Microsoft Has Been Secretly Testing Its Bing Chatbot ‘Sydney’ For Years

According to The Verge, Microsoft has been secretly testing its Sydney chatbot for several years after making a big bet on bots in 2016. From the report: Sydney is a codename for a chatbot that has been responding to some Bing users since late 2020. The user experience was very similar to what launched publicly earlier this month, with a blue Cortana-like orb appearing in a chatbot interface on Bing. “Sydney is an old codename for a chat feature based on earlier models that we began testing in India in late 2020,” says Caitlin Roulston, director of communications at Microsoft, in a statement to The Verge. “The insights we gathered as part of that have helped to inform our work with the new Bing preview. We continue to tune our techniques and are working on more advanced models to incorporate the learnings and feedback so that we can deliver the best user experience possible.”

“This is an experimental AI-powered Chat on Bing.com,” read a disclaimer inside the 2021 interface that was added before an early version of Sydney would start replying to users. Some Bing users in India and China spotted the Sydney bot in the first half of 2021 before others noticed it would identify itself as Sydney in late 2021. All of this was years after Microsoft started testing basic chatbots in Bing in 2017. The initial Bing bots used AI techniques that Microsoft had been using in Office and Bing for years and machine reading comprehension that isn’t as powerful as what exists in OpenAI’s GPT models today. These bots were created in 2017 in a broad Microsoft effort to move its Bing search engine to a more conversational model.

Microsoft made several improvements to its Bing bots between 2017 and 2021, including moving away from individual bots for websites and toward the idea of a single AI-powered bot, Sydney, that would answer general queries on Bing. Sources familiar with Microsoft’s early Bing chatbot work tell The Verge that the initial iterations of Sydney had far less personality until late last year. OpenAI shared its next-generation GPT model with Microsoft last summer, described by Jordi Ribas, Microsoft’s head of search and AI, as “game-changing.” While Microsoft had been working toward its dream of conversational search for more than six years, sources say this new large language model was the breakthrough the company needed to bring all of its its Sydney learnings to the masses. […] Microsoft hasn’t yet detailed the full history of Sydney, but Ribas did acknowledge its new Bing AI is “the culmination of many years of work by the Bing team” that involves “other innovations” that the Bing team will detail in future blog posts.

Read more of this story at Slashdot.

Spotify Is Testing Playlists That Could Be Unlocked By NFT Holders

Unlocking exclusive access has been a long-held promise of a lot of NFT-based communities. And now, Spotify is helping some of them realize that claim with token-gated playlists. TechCrunch reports: According to a series of tweets by Kingship, a metaverse band signed to Universal Music Group (UMG), the streaming company is piloting playlists that could be unlocked through NFTs in certain geographies. Under the pilot, Kingship has released a special playlist that could be accessed only by Kingship key card NFT holders. The group posted a series of steps that involves linking a crypto wallet like Metamask, Trust Wallet, Rainbow, Ledger Live, or Zerion to authenticate the NFT that unlocks the playlist. Kingship said that currently, this experience is only available to Android users in the U.S., the U.K., Germany, Australia and New Zealand. “At Spotify, we routinely conduct a number of tests in an effort to improve our user experience. Some of those end up paving the path for our broader user experience and others serve only as important learnings. We have no further news to share on future plans at this time,” a Spotify spokesperson said.

Read more of this story at Slashdot.

At Least One Open Source Vulnerability Found In 84% of Code Bases, Report Finds

L.Kynes shares a report from CSO Online: At a time when almost all software contains open source code, at least one known open source vulnerability was detected in 84% of all commercial and proprietary code bases examined by researchers at application security company Synopsys. In addition, 48% of all code bases analyzed by Synopsys researchers contained high-risk vulnerabilities, which are those that have been actively exploited, already have documented proof-of-concept exploits, or are classified as remote code execution vulnerabilities. The vulnerability data — along with information on open source license compliance — was included in Synopsys’ 2023 Open Source Security and Risk Analysis (OSSRA) report (PDF), put together by the company’s Cybersecurity Research Center (CyRC). “Of the 1,703 codebases that Synopsys audited in 2022, 96% of them contained open source,” adds L.Kynes, citing the report. “Aerospace, aviation, automotive, transportation, logistics; EdTech; and Internet of Things are three of the 17 industry sectors included in the report that had open source in 100% of their audited codebases. In the remaining verticals, over 92% of the codebases contained open source.”

