Latest News - Page 920

The Northern Lights, commonly spotted in the Arctic, made its way as far south as Arizona last night, stunning onlookers that happened to catch a glance at the rare sighting.

Over March 23 and March 24, Earth was rocked with an unexpected Coronal Magnetic Ejection (CME) from an eruption that occurred on the Sun. The National Oceanic And Atmospheric Administration (NOAA) issued a warning for G4 (severe) geomagnetic storms across the entire North American region, writing in its warning that the following pieces of technology may be affected; power grids, spacecraft operations, atmospheric drag on satellites operating in Low Earth Orbit (LEO) and GPS operations.

Spaceweather.com reports that as of March 25, the origin of the unexpected CME blast remains unknown, but their suspicions lie with the near-miss March 23 CME that may have caused a ripple effect that hit Earth. With the impact of the CME comes auroras as the charged particles from the Sun interact with Earth's atmosphere and magnetic field, causing the incredible phenomena we call auroras. Auroras typically occur close to Earth's poles, but during a geomagnetic storm, the phenomena can move closer to Earth's equator.

It was only a few days ago that a team of Stanford researchers built a clone of OpenAI's ChatGPT for just $600. Now those researchers have taken the demo.

The release of OpenAI ChatGPT put artificial intelligence in the limelight and demonstrated the widespread demand for language models, the underlying technology powering these AI tools. Following the release of ChatGPT, which quickly adopted millions of users, other companies such as Google, Microsoft, Facebook, and Amazon began dropping information about their own language models that are currently in development. Microsoft quickly hopped on the OpenAI train by investing billions of dollars into the company in return for its propriety GPT language model.

Researchers at Stanford decided to see how difficult and costly it would be to create their own language model and decided to try and replicate OpenAI's GPT. As previously reported, the Stanford professors took Meta's open-source LLaMA 7B language model and trained it on trillions of tokens of data. The results were the almost-creation of a ChatGPT clone named Alpaca, which came with some key differences. While Alpaca was trained on a large amount of data, it wasn't trained on how to sift through that data to acquire an answer quickly. This led researchers to conclude that Alpaca was much slower than GPT.

Internal Microsoft researchers have published a paper that explored OpenAI's newest language model, GPT-4, and according to those researchers, it may be the very first step toward Artificial General Intelligence (AGI).

A new research paper has been published on the arXiv pre-print server by internal Microsoft AI researchers that explored the capabilities of OpenAI's newest language model called GPT-4. According to the team behind the paper, an analysis was conducted on an early iteration of the GPT-4 language model, and based on the results, the researchers believe that GPT-4 could be viewed as an early and incomplete version of AGI. Microsoft's testers write that GPT-4 shows a clear leap in several areas compared to OpenAI's previous model, GPT-3.5.

The team found that GPT-4 achieved close to human-level performance in a range of categories that its previous generation lacked. Those categories were mathematics, coding, vision, medicine, law, psychology, and more. According to the paper, GPT-4 also performed exceptionally well in several exams, scoring, respectively, in the 90th, 88th, and 86th percentiles on the Bar exam, LSAT, and Certified Sommelier theory test.

Epic Games has revealed at its State of Unreal keynote at GDC 2023 its new MetaHuman creator tool that will enable developers to animate their MetaHumans with just an iPhone.

The announcement featured the reveal of the above video that, demonstrates the ease of developers using video footage shot with their iPhones to animate characters. The results are very impressive, with the animation tool showing realistic movements in subtle facial expressions, eye tracking, and graphics quality.

Epic showed off what's possible with the technology by releasing a clip of Ninja Theory's game Senua's Saga: Hellblade II, in which performance capture artist Melina Juergens that plays the game's main character, acts out a scene that is then transformed into a gorgeous animation. The Senua's Saga: Hellblade II technology demonstration was created using expensive motion capture cameras within a studio, hence the impressive end result.

Over the past couple of days, the big news was that Linus Tech Tips, a huge YouTube channel with over 15 million subscribers, was hacked. The hackers were able to change the channel name and live stream a fake Elon Musk video trying to get viewers to send Bitcoin to them.

