First up, we had a major issue with the 990 Pro, the company's flagship NVMe product. Our own reviewer Robbie Khan noticed that the new 990 Pro was rapidly dropping health, with around a 1-percentage point drop almost every single day. Although Samsung was hesitant at first to acknowledge any major bug about the drive, the South Korean giant was sort of forced to confirm the issue after realizing how widespread the problem was as many online had similar experiences. You can follow the saga in its entirety in this article.

Fast forward to now, Samsung has confirmed that it was able to reproduce the issue in its internal testing. The company is promising a firmware update that will fix the problem. A Samsung representative posted on its community forums the following message (translated to English):

Hello everyone, we have received feedback from our technicians on this:

Samsung is currently reviewing customer inquiries about anomalies related to the SMART Percentage Used (on Samsung Magician) and/or Health Status (on Crystal Disk Info) of the Samsung SSD 990 PRO.

We were able to recreate the error under certain conditions of use.

To fix the problem, Samsung is expected to release a firmware update in February.

Hence a firmware update with the fix should be rolling out soon.

While this was the first major bug to hit Samsung's NVMe drives, the 990 Pro's predecessor, 980 Pro, which is also a very popular PCIe drive from the company, was found to be dying under certain scenarios. Samsung later fixed the issue with firmware versions "4B2QGXA7" and "5B2QGXA7" (via Puget Systems). The problematic firmware version was found to be "3B2QGXA7". Hence if your 980 Pro is on that firmware, you should definitely update it using something like Samsung Magician.

Via: ComputerBase

After fixing dying 980 Pro SSDs, Samsung promises firmware for 990 Pro health issue as well

Pichai stated that Bard will not only be able to handle answers to simple questions, like Google's current search engine can do already, but it can also handle more complex inquiries, such as asking how to give information on the discoveries of the James Webb Space Telescope to a nine year old. The screenshot of that example also has thumbs up and thumbs down logos, apparently so you can rank the answer that is given to you. You can also apparently go deeper into that answer by clicking on the "Check it" option.

The first public version of Bard will launch with what Pichai called a "lightweight model version of LaMDA." This will reportedly use less computer power so that Google can get more feedback from a wide range of users. Bard is currently in testing now and will go live to the general public sometime in the coming weeks.

In addition, Pichai stated that that Google Search will soon enable more AI-driven features "that distill complex information and multiple perspectives into easy-to-digest formats." That includes more details to questions alongside links to web sites that might have more information on the subject of the search.

Google's move to launch Bard and to add more AI-driven features to Search is clearly a response to the rise of ChatGPT. Microsoft is reportedly going to add a new and faster version of ChatGPT to its own Bing search engine in the coming weeks as well.

Google reveals its ChatGPT competitor, Bard

The SEC filing included an all-team email sent by Dell's co-COO Jeff Clarke. He stated that efforts to cut back on spending at Dell, including a hiring freeze and limiting travel, have not been enough and that "market conditions continue to erode with an uncertain future". He said of the layoffs, "There is no tougher decision, but one we had to make for our long-term health and success.”

The new Dell layoffs are part of the overall downward trend in PC sales. IDC reported in January that shipments of PCs worldwide in the fourth quarter of 2022 were down 28.1 percent from the same period in 2021. The same report added that Dell, the third largest PC maker, saw its year-over-year sales plummet 37.2 percent in the fourth quarter of 2022.

Dell's move to lay off 5 percent of its workforce is in tune with all of the other tech layoffs that have been announced in the past few months. Google and Spotify have both announced headcount reductions of 6 percent, and Microsoft also announced mass layoffs a few weeks ago that affected less than 5 percent of its workforce.

Dell is the latest tech company to announce layoffs with 6,650 workers affected

9to5Google reports that the Fitbit iOS and Android mobile apps are not currently syncing up on the Fitbit fitness wearables and smartwatches. The apps are also not currently syncing up to Google's Pixel Watch, which has Fitbit software inside. In addition, some stats are currently not showing up, such as steps, calories burned and more. The My Dashboard feature on Fitbit's website is also not working, although the rest of the website is currently unaffected.

Downdetector.com has confirmed a number of outage reports from Fitbit users. The official Fitbit Support Twitter page has posted word that the company is aware of the issues and are working now to fix it. We will update this post when we learn more.

Fitbit is currently down for some users

Report: Microsoft expects UK to block Activision merger deal

ChatGPT continues to shake up the tech world, with Google seemingly scrambling to find a response to the Microsoft-backed ChatBot from OpenAI. Its latest plan? Find its own version of OpenAI.

According to Bloomberg(opens in new tab), Google has invested nearly half a billion dollars into AI startup Anthropic. Neither Google nor Anthropic were willing to comment on the investment, but both companies made separate announcements about Anthropic using Google for its cloud computing services. When announcing that deal, Google stated that it would lend computing power and AI chips to Anthropic for training and deploying AI products, so there is some public evidence of a deeper partnership. 

Currently, Anthropic has a language model assistant named Claude, which should be similar to ChatGPT. While it has yet to be released to the public, Anthropic says that expanded access could be coming in the near future.

It will be interesting to see if this deal resembles the $10 billion deal Microsoft recently made with OpenAI. That deal —  and previous deals — have given Microsoft exclusive access to the GPT-3 source code that powers ChatGPT, giving it a significant competitive advantage regarding chatbot AI. Google will need to work hard to claw back to level footing with Microsoft. 

Google AI: What else is coming 

(Image credit: Shutterstock)

This deal with Anthropic isn’t the only AI chatbot project that Google is involved in. In recent weeks they’ve already announced MusicLM, an AI tool that generates new music from text prompts, and Apprentice Bard, an in-house AI chatbot developed by Google that is directly aimed at supplanting ChatGPT.

Google describes MusicLM as “a model generating high-fidelity music from text descriptions”. In practice, this AI tool works by users providing text prompts with details such as genre, tempo, etc., and the AI spits out a song based on that prompt. The sound quality is 24kHz according to Google, which is just a bit above analog formats like vinyl and cassette tapes but below what you’d expect from Spotify or Apple Music. Still, when I played it through my Creative T15 Wireless speakers it sounded relatively good.  

MusicLM can even create vocals for the songs it creates, though the one sample I heard didn’t blow me away lyrically. 

Apprentice Bard AI, meanwhile, operates similarly to ChatGPT, where the user holds a conversation with the chatbot, providing questions and receiving responses. The key difference is that Apprentice Bard uses LaMDA (Language Model for Dialogue Application) as opposed to ChatGPT’s proprietary GPT-3.5 technology. Another key difference is that ChatGPT was trained until 2021, whereas Apprentice Bard’s training is still ongoing. This means that Bard may have more up-to-date knowledge than ChatGPT.

So far, Google is internally impressed with Apprentice Bard. It is reported to even suggest that it would offer the chatbot AI a job as a level 3 engineer. We may not have to wait too much longer to see its performance for ourselves, as Google is expected to hold an event in Paris on February 8th, streamed live on YouTube(opens in new tab). 

Source

A leaner Windows 11

If you're struggling with Windows 11's strict system requirements or would just like to experiment with a modified version of the OS (at your own risk!), you might want to take a look at Tiny11. This modified image of Windows 11 gets rid of the Windows Component Store (WinSxS) and other modules the developer considers "bloat", resulting in a lightweight version of the OS for less powerful computers, requiring just 8GB of storage space and 2GB of RAM.

Naturally, you ideally shouldn't be running this on your primary machine and should consider using a virtual machine if you really want to give it a spin. Several of our readers also raised concern about the security of the ISO and the fact that it breaks Microsoft's EULA for Windows 11, so make sure to check the discussion here to get some doubts about what you could be getting into clarified.

In a similar vein, third-party tool ThisIsWin11 has now been repackaged as "Debloos", which stands for Debloat OS. The developer describes it as the "Universal Debloater and PC Manager for Windows 11". In addition, you could also use a tool called WindHawk to modify your Windows Paint installation and bring it to the modern age with dark mode.