Read more of this story at Slashdot.

Apple Is Reportedly Closer To Bringing No-Prick Glucose Monitoring To the Watch

According to Bloomberg, Apple’s quest to bring blood glucose monitoring to the Apple Watch is now at a “proof-of-concept stage.” The last remaining hurdle is for it to be made smaller. Engadget reports: The technology, which uses lasers to gauge glucose concentration under the skin, was previously tabletop sized but has reportedly advanced to the point where an iPhone-sized wearable prototype is in the works. The system would not only help people with diabetes monitor their conditions, but would ideally alert people who are prediabetic, the insiders say. They could then make changes that prevent Type 2 (adult onset) diabetes.

The project has supposedly been in development for a long time. It began in 2010, when an ailing Steve Jobs had his company buy blood glucose monitoring startup RareLight. Apple is said to have kept the effort secret by operating it as a seemingly isolated firm, Avolonte Health, but folded it into a previously unknown Exploratory Design Group (XDG). CEO Tim Cook, Apple Watch hardware lead Eugene Kim and other top leaders have been involved.

Any real-world product is likely years away, according to Bloomberg. The industry also doesn’t have a great track record of bringing no-prick monitors to market. In 2018, Alphabet’s health subsidiary Verily scrapped plans for a smart contact lens that would have tracked glucose using tears. Even major brands with vast resources aren’t guaranteed success, in other words, and it’s not clear how accurate Apple’s solution would be.

Read more of this story at Slashdot.

Instagram Co-Founders Launch Personalized News App ‘Artifact’

Artifact, the personalized news reader built by Instagram’s co-founders, is now open to the public, no sign-up required. TechCrunch reports: With today’s launch, Artifact is dropping its waitlist and phone number requirements, introducing the app’s first social feature and adding feedback controls to better personalize the news reading experience, among other changes. […] With today’s launch, Artifact will now give users more visibility into their news reading habits with a newly added stats feature that shows you the categories you’ve read as well as the recent articles you read within those categories, plus the publishers you’ve been reading the most. But it will also group your reading more narrowly by specific topics. In other words, instead of just “tech” or “AI,” you might find you’ve read a lot about the topic “ChatGPT,” specifically.

In time, Artifact’s goal is to provide tools that would allow readers to click a button to show more or less from a given topic to better control, personalize and diversify their feed. In the meantime, however, users can delve into settings to manage their interests by blocking or pausing publishers or selecting and unselecting general interest categories. Also new today is a feature that allows you to upload your contacts in order to see a signal that a particular article is popular in your network. This is slightly different from Twitter’s Top Articles feature, which shows you articles popular with the people you follow, because Artifact’s feature is more privacy-focused.

“It doesn’t tell you who read it. It doesn’t tell you how many of them read it, so it keeps privacy — and we clearly don’t do it with just one read. So you can’t have one contact and like figure out what that one contact is reading … it has to meet a certain minimum threshold,” notes [Instagram co-founder Kevin Systrom]. This way, he adds, the app isn’t driven by what your friends are reading, but it can use that as a signal to highlight items that everyone was reading. In time, the broader goal is to expand the social experience to also include a way to discuss the news articles within Artifact itself. The beta version, limited to testers, offers a Discover feed where users can share articles and like and comment on those shared by others. There’s a bit of a News Feed or even Instagram-like quality to engaging with news in this way, we found.

Read more of this story at Slashdot.

AI-Created Images Lose US Copyrights In Test For New Technology

Images in a graphic novel that were created using the artificial-intelligence system Midjourney should not have been granted copyright protection, the U.S. Copyright Office said in a letter seen by Reuters. From the report: “Zarya of the Dawn” author Kristina Kashtanova is entitled to a copyright for the parts of the book she wrote and arranged, but not for images she made using Midjourney, the office said in its letter, dated Tuesday. The decision is one of the first by a U.S. court or agency on the scope of copyright protection for works created with AI, and comes amid the meteoric rise of generative AI software like Midjourney, Dall-E and ChatGPT.

The Copyright Office said in its letter that it would reissue its registration for “Zarya of the Dawn” to omit images that “are not the product of human authorship.” […] Midjourney is an AI-based system that generates images based on text prompts entered by users. Kashtanova wrote the text of “Zarya of the Dawn,” and Midjourney created the book’s images based on her prompts.

Read more of this story at Slashdot.