Thankfully, Google helped Linus and his team recover his channel. Still, I can only imagine it would have been an incredibly stressful and nightmare-type situation while the recovery was in process. And it's not just Linus Tech Tips that was hacked. A few weeks ago, Andy from eTeknix suffered the same fate after being conned by a fake video sponsorship scheme where the victim is tricked into opening what appears to be an agreement PDF - the "PDF" file is the malware.

Once opened, the malware sends the user's data to the hacker. It does not matter how strong your password is or if you have enabled two-factor authentication. It's not entirely clear which data is sent, but the critical data we know that is sent includes the user's browser data, including actively logged-in session tokens and cookies. Once obtained, the hacker can carefully plan an attack on the unsuspecting victim, usually when they are asleep.

Blizzard is looking into reported issues with NVIDIA's RTX 3080 Ti, and possibly other GPUs, in the Diablo IV beta.

This is clearly something to bear in mind for PC gamers who own an NVIDIA graphics card - especially that model - who are planning to hit the open beta, which has just kicked off.

In some cases, there are complaints that the beta test of the action RPG has bricked RTX 3080 Ti cards, with other reports coming in of GPUs overheating to alarming levels (and fans kicking into super-high gear, as you might imagine).

The Competition and Markets Authority (CMA) has issued a statement which takes a more positive attitude towards Microsoft's Activision Blizzard merger.

This is a huge move for the gaming world - a $68.7 billion takeover for Microsoft - so the regulatory scrutiny is intense, but the CMA has let go of some of its concerns about the deal.

Based on reviewing new evidence, the regulator is now satisfied that the Microsoft takeover won't damage the console gaming market in the UK in terms of competitiveness.

Microsoft's idea for a 'Cloud PC' feature appears to be progressing, at least going by hidden clues unearthed in a Windows 11 test build.

This comes courtesy of well-known Microsoft leaker Albacore, who has been peeking under the hood in the recent Windows 11 build 23419, and found that the feature is being fleshed out in the background.

There's now an option to access the Cloud PC in two different ways - you can do so either directly from the desktop, or via an app.

Bing search traffic is up appreciably, with Microsoft's new ChatGPT-powered Bing AI successfully helping to boost search numbers in the battle against Google.

The Bing chatbot may not have been on the scene for long, but the early promising result for Microsoft's Bing search engine is that it's up almost 16% with page visits.

That's according to stats compiled by analytics outfit Similarweb (spotted by Reuters), with the sizeable uptick happening since the AI was introduced to Bing search on February 7. Meantime, Google has slipped by 1% in terms of its numbers.

Jackson Greathouse Fall, a brand designer, has created the Hustle GPT challenge, where he acts as an instrument to the instructions of ChatGPT on how to start a business with just $100.

Fall explains in a now-viral Twitter thread that he told ChatGPT that it was now called "HustleGPT" and that it's an "entrepreneurial" artificial intelligence that's goal is to take the initial $100 budget and turn it into as much money as possible within the shortest timeframe possible. Fall added some needed caveats to the prompt, informing ChatGPT that it wasn't able to provide any methods that would involve doing anything illegal and that Fall was going to "act as a liaison" between the AI and the required physical work.

Additionally, HustleGPT was told by Fall that it would do "everything" it says and that he would update the AI on the current cash total of the business as time went on. Following that prompt, Fall explained that HustleGPT told him to set up an affiliate marketing site that makes content around eco-friendly/sustainable living products. HustleGPT provided the steps on how to get started with this business, instructing Fall to buy a domain, hosting, leveraging social media to acquire leads, website design, and optimizing for search engines. Notably, HustleGPT provided the costs for each of the steps and the grand total spent.

A journalist has received a USB drive in the mail, and it turned out to be a bomb that detonated when they plugged it into a computer at the newsroom.