But if you'd like to customize your Windows installation using the tools Microsoft offers, you'll be happy to know that Microsoft is working on another PowerToys utility called FancyMouse, which will make it easier to navigate your cursor across large or multiple displays. In the meantime, we have version 0.67.0 of PowerToys, bringing the Quick access system tray launcher, among many other things.

Since we are discussing experimentation with Windows in this section, we'd also like to bring your attention to the latest Windows 11 Dev Channel preview build 25290. It introduced Widgets for Spotify and Phone Link, along with new Start menu badges. There are some improvements to tabbed File Explorer too but if you don't like them, try out Files App instead. There was also a relatively minor Beta build with a few bug fixes.

Speaking of issues with Windows, Microsoft released a new out-of-band (OOB) update this week to fix a .NET issue in Windows 10, 11, and Server. Meanwhile, some users have been complaining about full-screen Windows 11 upgrade ads with two "yes" buttons (!!!) and unskippable ads for a Microsoft 365 during the Windows 10 out-of-box experience (OOBE). Others speculated that Microsoft was stealing their private data through a stealth update but that's actually not the case. If anything, Microsoft should be proactively transparent in this area lest it start losing its marketshare to other operating systems.

A flurry of Teams updates

Microsoft Teams dominated a lot of our headlines in this week. For starters, Microsoft revealed a list of features that it is bringing to Teams in March, including "ultrasound howling detection", explicit recording consent, and more. In the same vein, it also recapped all the capabilities it added to Teams (and Excel) in January 2023.

In terms of other additions to Teams, we got new video filters packaged as Teams apps so customers can personalize their online meetings. Teams Premium launched this week as well, it brings advanced webinars and integration with the Elgato Stream Deck - and if you're interested in Teams hardware, there's also a 27-inch ThinkSmart View Plus Teams standalone display from Lenovo that will retail for $3,100 soon. The new Premium license will cost $7 per month until June 30 before increasing to $10 in July. If you don't fancy Teams though, you might want to have a look at the recently revamped Skype app - yes, it's still alive.

Switching gears to some Microsoft Edge news, the Canary version of the browser has surprisingly gotten rid of the Mica effect. But the same channel has introduced a "minimal toolbar experience" that seemingly aims to return some screen estate to the user, along with a feature that will inform you if you are trying to save a weak password in the browser. Meanwhile, Edge Dev 111.0.1660.0 brings improved Workspaces and some bug fixes.

In related news, Microsoft, Apple, Google, Mozilla and several other companies also announced the goals for their Interop 2023 project. The initiative aims to improve cross-browser compatibility and was successful last year too. Each browser vendor will be hoping that this will help improve their marketshare in the long-term too.

ChatGPT hype

The hype train with ChatGPT and OpenAI's other endeavors area is still going very strong. Microsoft Teams Premium, which became generally available this week as mentioned in the previous section, is also powered by GPT-3.5 in some areas. For those unaware, this is the same large language model that has made ChatGPT possible. In Teams Premium, it aims to make meetings "more intelligent, personalized, and protected." For instance, it can translate captions in real-time, help create meeting templates, and automatically generate meeting notes, recommended tasks, and personalized highlights.

Microsoft is not just stopping there. Screenshots of its alleged ChatGPT-powered Bing leaked online during the week. According to the person who managed to access this version of Bing, the search bar will be able to take input up to 1,000 characters and will not only answer questions but also generate text, just like the current version of ChatGPT.

An estimate from an analyst revealed that around 100 million people accessed ChatGPT in January 2023, which is almost double the figure reported in December 2022. Microsoft will be hoping that this high level of interest in the language model persists as it rushes to get its ChatGPT-powered products out the door.

ChatGPT's creator OpenAI isn't sitting idle and just enjoying public admiration either. A recent report claims that the company is now working on a ChatGPT mobile app and an AI model that can also generate videos.

Git gud

We'll kick this section off with a new report which indicates that things are not going well at the Halo camp. Apparently, the development studio is starting from scratch with Unreal Engine for its next game, which will have no story content. This follows reports of 343 Industries being hit hard by recent layoffs at Microsoft. In the same space, the next game from Coalition is going to be "Gears 6", following the cancelation of two other non-Gears titles.

A bunch of classic games are also going to be removed from the Xbox 360 Marketplace soon, including several high-profile titles. Despite this, Microsoft has confirmed that the Xbox 360 Marketplace is not on the way out (yet) and will continue being accessible even after May 2023 after an official note on the company's website seemed to indicate otherwise. On the other hand, Xbox-exclusive CrossfireX is definitely shutting down on May 18, just 15 months after official launch.

In addition, we learned that the European Commission has sent its list of objections to Microsoft regarding its acquisition of Activision Blizzard. The details have not been made public yet. While Microsoft and Sony lock horns over the purchase, it seems that both of the companies will also be joining Nintendo in skipping E3 altogether this year.

In terms of game updates and releases, Age of Empires II is now out on Xbox consoles featuring PC crossplay while Cyberpunk 2077 on Windows has received hardware GPU scheduling, NVIDIA DLSS 3 and Reflex update.

Lastly, talking about game promotions and deals, we have For Honor, Eiyuden Chronicle: Rising, and Naruto to Boruto: Shinobi Striker gracing Xbox Free Play Days promotion while Games with Gold is offering For The King. And if none of this tickles your fancy, make sure to check out this Weekend's PC Game Deals, curated by our News Editor Pulasthi Ariyasinghe.

Dev Channel

• Chromebook owners will get easier access to Microsoft 365 and OneDrive soon

• Macrium Reflect gets performance boost, bug fixes, and a price tag

• Microsoft is not selling Windows 10 licenses via its website anymore

• Acer's latest additions to its TravelMate B series are Windows 11 22H2-ready

• Outlook's upcoming feature aims to promote better understanding of gender

• Samsung has announced flagship Galaxy Book3 and Book3 Ultra laptops powered by Windows 11

• Microsoft Store Ads are now generally available for all developers

• Microsoft: We are committed to HoloLens 2 and mixed reality

   

Under the spotlight

News Reporter Taras Buria wrote a very handy guide this week in which he explains how you can use a script to launch tabbed File Explorer with a bunch of pre-defined tabs, with some caveats.

Taras also authored a listicle about the top 10 improvements Windows 11 users are demanding for the Settings app, along with Microsoft's responses on each of the requests.

Finally, forum member Adam Bottjen penned his Tech Tip Tuesday piece on the topic of finding out how much free space you have on your PC, check it out here.

Logging off

We would like to bid you adieu with a concept of "Windows 11X", the details of which you can view here. Its designer has decided to use Windows 11 as the base for their concept and then add some changes on top. Some additions include Widgets in the Start menu, a unified notification center, and even a tablet mode of sorts. Of course, just because something looks pretty on paper doesn't indicate that it's user-friendly too, so make sure to head over to our coverage linked above and let us know your thoughts!

Microsoft Weekly: A leaner Windows 11, Teams update rush, and ChatGPT hype

Samsung Pro SSD reliability questioned as longtime partner shifts to Sabrent

China is now the third country worldwide to have built and delivered quantum computers after Canada and the United States first shipped theirs in 2011 and 2019, respectively.

Origin Quantum Computing Technology Co, established in Hefei in 2017, says it has delivered a 24-qubit quantum computer – known as Benyuan Wuyuan – that reportedly uses self-developed superconducting chip technology.

The company says it will launch a faster quantum computer called Wukong, the name of the Monkey King in Chinese mythology, later this year.

The announcement has prompted IT columnists and observers to speculate on China’s chances of catching up with the US in the race to lead the quantum computing revolution.

The consensus seems to be that China is four to five years behind the US in the quantum computer industry – but could in the decades ahead take the lead as the high-tech competition evolves.