Amazon Closes $3.9 Billion Deal To Acquire One Medical

An anonymous reader quotes a report from CNBC: Amazon on Wednesday said it had closed its $3.9 billion deal for primary care provider One Medical. Amazon agreed last July to acquire One Medical to deepen its presence in health care, and “dramatically improve” the experience of getting medical care. Amazon has long had ambitions to expand into health care, buying online pharmacy PillPack in 2018 for $750 million, then launching its own virtual clinic for chronic conditions, and prescription perks for Prime members. The deal gives Amazon access to One Medical’s more than 200 brick-and-mortar medical offices in 26 markets, and roughly 815,000 members.

The purchase was the first major deal announced since CEO Andy Jassy took the helm from founder Jeff Bezos in July 2021, and Jassy has indicated he sees health care as a major area of expansion. In a statement, he said health care is ripe for disruption, citing long appointment times and the complexities of primary care. “Customers want and deserve better, and that’s what One Medical has been working and innovating on for more than a decade,” Jassy said in a statement. “Together, we believe we can make the health care experience easier, faster, more personal, and more convenient for everyone.” Amazon said it would discount One Medical memberships for U.S. users to $144 from $199 for the first year, regardless of whether they’re a Prime subscriber.

Read more of this story at Slashdot.

Biden Won’t Stop a Potential Ban On Importing Apple Watches

Medical device maker AliveCor announced today that President Biden has upheld an International Trade Commission ruling that could result in a potential import ban on the Apple Watch over its EKG feature. The Verge reports: Back in December, the ITC issued a final determination (PDF) that Apple had infringed on AliveCor’s wearable EKG tech. In the ruling, the ITC recommended a limited exclusion order and a cease-and-desist order for Apple Watch models with EKG features. If enforced, that would mean that Apple would no longer be able to import Apple Watch with EKG capabilities into the US for sale. According to Apple spokesperson Hannah Smith, the company will appeal the ITC’s decision to the Federal Circuit.

A veto from Biden would have rendered the issue moot. According to The Hill, while presidents generally don’t interfere with ITC rulings, in 2013, former President Obama vetoed a similar import ban after the ITC ruled that iPhones and iPads infringed on Samsung tech. It’s possible that Apple was hoping for history to repeat itself, as it reportedly amped up lobbying last week ahead of Biden’s decision.
https://www.theverge.com/2023/1/11/23550036/the-apple-watchs-blood-oxygen-feature-is-at-the-center-of-a-potential-import-ban
Biden’s decision doesn’t mean every Apple Watch from the Series 4 to the Apple Watch Ultra (excluding both generations of the SE) is about to disappear off shelves. Apple’s Smith told The Verge the ITC’s ruling doesn’t have any real impact at the moment. That’s because the Patent Trial and Appeal Board recently ruled that AliveCor’s EKG tech isn’t actually patentable, and AliveCor would have to win its appeal (PDF) to that ruling for any potential ban to take effect. However, AliveCor isn’t the only medical tech company that’s seeking an import ban on the Apple Watch via the ITC. Masimo also sued Apple for allegedly infringing on five of its pulse oximetry patents. Last month, an ITC judge also ruled in Masimo’s favor and will decide whether a potential import ban is warranted in May. If so, that import ban would impact any Apple Watch with an SpO2 sensor (i.e., the Series 6 or later, excluding the SE.)

Read more of this story at Slashdot.

Microsoft To Test Bing Chat Tones, Allows Longer Chat Sessions

Microsoft will start testing Bing Chat tones, enabling users to switch between receiving answers that are either more creative or more focused on their queries. Once the new option is available, you can choose between precise, balanced, or creative replies from Redmond’s AI-powered Bing chat bot. BleepingComputer reports: We are “going to begin testing an additional option that lets you choose the tone of the Chat from more Precise — which will focus on shorter, more search focused answers — to Balanced, to more Creative — which gives you longer and more chatty answers,” the Bing Team said. “The goal is to give you more control on the type of chat behavior to best meet your needs.”

Microsoft also says chat limits (5 chat turns per session and a total of 50 per day) that were introduced on Friday to address Bing Chat’s confusing and, at times, aggressive replies have now been expanded to allow for a more “natural daily use of Bing.” Starting today, the number of chat turns has been increased to six per session, to 60 chats per day. Next, Redmond plans to increase the daily cap to 100 total chats and ensure that normal searches will no longer count against the total number of chats.

Read more of this story at Slashdot.