the project shared an update on the Bluesky blog, detailing the status of the social protocol that powers its new Twitter-like app, also called Bluesky. AT (originally called ADX, or "Authenticated Transfer Protocol,") is Bluesky's main effort while the Bluesky mobile app serves to showcase the protocol in action. [...]

We received an invite to the service and found it to be a functional, if still rather bare-bones, Twitter-like experience. Users create a handle which is then represented as @username.bsky.social as well as the display name that appears more prominently in bold text, as on Twitter. As a brand-new app, Bluesky's suggested user list didn't immediately impress with big names of public figures during onboarding. The app itself presents a simplified user interface where you can click a plus button to create a post of 256 characters, which can include photos. Where Twitter asks "What's happening?," Bluesky asks "What's up?" You can search for and follow other individuals, much like on Twitter, then view their updates in a Home timeline. User profiles contain the same sort of features you'd expect: a profile pic, background, bio and metrics, like the number of followers and posts a user has, as well as how many people they're following. Profile feeds are also divided into two sections, like Twitter: posts and posts & replies.

Bluesky users can share, mute and block accounts, but advanced tools, like adding them to lists, are not yet available. The discover tab in the bottom center of the app's navigation is useful, offering more "who to follow" suggestions and a running feed of recently posted Bluesky updates. The latter gives you the opportunity to find more people who you might like to follow, based on their posts rather than just a bio. Posts themselves can be replied to, retweeted, liked and, from a three-dot menu, reported, shared via the iOS Share Sheet to other apps, or copied as text. Another tab lets you check on your Notifications, including likes, reposts, follows and replies, also much like Twitter. There are no DMs. You can download the app here, but you're still going to need an invite code.

Gmail client-side encryption (CSE) is now generally available for Google Workspace Enterprise Plus, Education Plus, and Education Standard customers. BleepingComputer reports: The feature was first introduced in Gmail on the web as a beta test in December 2022, after being available in Google Drive, Google Docs, Sheets, Slides, Google Meet, and Google Calendar (in beta) since last year. Once enabled, Gmail CSE ensures that any sensitive data sent as part of the email's body and attachments (including inline images) will be unreadable and encrypted before reaching Google's servers. It's also important to note that the email header (including subject, timestamps, and recipients lists) will not be encrypted. "Client-side encryption takes this encryption capability to the next level by ensuring that customers have sole control over their encryption keys -- and thus complete control over all access to their data," Googled explained.

"Starting today, users can send and receive emails or create meeting events with internal colleagues and external parties, knowing that their sensitive data (including inline images and attachments) has been encrypted before it reaches Google servers. As customers retain control over the encryption keys and the identity management service to access those keys, sensitive data is indecipherable to Google and other external entities."

Tesla has finally started to open some Supercharger stations to non-Tesla electric cars owners in the US and it explained how it works. Electrek reports: As we previously reported, everything is handled through the app. Non-Tesla EV owners simply have to download the Tesla app, create an account, add a credit card for payment, and then they can roll up to some of the select few Supercharger stations now equipped with a Magic Dock -- primarily in New York for now. In the app, electric car owners can see the station and select the stall where they park. After, they simply have to grab the handle where the CCS adapter will come out of the Magic Dock:

At the moment, it appears that only half a dozen stations in the state of New York are available to non-Tesla EV owners, but the number is expected to grow rapidly as Tesla deploys the Magic Dock (the integrated CCS adapter to work with non-Tesla EVs) at more stations and builds new ones.

Nvidia is releasing new GPU drivers today that will upscale old blurry web videos on RTX 30- and 40-series cards. The Verge reports: RTX Video Super Resolution is a new AI upscaling technology from Nvidia that works inside Chrome or Edge to improve any video in a browser by sharpening the edges of objects and reducing video artifacts. Nvidia will support videos between 360p and 1440p up to 144Hz in frame rate and upscale all the way up to 4K resolution.

This impressive 4K upscaling has previously only been available on Nvidia's Shield TV, but recent advances to the Chromium engine have allowed Nvidia to bring this to its latest RTX 30- and 40-series cards. As this works on any web video, you could use it to upscale content from Twitch or even streaming apps like Netflix where you typically have to pay extra for 4K streams.

An anonymous reader writes quotes a report from Ars Technica: Did you ever see that movie The Ring? People who watched a cursed, creepy video would all mysteriously die in seven days. Somehow Google seems to have re-created the tech version of that, where the creepy video is this clip of the 1979 movie Alien, and the thing that dies after watching it is a Google Pixel phone. As noted by the user 'OGPixel5" on the Google Pixel subreddit, watching this specific clip on a Google Pixel 6, 6a, or Pixel 7 will cause the phone to instantly reboot. Something about the clip is disagreeable to the phone, and it hard-crashes before it can even load a frame. Some users in the thread say cell service wouldn't work after the reboot, requiring another reboot to get it back up and running.

The leading theory floating around is that something about the format of the video (it's 4K HDR) is causing the phone to crash. It wouldn't be the first time something like this happened to an Android phone. In 2020, there was a cursed wallpaper that would crash a phone when set as the background due to a color space bug. The affected phones all use Google's Exynos-derived Tensor SoC, so don't expect non-Google phones to be affected by this. Samsung Exynos phones would be the next most-likely candidates, but we haven't seen any reports of that. According to CNET, the issue has been addressed and a full fix will be deployed in March.