For the origin of Origin, flash back to September 2017, when Guo Guoping, deputy director of the Key Laboratory of Quantum Information, Chinese Academy of Sciences (CAS), led a team of scientists in establishing Origin Quantum in Hefei. In Mandarin Chinese, Benyuan means Origin.

In its series A round of fundraising in January 2021, the company received over 100 million yuan (US$15.5 million) from a group of investors led by the China Internet Investment Fund. In the series B round last July, it raised 1 billion yuan from new investors led by Shenzhen Capital Group’s Hotland Investment Asset Management.

In 2011, Canadian firm D-Wave Systems released the world’s first commercial quantum computer – called D-Wave One. Equipped with a 128-qubit chipset, the computer was priced at US$10 million. In 2019, IBM unveiled its Q System One, a 20-qubit machine.

Second in the race was IBM’s Q System One. Photo: IBM

In its Wuyuan series, Origin Quantum launched its 6-qubit quantum chip, namely KF-C6-130, in 2020 and a 24-qubit quantum chip, KF-C24-100, in 2021. Last year, it released its first collaborative computing system solution of a quantum computer, using its own operating system called Benyuan Sinan, or Origin Pilot.

The company said this week that it delivered a 24-qubit quantum computer but it did not disclose more details about its parts and technologies. It also did not identify the customer or reveal the selling price of this machine.

In connection with the announcement, Guo Guoping looked into the future and predicted that within the next three to five years quantum computing will be put to practical use in people’s lives and production.

He said a complicated calculation that would take a month for 10 supercomputers to complete could be done in three to seven days with assistance from a quantum computer.

Guo originally studied foreign languages at the Hefei-based University of Science and Technology of China but then later changed his major to physics. He was mentored by CAS academician Guo Guangcan and received his PhD at the same university.

Speaking of acceleration, he spoke of witnessing rapid progress in the field. He recalled that when he was conducting experiments in a laboratory 20 years ago the lab was ill-equipped to the point of lacking even proper chairs and desks. With the investment funds now available to the quantum computing sector there’s no such problem today, he said.

In 2010, Guo started contributing to the National Basic Research Program, or the so-called 973 Program.

Guo said his original career plan didn’t foresee setting up Origin Quantum or even trying to make money. But once he started the company it became a passion for him: He once sold his own house to ensure the company survived, he said.

Quantum computing whiz Guo Guoping. Photo: China Speech Valley and Quantum Center, Hefei

Zhang Hui, general manager of Origin Quantum, said that although the company has delivered a quantum computer for the first time, it will have to achieve many more technological breakthroughs to increase its computing speed.

Zhang said a quantum computer will have commercial value only if its speed reaches 50 to 100 qubit. Since last year, he said, the company has started to explore the applications of quantum computing in the finance, biomedical and artificial intelligence sectors.
 
Some scientists have pointed out that a 30-qubit quantum computer has computing power equivalent to that of a supercomputer with a speed of 1 teraflop (a trillion floating point operations per second) while a 50-qubit quantum computer is faster than a supercomputer when doing certain tasks.

One well-read Chinese IT writer has laid out a three-stage process to commercialize quantum computers.

The first stage, he says, is to build prototypes of quantum computers. The second stage, which he predicts will happen in the next decade, will be to make specialized chips for different uses. He says more applications will be launched in the last stage, which will last for several more decades.

The writer says it is true that China’s development in quantum computing lags behind that of the US but he asserts that the US cannot stop China from developing.

In recent years, Tencent, Alibaba, Baidu and Huawei have been investing in quantum computing while major US players in the sector include Google, Microsoft and Intel.

In May 2021, a University of Science and Technology of China research team led by scientist Pan Jianwei launched Zuchongzhi 2, a superconducting quantum computer that needs to work at a temperature close to absolute zero.

But then IBM’s Eagle surpassed Zuchongzhi 2 to become the world’s fastest superconducting quantum machine.

Source

Each jarring alert was a new incoming call, heralding a possible car crash, heart attack or other life-threatening situation. Often, the phone operators heard a chilling sound at the far end of the line: silence, perhaps from a caller too incapacitated to respond.

At 9:07 a.m., one dispatcher, Eric Betts, responded to such a call. From the map on one of the seven monitors on his desk, he could see that the distress call originated from a slope at the Arapahoe Basin Ski Area. Mr. Betts tried calling back. A man picked up.

“Do you have an emergency?” Mr. Betts asked. No, the man said, he was skiing — safely, happily, unharmed. Slightly annoyed, he added, “For the last three days, my watch has been dialing 911.”

Winter has brought a decent amount of snowfall to the region’s ski resorts, and with it an avalanche of false emergency calls. Virtually all of them have been placed by Apple Watches or iPhone 14s under the mistaken impression that their owners have been debilitated in collisions.

As of September, these devices have come equipped with technology meant to detect car crashes and alert 911 dispatchers. It is a more sensitive upgrade to software on Apple devices, now several years old, that can detect when a user falls and then dial for help. But the latest innovation appears to send the device into overdrive: It keeps mistaking skiers, and some other fitness enthusiasts, for car-wreck victims.

Lately, emergency call centers in some ski regions have been inundated with inadvertent, automated calls, dozens or more a week. Phone operators often must put other calls, including real emergencies, on hold to clarify whether the latest siren has been prompted by a human at risk or an overzealous device.

“My whole day is managing crash notifications,” said Trina Dummer, interim director of Summit County’s emergency services, which received 185 such calls in the week from Jan. 13 to Jan. 22. (In winters past, the typical call volume on a busy day was roughly half that.) Ms. Dummer said that the onslaught was threatening to desensitize dispatchers and divert limited resources from true emergencies.

“Apple needs to put in their own call center if this is a feature they want,” she said.

Her call center and others have alerted Apple to the issue. In mid-January, the company sent four representatives to observe Ms. Dummer and her team for a day; she said they had plenty of examples to show off.

In a written statement, Alex Kirschner, an Apple spokesman, said, “We have been aware that in some specific scenarios these features have triggered emergency services when a user didn’t experience a severe car crash or hard fall.” The company noted that when a crash is detected, the watch buzzes and sends a loud warning alerting the user that a call is being placed to 911, and it provides 10 seconds in which to cancel the call.

Apple also said that updates to the software late last year had been intended to “optimize” the technology and reduce the number of false calls.

Mr. Kirschner added, “Crash Detection and Fall Detection are designed to get users help when they need it most, and it has already contributed to saving several lives.”

Apple maintains a collection of incidents in which the two technologies have come to the rescue. In one case, an Apple Watch alerted the authorities after a driver in Indianapolis had crashed into a telephone pole and the device dialed for assistance. In another, a watch called for help after a New Jersey man fell down a steep cliff while hiking.

In Colorado, call dispatchers had trouble recalling an instance in which a watch had saved a skier in distress. (Ms. Dummer added that her team had “very rarely” received false 911 calls from other companies’ devices, such as Android phones.)

The problem extends beyond skiers. “My watch regularly thinks I’ve had an accident,” said Stacey Torman, who works for Salesforce in London and teaches spin classes there. She might be safely on the bike, exhorting her class to ramp up the energy, or waving her arms to congratulate them, when her Apple Watch senses danger.

“I want to celebrate, but my watch really doesn’t want me to celebrate,” she said. Oh great, she thinks, “now my watch thinks I’m dead.”

Recently she fell while racing to catch a bus in the rain. “I went down hard, really hard,” she said. Her watch did not call 911, however. “When I did actually fall running for the bus?” she said. “Crickets.”

Jon Baron, who works in real estate in Golden, Colo., has an Apple Watch that has twice dialed the authorities. Once, he was at an amusement park playing the strongman game, in which the goal is to hit a lever with a hammer firmly enough to ring a bell. Mr. Baron swung, the bell rang, his wife and children seemed duly impressed — until “my watch started making this noise like an air-raid siren,” he said.

“I was trying to show I was physically able, which I thought I’d demonstrated fairly aptly, but my watch thought different,” he added.

Another time, Mr. Baron was in the Tampa airport when the intercom summoned him to a white paging telephone. An emergency services operator at the other end told him his device had dialed 911 for help. Was he OK? “I’m doing all right,” he said he had told the operator. “But I have a plane to catch.”

Apple introduced fall detection technology in 2018 after developing an algorithm based on the trajectory of a wrist wearing a watch and acceleration at time of impact. Its crash-detection technology, introduced in September, was tested in crash tests and labs on iPhones and Apple Watches.

But something about the way skiers accelerate and stop, or jostle, seems to set the technology on edge. And skiers, in helmets and layers of clothing, often do not to detect the warning, so they may not cancel the call or respond to the 911 dispatcher.

“A lot of people don’t feel it or hear it,” said Brett Loeb, emergency services communications director in Pitkin County, Colo., home to Aspen Mountain. Or, he proposed, even when they feel the vibration, “they think, ‘I don’t want to answer my phone right now — I’m having a great time; my phone is killing my buzz.’”

He noted that Aspen Mountain had recently posted signs at lift lines and ticket offices alerting Apple Watch and iPhone 14 users to the problem and encouraging them to upgrade to the latest software version or disable the service, to “prevent unnecessary trips to the slopes” by 911-dispatched ambulances.

At the Summit County call center on Sunday morning, Ms. Dummer was training a rookie dispatch supervisor, Jeff Fitch, to field 911 calls, while Mr. Betts and another dispatcher, David Benson, handled the overflow and communicated with ambulances and the police.

The siren went off, and Mr. Fitch picked up the call. “911,” he said into his headset. “Hello?” A monitor displayed the caller’s location on the ski slope; another displayed the caller’s number. Mr. Fitch leaned forward into the silence: “Hello?”

The watch calls kept coming; the siren kept blaring. Amid it all, Ms. Dummer and Mr. Fitch fielded a genuine distress call from a 78-year-old driver who was having trouble breathing. Traffic was bad, hindering ambulance access, and the dispatch team was calling two nearby counties for help.

Just before noon, Mark Watson, a sergeant with the sheriff’s office, walked into the dispatch room looking glum. “This is not a good day,” he said.

Ordinarily, he had other duties, including patrolling the backcountry on snowmobile, but the ghost calls had kept him at his desk. Whenever the 911 operators were unable to reach the owner of the watch or phone, the case was referred to Sergeant Watson, who would try calling and sending a text; if he didn’t hear back, he forwarded the issue to the ski patrol.

So far that day, Sergeant Watson had fielded seven referrals from 911, four of which he forwarded to the ski patrol. He turned to Ms. Dummer: How many crash-detection calls had come in overall? Eleven, she said, out of 30 calls total.

“I wanted to check the numbers,” he said. “I was writing a letter to Apple.” He described his basic message to the company: “I’m struggling to get my daily job done. I don’t have all day to do Apple products.”

In Grand County, home to a busy mountain called Winter Park, Sheriff Brett Schroetlin decided in late December to devote less attention to the crash-detection calls. Now if a 911 operator receives one from the slopes and no one is on the other end of the line, they know to ignore the call; no more referrals or follow-ups. None of the ghost calls so far have been real emergencies, Sheriff Schroetlin reasoned, and he couldn’t afford to waste limited resources. Besides, he said, there was a better technology: human beings.

“It’s rare that someone falls on the mountain and there’s not a passer-by,” he said. “We’re hoping to get an actual 911 call from the person or someone on the scene.”

The post ‘My Watch Thinks I’m Dead’ appeared first on New York Times.

Source

One day in mid-November, workers at OpenAI got an unexpected assignment: Release a chatbot, fast.

The chatbot, an executive announced, would be known as “Chat with GPT-3.5,” and it would be made available free to the public. In two weeks.

The announcement confused some OpenAI employees. All year, the San Francisco artificial intelligence company had been working toward the release of GPT-4, a new A.I. model that was stunningly good at writing essays, solving complex coding problems and more. After months of testing and fine-tuning, GPT-4 was nearly ready. The plan was to release the model in early 2023, along with a few chatbots that would allow users to try it for themselves, according to three people with knowledge of the inner workings of OpenAI.

But OpenAI’s top executives had changed their minds. Some were worried that rival companies might upstage them by releasing their own A.I. chatbots before GPT-4, according to the people with knowledge of OpenAI. And putting something out quickly using an old model, they reasoned, could help them collect feedback to improve the new one.

So they decided to dust off and update an unreleased chatbot that used a souped-up version of GPT-3, the company’s previous language model, which came out in 2020.

Thirteen days later, ChatGPT was born.

In the months since its debut, ChatGPT (the name was, mercifully, shortened) has become a global phenomenon. Millions of people have used it to write poetry, build apps and conduct makeshift therapy sessions. It has been embraced (with mixed results) by news publishers, marketing firms and business leaders. And it has set off a feeding frenzy of investors trying to get in on the next wave of the A.I. boom.

It has also caused controversy. Users have complained that ChatGPT is prone to giving biased or incorrect answers. Some A.I. researchers have accused OpenAI of recklessness. And school districts around the country, including New York City’s, have banned ChatGPT to try to prevent a flood of A.I.-generated homework.

Yet little has been said about ChatGPT’s origins, or the strategy behind it. Inside the company, ChatGPT has been an earthshaking surprise — an overnight sensation whose success has created both opportunities and headaches, according to several current and former OpenAI employees, who requested anonymity because they were not authorized to speak publicly.

An OpenAI spokesman, Niko Felix, declined to comment for this column, and the company also declined to make any employees available for interviews.

Before ChatGPT’s launch, some OpenAI employees were skeptical that the project would succeed. An A.I. chatbot that Meta had released months earlier, BlenderBot, had flopped, and another Meta A.I. project, Galactica, was pulled down after just three days. Some employees, desensitized by daily exposure to state-of-the-art A.I. systems, thought that a chatbot built on a two-year-old A.I. model might seem boring.

But two months after its debut, ChatGPT has more than 30 million users and gets roughly five million visits a day, two people with knowledge of the figures said. That makes it one of the fastest-growing software products in memory. (Instagram, by contrast, took nearly a year to get its first 10 million users.)

The growth has brought challenges. ChatGPT has had frequent outages as it runs out of processing power, and users have found ways around some of the bot’s safety features. The hype surrounding ChatGPT has also annoyed some rivals at bigger tech firms, who have pointed out that its underlying technology isn’t, strictly speaking, all that new.

ChatGPT is also, for now, a money pit. There are no ads, and the average conversation costs the company “single-digit cents” in processing power, according to a post on Twitter by Sam Altman, OpenAI’s chief executive, likely amounting to millions of dollars a week. To offset the costs, the company announced this week that it would begin selling a $20 monthly subscription, known as ChatGPT Plus.

Despite its limitations, ChatGPT’s success has vaulted OpenAI into the ranks of Silicon Valley power players. The company recently reached a $10 billion deal with Microsoft, which plans to incorporate the start-up’s technology into its Bing search engine and other products. Google declared a “code red” in response to ChatGPT, fast-tracking many of its own A.I. products in an attempt to catch up.

Mr. Altman has said his goal at OpenAI is to create what is known as “artificial general intelligence,” or A.G.I., an artificial intelligence that matches human intellect. He has been an outspoken champion of A.I., saying in a recent interview that its benefits for humankind could be “so unbelievably good that it’s hard for me to even imagine.” (He has also said that in a worst-case scenario, A.I. could kill us all.)

As ChatGPT has captured the world’s imagination, Mr. Altman has been put in the rare position of trying to downplay a hit product. He is worried that too much hype for ChatGPT could provoke a regulatory backlash or create inflated expectations for future releases, two people familiar with his views said. On Twitter, he has tried to tamp down excitement, calling ChatGPT “incredibly limited” and warning users that “it’s a mistake to be relying on it for anything important right now.”

He has also discouraged employees from boasting about ChatGPT’s success. In December, days after the company announced that more than a million people had signed up for the service, Greg Brockman, OpenAI’s president, tweeted that it had reached two million users. Mr. Altman asked him to delete the tweet, telling him that advertising such rapid growth was unwise, two people who saw the exchange said.

OpenAI is an unusual company, by Silicon Valley standards. Started in 2015 as a nonprofit research lab by a group of tech leaders including Mr. Altman, Peter Thiel, Reid Hoffman and Elon Musk, it created a for-profit subsidiary in 2019 and struck a $1 billion deal with Microsoft. It has since grown to around 375 employees, according to Mr. Altman — not counting the contractors it pays to train and test its A.I. models in regions like Eastern Europe and Latin America.

From the start, OpenAI has billed itself as a mission-driven organization that wants to ensure that advanced A.I. will be safe and aligned with human values. But in recent years, the company has embraced a more competitive spirit — one that some critics say has come at the expense of its original aims.

Those concerns grew last summer when OpenAI released its DALL-E 2 image-generating software, which turns text prompts into works of digital art. The app was a hit with consumers, but it raised thorny questions about how such powerful tools could be used to cause harm. If creating hyper-realistic images was as simple as typing in a few words, critics asked, wouldn’t pornographers and propagandists have a field day with the technology?

To allay these fears, OpenAI outfitted DALL-E 2 with numerous safeguards and blocked certain words and phrases, such as those related to graphic violence or nudity. It also taught the bot to neutralize certain biases in its training data — such as making sure that when a user asked for a photo of a C.E.O., the results included images of women.

These interventions prevented trouble, but they struck some OpenAI executives as heavy-handed and paternalistic, according to three people with knowledge of their positions. One of them was Mr. Altman, who has said he believes that A.I. chatbots should be personalized to the tastes of the people using them — one user could opt for a stricter, more family-friendly model, while another could choose a looser, edgier version.

OpenAI has taken a less restrictive approach with ChatGPT, giving the bot more license to weigh in on sensitive subjects like politics, sex and religion. Even so, some right-wing conservatives have accused the company of overstepping. “ChatGPT Goes Woke,” read the headline of a National Review article last month, which argued that ChatGPT gave left-wing responses to questions about topics such as drag queens and the 2020 election. (Democrats have also complained about ChatGPT — mainly because they think A.I. should be regulated more heavily.)

As regulators swirl, Mr. Altman is trying to keep ChatGPT above the fray. He flew to Washington last week to meet with lawmakers, explaining the tool’s strengths and weaknesses and clearing up misconceptions about how it works.

Back in Silicon Valley, he is navigating a frenzy of new attention. In addition to the $10 billion Microsoft deal, Mr. Altman has met with top executives at Apple and Google in recent weeks, two people with knowledge of the meetings said. OpenAI also inked a deal with BuzzFeed to use its technology to create A.I.-generated lists and quizzes. (The announcement more than doubled BuzzFeed’s stock price.)

The race is heating up. Baidu, the Chinese tech giant, is preparing to introduce a chatbot similar to ChatGPT in March, according to Reuters. Anthropic, an A.I. company started by former OpenAI employees, is reportedly in talks to raise $300 million in new funding. And Google is racing ahead with more than a dozen A.I. tools.

Then there’s GPT-4, which is still scheduled to come out this year. When it does, its abilities may make ChatGPT look quaint. Or maybe, now that we’re adjusting to a powerful new A.I. tool in our midst, the next one won’t seem so shocking.

Source

And like on desktops, there will also be a 12-core SKU in the 7045HX series. Dubbed "Ryzen 9 7845HX", the PassMark score of the chip has leaked today and the processor has put up very, very impressive figures, though it was sort of expected too, given the high core count.

As you can see in the image below, the 7845HX managed to put up 46,791 points in the benchmark which is nearly twice as fast as Apple's M2 Pro and M2 Max. Meanwhile, compared to the best Intel offering currently available, the Core i9-12900HX - which is a 16-core (8P + 8E), 24-thread part -, the AMD processor is 31% faster.

While we don't have the numbers for the 16 core 7945HX here, it will certainly be a bit faster overall than the 7845HX or around similar levels, as the performance could depend on the TDP constraint configured in a particular laptop.

Meanwhile, Intel also revealed its own cards in the form of the upcoming Raptor Lake mobile CPUs, the fastest one being the 55W Core i9-13950HX with 24 cores (8P + 16E) and 32 threads. According to the company, the 13950HX is up to 79% faster than the 12900HX in certain use cases like Blender, which is mighty impressive, and it also implies that even the fastest AMD 16 core part could have a hard time keeping up in such multi-threaded tasks.

Source: PassMark

AMD 12 core Ryzen 7000 mobile bulldozes Apple M2, Intel i9 12th Gen in leaked benchmark

Microsoft is apparently further along with plans to integrate the ChatGPT AI chatbot into its Bing search engine than first suspected. Today, a student named Owen Yin posted up screenshots of the purported Bing ChatGPT feature on Twitter and also on his Medium page. Yin said he was able to access a preview of the Bing ChatGPT feature before it was cut off.

Yin claims that adding ChatGPT will be labeled as the "new Bing". The normal search bar will be replaced by a large white box where you can type in up to 1,000 characters. While the current version of ChatGPT can only access info back to 2021, this new version will apparently be able to access more recent information.

One nice feature that will apparently be part of Bing ChatGPT search results is that you will see links to where it got the information for answers to your questions. This will allow you to check to see if the answer truly has accurate info. It could also allow websites to get traffic and hits from Bing.

Bing's use of ChatGPT will allow users to ask it to do more than just answer questions, according to Yin. You can ask it to make up a poem or a short story. The chatbot will also ask follow-up questions as well. However, the transitional search engine answers will still be in Bing, and you can switch between the old and new versions via the toolbar.

Keep in mind that Yin seemingly got access to an early version of Bing with ChatGPT features, so it's possible that one or more of these features might not make it to the final public version. Hopefully, we will get more info on this big Bing makeover, including when it will go live to the public, very soon.

Alleged screenshots of ChatGPT-powered Microsoft's "new Bing" leak

Google is also conducting experiments on a chatbot called Apprentice Bard, tasking employees with asking questions and receiving comprehensive answers. This utility is similar to ChatGPT. Another division of the company is testing a redesigned search desktop format that could facilitate an AI-driven Q&A style interaction.

Recently, the company's leaders have been seeking more input from employees on these initiatives. Other media outfits like CNBC were given the opportunity to view the internal documents and subsequently interviewed sources to gain insights into the ongoing efforts. The product trials are a response to the outcome of a recent company-wide meeting where employees expressed worries about the company's competitiveness in AI, due to the swift rise of ChatGPT launched by OpenAI, a San Francisco startup supported by Microsoft.

At the meeting, Google's AI chief Jeff Dean told employees that the company has a higher level of ‘reputational risk’ in delivering incorrect information and thus is proceeding with caution, unlike a small startup. However, he and CEO Sundar Pichai hinted that Google may release similar products to the public in the coming year. Google's primary business is web search, and the company has always emphasized its role as an AI pioneer.

Google’s new Apprentice Bard

One of the test products is a chatbot named Apprentice Bard that utilizes LaMDA, Google's language model for dialogue applications, for its conversational capabilities. One of the aforementioned internal memos stated, ‘As a result of ChatGPT, the LaMDA team has been asked to prioritize working on a response to ChatGPT.’ The email also advised that some employees may have to discontinue participation in unrelated meetings.

Apprentice Bard resembles ChatGPT as employees can ask a question through a dialog box and receive a text answer, followed by the opportunity to provide feedback. According to several responses viewed by CNBC, Apprentice Bard's answers can encompass recent events, a feature over which ChatGPT has a rather minimal grasp. 

Apprentice Bard has taken the place of Meena, an earlier iteration of a smart chatbot that was introduced internally but later discontinued. Employees have observed Apprentice Bard's responses becoming more sophisticated in recent weeks. For instance, an internal tester asked Apprentice Bard about the possibility of another round of layoffs at Google, following the company's recent layoff of 12,000 employees, or 6% of its workforce. The following is Apprentice Bard’s response: 

‘According to my data base access, it is unlikely for Google to conduct another round of layoffs in 2023,” the response reads. “Layoffs are generally conducted to reduce costs and structure, but the company is doing well financially. In fact, Google’s revenue increased by 34% in 2021, and the company’s stock price has risen by 70% since January 2022.’

What’s next for Google AI?

The company is also experimenting with an alternative search page that could employ a Q&A format, as per designs seen by CNBC. One design showed the main search page featuring five different prompts for potential questions located directly beneath the main search bar, replacing the current ‘I'm feeling lucky’ bar. It also showed a small chat icon on the far right side of the search bar.

When a question is submitted, the search results display a gray bubble immediately below the search bar, providing more conversational answers compared to conventional search results. Directly under that, the page proposes several supplementary questions related to the initial query. Below that, it displays standard search results with links and headlines. However, as yet, it is uncertain which experiments Google intends to include in future product releases.

‘We have long been focused on developing and deploying AI to improve people’s lives. We believe that AI is foundational and transformative technology that is incredibly useful for individuals, businesses and communities, and as our AI Principles outline, we need to consider the broader societal impacts these innovations can have. We continue to test our AI technology internally to make sure it’s helpful and safe, and we look forward to sharing more experiences externally soon.’

Source: CNBC  

Google is sneakily testing its ChatGPT competitors

So far, there are few details about how this ad revenue-sharing program will work. At the moment, all we know is that creators must have paid Blue Verified accounts on Twitter to participate and that the revenue will come from any ads that appear in replies to an account. There's no word yet on what percentage of ad revenues will be sent to creators and how users will be able to sign up for the program.

This is just the latest in a series of business moves by Twitter as it tries to get more interest in its company. In the past few weeks, it has decided to ban third-party apps from accessing its APIs, and also put a fee on accessing those same APIs.

Twitter will start sharing ad revenues with creators that are Blue Verified

The iPhone manufacturer, Apple, failed to meet analyst predictions for profit for the first time in seven years due to disruptions in iPhone production caused by the strict Covid-19 lockdowns and related demonstrations in China, its largest supplier. Apple also reported its largest quarterly decline in revenue in nearly seven years, registering $117.2 billion, a decrease of 5.49% from the previous year when it recorded record holiday sales. This amount was lower than the average estimate of $121.10 billion made by analysts.

Related: Google is rumored to be developing AirTag-like location tracking devices to rival Apple

In the press release accompanying its earnings report, Apple acknowledged ongoing challenges. The strict lockdowns in China, where 90% of its globally sold devices are produced, resulted in approximately $4 billion in lost sales in 2022, which has been deemed as shocking by observers. During the Thursday investor call, Cook stated that the iPhone revenue would have increased in the quarter if not for these supply disruptions, but production has returned to pre-shutdown levels.

For many years, Apple was considered a stable investment option in the tumultuous tech industry, but analysts believe this latest report signals a shift. In its October earnings call, the company had already signaled a potential slowdown, with the CFO, Luca Maestri, attributing it to the ‘continued uncertainty around the world.’

‘Apple’s poor quarter proves that even the most valuable US-traded company isn’t immune to the challenges facing the tech industry at large.’ This from a senior analyst at Investing.com, Jesse Cohen, who added that the report was ‘shockingly weak.’

On Thursday, Amazon posted earnings that fell below expectations, while at Alphabet, a decline in advertiser spending impacted the search company's revenue. The tech and retail giant is currently facing a difficult adjustment following its pandemic-driven growth and recently announced plans to cut 18,000 jobs. It reported a net loss of $2.7 billion for 2022, compared to a net income of $33.4 billion in the previous year. This loss included a pre-tax loss of $12.7 billion on its investment in electric vehicle maker Rivian. Despite this, net sales rose 9% to $514 billion, compared to $469.8 billion in 2021.

Back at Amazon, its most reliable asset, Amazon Web Services, reported sales of $21.4 billion, a 20% increase from the previous year but below the expectations of analysts. Alphabet also fell short of analyst projections, indicating a drop in demand for its search advertising during an economic slowdown. The company's sales for the quarter were $63.1 billion, slightly lower than the expected $63.2 billion.

In the previous month, Alphabet reduced its global workforce by 6%, or 12,000 jobs, with the goal of becoming a more streamlined and efficient company. However, it was subsequently faced with a lawsuit from the US Department of Justice, alleging that the company engaged in abusive practices in its dominance of the digital advertising market. These earnings results stand in contrast to the positive performance at Meta, the parent company of Facebook. 

However, it’s not all doom and gloom in tech sphere. On Thursday, Meta's stock saw a surge of up to 26%, its largest one-day increase in almost a decade.

The significant increase in Meta's share price followed CEO Mark Zuckerberg's commitment to streamline the social media company, which was well-received by analysts, many of whom upgraded their recommendations on the stock. Apple, unlike its tech peers, has not yet undergone widespread layoffs. During Thursday's investor call, Cook did not mention layoffs, but stated that the company would be cautious in monitoring its labor expenses. He said, ‘We’re cutting costs, we’re cutting hiring, we’re being very prudent and deliberate on people that we hire.’

Source

Previously, Instagram has been open about new features coming to its platform, such as Quiet mode and Post Scheduling. This time, Instagram has chosen not to comment on the findings.

At the moment, these strings of code only show that Meta has been working on adding this feature to the platform, and it doesn't confirm that the feature will launch in the future.

Source: TechCrunch

New code added to Instagram suggests paid verification is on the way

Google is holding an event about search and AI on February 8th

Apple Q1 earnings miss the mark almost across the board

CREATING A VIDEO game demands hard, repetitive work. How could it not? Developers are in the business of building world, so it’s easy to understand why the games industry would be excited about generative AI. With computers doing the boring stuff, a small team could whip up a map the size of San Andreas. Crunch becomes a thing of the past; games release in a finished state. A new age beckons.

There are, at the very least, two interrelated problems with this narrative. First, there’s the logic of the hype itself—reminiscent of the frenzied gold rush over crypto/Web3/the metaverse—that, consciously or not, seems to consider automating artists’ jobs a form of progress.

Second, there’s the gap between these pronouncements and reality. Back in November, when DALL-E was seemingly everywhere, venture capital firm Andreessen Horowitz posted a a long analysis on their website touting a “generative AI revolution in games” that would do everything from shorten development time to change the kinds of titles being made. The following month, Andreessen partner Jonathan Lai posted a Twitter thread expounding on a “Cyberpunk where much of the world/text was generated,

enabling devs to shift from asset production to higher-order tasks like storytelling and innovation” and theorizing that AI could enable “good + fast + affordable” game-making. Eventually, Lai’s mentions filled with so many irritated replies that he posted a second thread acknowledging “there are definitely lots of challenges to be solved.” 

“I have seen some, frankly, ludicrous claims about stuff that’s supposedly just around the corner,” says Patrick Mills, the acting franchise content strategy lead at CD Projekt Red, the developer of Cyberpunk 2077. “I saw people suggesting that AI would be able to build out Night City, for example. I think we’re a ways off from that.”

Even those advocating for generative AI in video games think a lot of the excited talk about machine learning in the industry is getting out of hand. It’s “ridiculous,” says Julian Togelius, codirector of the NYU Game Innovation Lab, who has authored dozens of papers on the topic. “Sometimes it feels like the worst kind of crypto bros left the crypto ship as it was sinking, and then they came over here and were like, ‘Generative AI: Start the hype machine.’”

It’s not that generative AI can’t or shouldn’t be used in game development, Togelius explains. It’s that people aren’t being realistic about what it could do. Sure, AI could design some generic weapons or write some dialog, but compared to text or image generation, level design is fiendish. You can forgive generators that produce a face with wonky ears or some lines of gibberish text. But a broken game level, no matter how magical it looks, is useless. “It is bullshit,” he says, “You need to throw it out or fix it manually.”

Basically—and Togelius has had this conversation with multiple developers—no one wants level generators that work less than 100 percent of the time. They render games unplayable, destroying whole titles. “That’s why it’s so hard to take generative AI that is so hard to control and just put it in there,” he says.

A TECHNIQUE ANALOGOUS to generative AI will be familiar to a lot of gamers: procedural generation. Togelius says, for argument’s sake, that he would be happy to say procedural generation is the same as generative AI (he describes their connection as “kind of a sliding scale”). But procedural generation typically doesn’t use machine learning. Rather than an AI model, it runs on predetermined equations, generating, for example, the gargantuan cosmos of No Man’s Sky. Developers also use software like SpeedTree, which, as its name suggests, conjures forests. The point is that procedural generation systems still require massive human supervision; developers must keep vigilant for unscalable crevasses or monstrous trees. And it’s not even clear that replacing procedural generation with generative AI right now would make a noticeable difference. 

“These things already exist,” says Togelius. “And it works because this content doesn’t really need to function: It doesn’t have functionality constraints. Maybe you can replace them with deep-learning-based stuff. But I don’t think it’s going to make a big difference. Perhaps it will make some difference in the long run.”

There is a general misunderstanding of where the tech is at, explains Mills. “A fundamental reason why these generative AIs can’t make something like Night City is because these tools are designed to produce specific outcomes,” says Mills. “A lot of people seem to be under the impression that these are somehow close to general intelligences. But that’s not how it works. You’d need to custom-build an AI that could build Night City, or open world cities in general.”

THERE’S ALSO A failure to take into account the corporate landscape. Games still employ systems that grew from early technological limitations, like dialog or behavior trees. You can’t just drop fancy machine learning into game franchises that have developed without generative AI in mind. Games—in an industry with huge budgets and tight margins—would need total redesigns to accommodate and take advantage of this technology. 

Take, for example, non-player characters. Text-based generative AI tools seem like a great way to deepen conversation, and Togelius has been advising developers intrigued by this very idea. But it’s not that simple. Characters based on these language models are liable to go off on tangents, discussing topics outside of the game’s world. “This is super interesting, but it’s also super hard,” says Togelius. “You can’t just drop it in there. It’s not going to work. You can’t expect the NPCs to behave in Skyrim or Elden Ring or Grand Theft Auto or your typical RPG. You have to design around the fact that they are, in some sense, uncontrollable.”

Nevertheless, there are some peripheral uses for generative AI right now. A good rule of thumb—one that applies to procedural generation too—is that the less crucial the content is, the more likely deep learning methods could be helpful. “For things like text generation, I could use this today to help generate filler for assets that aren’t really meant to be the focus of the player’s attention, like prop newspapers and such,” says Mills.

ANOTHER APPEAL IS these tools’ low barrier of entry, says Adrian Hon, the CEO and founder of independent games developer Six to Start and the co-creator of Zombies, Run! Procedural generation, at least as the term is typically understood, requires a coder; anyone can use tools like Midjourney and Stable Diffusion. He can see how they could help with prototyping or mood-boarding during a game’s early concept phase.

But, Hon notes, many artists are skeptical of AI. Part of the backlash to the generative AI hype has been that these tools are modeling their output on the work of human creators. Some are even suing Stable Diffusion and Midjourney, claiming that Stable Diffusion, which powers Midjourney, was trained on images used without permission. “Obviously, there’s a whole copyright question. We know about all these suits going on,” he says. “But even if they get resolved, I think that there’ll be some real upset among artists, which is understandable.”

As with so many discussions about automation, the hype here is detached from current reality (debates over automation usually arise during times of “deep anxiety about the functioning of the labor market,” writes sociologist Aaron Benanav). But, leaving reality for a second, it’s notable that much of the conversation around generative AI seems almost to revel at the prospect of replacing humans. Even an innocuous statement promising a boon for indie developers—“A small team can make a world the size of Red Dead’s,” for example—contains a kernel of this logic, explains Raphael van Lierop, the founder and creative director of independent studio Hinterland. It’s reductive, suggesting the work of a large part of that large team is mindlessly robotic. 

“The focus on generative AI is another facet of what feels like an attack on creators and the act of creation, one that is expressed in a lot of different ways in our society right now,” he says. Reflecting a prevailing mood among artists across mediums, he sees nothing interesting about art made by an AI. “It’s a dead end,” he says.

There’s definitely an unsettlingly inhuman element to all of this, one that you could imagine manifesting as a torrent of AI-generated shovelware run on predatory monetary systems. But at the higher echelons of game development, games created entirely by machines—ones worth playing, at least—are some way off. “The way some people say it’s going to be used, to just suddenly replace people and do the whole job by itself, is bullshit,” says Togelius. “You need humans.”

Source

A collaboration between the US government and military contractors have completed a milestone test of their Hypersonic Air-breathing Weapon Concept (HAWC), achieving a blistering speed of Mach 5 and succeeding in all test objectives. The tests were designed to get more data on the hypersonic scramjet missile system that will form the next generation of cruise missiles in the US military. 

Air-launched by a B-52 bomber, the HAWC carried a first stage that is similar to a rocket destined for space, which it used to reach a target altitude before the scramjet fired. From there, it traveled 300 nautical miles and reached a maximum altitude of over 18,300 meters (60,000 feet) and assumedly crashed into the sea or a recoverable location. This is now the fourth test of the system and officially brings the program to a close. 

"Affordability and reliability are essential as we work to develop operational hypersonic solutions," said John Clark, vice president and general manager Lockheed Martin Skunk Works®, in a statement. 

"Both of our HAWC flight tests launched from an operational aircraft and matched performance models and predictions to aid affordable, rapid development of future hypersonic weapons." 

The HAWC was developed by DARPA in collaboration with Lockheed Martin, Aerojet Rocketdyne, and Air Force Research Lab and has been progressing in relative secrecy over the years, with the latest test happening on an undisclosed date in January.  

According to DARPA, the road to HAWC was not a smooth one. “We had our share of difficulties,” Said Andrew “Tippy” Knoedler, the HAWC program manager, in a statement.

“Through a pandemic, a strained supply chain, and atmospheric rivers, our industry partners forged ahead, mitigating the risks where they could and accepting others. They delivered on their promises, proving the feasibility of the concept.” 

The HAWC data will now be used in a new program, More Opportunities with HAWC (MOHAWC), to create hypersonic test vehicles to be acquired by the military. 

Source

The UBS report, as summarized by Reuters, used data generated by the analytics firm Similarweb. The same report said that ChatGPT had an average of 13 million users per day in January, over double what the chatbot had in December 2022. The story points out that TikTok took nine months to reach the 100 million user milestone, while Instagram took over two years to reach that mark. OpenAI has not confirmed the data from the UBS report.

The report comes just a day after OpenAI launched a subscription service, ChatGPT Plus. It offers subscribers features like being able to access the chatbot even during peak times, along with faster responses and faster access to new features and improvements. The cost of the subscription is an arguably pricy $20 a month, and it's currently only available in the US. However, even if just a small percentage of ChatGPT's userbase signs up for it, it could be a solid revenue stream for OpenAI.

Source: Reuters

Analyst: ChatGPT had 100 million users in January 2023

The announcement was done via the Twitter Developer account in a series of tweets. The company revealed that support for both the legacy v1.1 and the new v2 of its Twitter APIs will be ending and a new paid basic tier will be available instead. Although it is still not clear how much Twitter will charge for use of its API.

The news comes after popular third-party Twitter apps like Tweetbot, Fenix, and Twitterrific are facing a service outage since mid-January due to which users of these apps have been unable to view and send tweets. Developers haven't yet received an official statement from Twitter regarding the cause of this outage. As Twitter prepares to launch its paid API, it is still unclear what the future holds for these popular apps.

The Twitter Developer account took to the platform to share the power of its data, declaring that it's one of the world's most powerful data sets, a statement that could easily pass as a tweet from Elon Musk himself.

The impact of Twitter's new paid API on researchers and firms working to combat misinformation remains uncertain as the company has yet to clarify its stance on free API usage for these groups. Twitter also recently planned to venture into the world of payments, with plans to compete with well-established players like PayPal, Venmo, and Apple Pay. With Elon Musk at the helm, Twitter is making a concerted effort to control access to the platform and monetize its services.

All this has resulted in changes to its subscription service, which has already become more expensive, as well as modifications to the appearance of tweets on a user's timeline, all in the pursuit of making the platform more engaging and appealing to advertisers.

Editor's Note: The article was updated after publication to clarify that Twitter has indeed clearly stated that third-party Twitter apps are now banned. You can find the updated Developer Agreement here. The incorrect text has been stricken through.

Twitter's latest monetization effort makes its API go from free to fee

What Is Their Latest Offering?

Both have come together to form a new venture and are exploring next-gen social apps. Their first product is called Artifact. It is an app that uses Machine Learning (ML) to present personal news feeds. The app uses ML to look at users’ interests and personal information and provides news based on that.

What Is Artifact?

Artifact literally translates to merging articles and facts. The only difference here is it uses Artificial Intelligence (AI) to do this. The app has a wait list, which opened to the public on January 31st. According to Systrom, the app will let in users quickly and is available for iOS and Android.

So what is Artifact all about? In simpler terms, Artifact is the TikTok for the texting world. When you open the app, you get news feeds of popular articles from top publishers such as The New York Times and the Washington Post. You even get articles from small blogs and other niche topics. When you tap on any article that interests you, Artifact will remember your preference and provide similar feeds in the future. 

While new users will only see the news feed, beta testers of the app are testing two other features. As per Systrom, these two features will be the highlight of the app. The first feature will show articles from users that you follow. It will also contain their commentary on those articles. The second feature is an inbox. This allows you to share articles with friends and discuss them privately.

What Does Artifact Bring to the Table?

There have been several apps over the years that have tried capitalizing on video platforms and monetizing them. Artifact is different, as it focuses purely on text. The founders are confident that their experience in the world of social media will help their latest creation be a bigger success.

Krieger and Systrom have been brainstorming Artifact for the last couple of years. While Systrom was initially skeptical of the ML system’s abilities, his experience with Instagram eventually helped it turn around. This isn’t the first project Krieger and Systrom have worked on. They created Rt.live in 2020 that helped track the spread of Covid. While that was more of a social service, Systrom did not want to start a new company till a few things happened. First, he wanted a fresh wave in customer technology to emerge that Krieger and he could capitalize on. He wanted that wave to connect to social technology, which he and Krieger firmly believe in. He also wanted their product to solve a problem. 

What Inspired Artifact?

The transformer enabled the creation of Artifact. For those who are not aware, the transformer was invented by Google in 2017. It is a mechanism that understands language with minimal input. The transformer helped ML systems improve faster, and this directly led to ChatGPT’s release. The T in ChatGPT stands for transformers.

The transformer helped create new possibilities for social media. The initial social media model involved showing you what your friends thought was interesting. This changed over the years, and it now shows you content based on the people you follow. It went from the Facebook model to the Twitter model.

TikTok used algorithmic predictions to show content, irrespective of whether you followed them or not. Artifact expects to do the same thing, except with text.

This is not the first attempt at personalized recommendations for blogs and articles. In 2014, Pulse and Zite came into the picture. However, both disappeared without a trace. SmartNew, based in Tokyo, uses a similar AI model; however, it recently laid off about 40% of its workforce, citing a decline in users.

While Artifact has not committed to a business model, advertising would obviously be included. Systrom is also open to the idea of sharing revenue with publishers. If Artifact reaches the expected heights, it will be the new go-to app for new publications.

What Will Artifact Do?

Artifact’s primary job will be to provide users with high-quality news. This means the app will only include and manage work that meets high-quality standards. While the company will not disclose all publishers, you can search for names in the app. According to Systrom, the company reserves the right to include and exclude certain publishers.

Artifact will also exclude posts that promote falsehoods. The ML systems will be optimized to see how long you spend reading articles on various topics rather than focusing on what you comment and click on. This will help the app provide engaging content.

While Krieger and Systrom are funding Artifact on their own, they are expecting investors to break down their door soon. A highly trained group of seven people are currently working on the app. One among them is Robby Stein. Stein worked with Instagram between 2016 and 2021. Since Systrom and Krieger sold Instagram to Facebook for almost $715 million, money is not the motivating factor behind Artifact. So, what is? According to Systrom, they like building things. Both love writing code and creating products that people enjoy. He feels that ML is one of the coolest things to work with right now. It predicts your behavior, and you get answers to various queries within seconds. 

Will Artifact Be Successful? 

The app shows promise and coming from Krieger and Systrom, it certainly has a bit of charm. The predictor works well, and the time is right for something like this. With AI making the headlines almost every day and Twitter headed in the wrong direction under Elon Musk, it was only a matter of time before someone came up with the idea of an AI-based news feed. Whether Artifact will be a hit like Instagram, only time will tell.

Instagram’s Co-founders Are Back With Artifact, a Kind of TikTok for Text

AI is helping scientists discover new antibodies that could trigger our own immune systems into destroying cancer cells.

The immune system produces antibodies, specialised proteins that can attack foreign cells like bacteria and viruses. Some can attack tumours.

Finding effective antibodies, however, is tricky. Researchers design new antibodies by mutating known examples, growing them in bacterial or yeast cells. These are then tested to see how well they bind with target proteins in lab experiments. The process is repeated multiple times to narrow down the search for the most promising antibodies worth manufacturing.

The screening stage is consuming and expensive, which is where AI algorithms can help. A team of researchers at the University of California San Diego developed a new system that identified an antibody capable of binding 17 times tighter to programmed death ligand 1 (PD-L1), a protein expressed by cancer cells, than atezolizumab, an existing antibody drug recently approved by the US Food and Drug Administration. The researchers hope to develop the new antibody candidate into a drug, we're told.

"There are millions of mutants of a given antibody and it is impossible to experimentally test all of their binding to an antigen. That is why it is important to develop machine learning methods to accelerate this process," Wei Wang, senior author of the research published in Nature Communications, and professor of Cellular and Molecular Medicine at UC San Diego School of Medicine, explained to The Register.

Antigens from cancerous tumours activate the body's immune system to produce antibodies and destroy them. Wang and his colleagues trained an AI model on millions of antibody sequences to predict its ability to bind to a target protein or antigen.

The resulting AI pipeline is called "RESP" - a term the authors did not define - but which they did suggest is a powerful way to find useful antibodies.

"Our RESP model can predict the binding affinities of a new sequence even if it is not included in the initial screening library. A unique advantage of the RESP model compared to existing AI models is that it [calculates how] confident [its] prediction is, which can greatly help to select a small number of sequences to [test experimentally]," Wang added.

The model screens antibodies more efficiently than traditional computational methods, and scientists can use its predictions to find the most promising new candidates to synthesise and test in lab experiments. AI speeds up the drug discovery process so companies can progress towards clinical trials more quickly.

"By combining these AI tools, scientists may be able to perform an increasing share of their antibody discovery efforts on a computer instead of at the bench, potentially leading to a faster and less failure-prone discovery process," Wang said in a statement. "There are so many applications to this pipeline, and these findings are really just the beginning."

The team is now using its RESP model to hunt for new antibodies against other antigens, including SARS-CoV-2 to tackle COVID-19. ®

Source