According to a report from the BBC, the journalist that received the USB drive was just one of many journalists across Ecuador that received similar USB drives. Luckily, other recipients either discarded the drive entirely or the bomb failed to detonate. The Ecuadorean attorneys-general's department confirmed it has opened a terrorism investigation following the explosion and explained that at least five news outlets were targeted.

The government body condemned the actions and said that any attempt to intimidate journalism and freedom of speech across Ecuador is a "loathsome action" that will be punished with "all rigor of justice". So, where did the devices come from? According to the interior minister, Juan Zapata, all of the devices were sent from the same town, with three USB drives being sent to media outlets located in Guayaquil, and two to newsrooms in the capital, Quito.

Researchers have honed the very sensitive instruments equipped to the James Webb Space Telescope on a planet in a distant star system, gathering data that informs researchers on the conditions of the planet.

The lucky planet to be a target of the highly advanced James Webb Space Telescope (JWST) is an exoplanet called VHS 1256 b, which researchers explain in a recent press release, is a massive brown dwarf planet that's located approximately 40 light-years away from Earth and orbits, not one, but two stars over a 10,000-year period. The researchers explain that VHS 1256 b orbits its stars four times further than Pluto orbits our Sun.

This aspect of VHS 1256 b makes it a prime target for Webb as the light from the planet isn't getting mixed up with the star it's orbiting, allowing Webb to gather a much more accurate reading. According to the press release, the higher altitudes of VHS 1256 b's atmosphere reach temperatures of 1,500 degrees Fahrenheit, and within those clouds, Webb detected both large and small silicate dust grains. The smaller grains are comparable to the size of tiny particles that appear in smoke, and the larger grains are similar to extremely hot grains of sand.

Yesterday OpenAI confirmed that ChatGPT was temporarily forced to shut down due to a bug that leaked chat history conversations with other users.

The developers behind the highly popular language model chatbot, ChatGPT, confirmed to Bloomberg that the chatbot was forced to shut down after users began reporting seeing email addresses and phone numbers in the ChatGPT Plus payment section of the website.

Users that were signing up for OpenAI's premium ChatGPT service, which features the newly released and next-generation GPT-4 language model, were seeing email addresses and mobile phone boxes being auto-filled with information that they didn't recognize, leading to the assumption which was promptly confirmed by OpenAI, that the auto-filled information was other ChatGPT user contact details. Additionally, some other users reported seeing the title of conversations appearing in their chat history.

NASA partner Axiom Space has revealed its next-generation spacesuit scheduled to be the safety barrier between the harsh lunar landscape and the NASA astronauts that will be walking on its surface.

NASA's Artemis 3 mission, scheduled to launch in December 2025, plans on sending the first astronauts back to the lunar surface in 50 years, and those astronauts will be wearing Axiom Space's Exploration Extravehicular Mobility Unit (xEMU). This next-generation spacesuit was developed under NASA's Exploration Extravehicular Activity Services, or xEVAS, contract, which also includes Collins Aerospace, which will be developing its own spacesuit that'll replace the current suits used by astronauts on the International Space Station (ISS).

As I explained in my previous article about these spacesuits, on March 15, Axiom Space revealed its first iteration of the new spacesuit design by having someone walk out on stage wearing it. The new suit comes with a range of improvements compared to Apollo spacesuits, which have been mostly the same for more than 40 years.

UPDATE: It looks like the Linus Tech Tips channel has been restored, with the channel re-appearing literally moments after this story went live. Here's the original article.

Popular and long-running tech-focused YouTube channel Linus Tech Tips (and those associated with the Linus Media Group) was hacked overnight and used as a platform to promote crypto scams. After a few tumultuous hours, the account was ultimately delisted.

The channel is still down, though a statement released by channel owner Linus Sebastian notes that "we are now on top of it with Google's team now." Adding that "we are getting to the bottom of the attack vector with the (hopeful) goal of hardening their security around YouTube accounts and preventing this sort of thing from happening to anyone in the future."

With DDR5 memory becoming a mainstay in the desktop space, especially for new builds, the next-gen high-speed memory is also making waves in the laptop gaming market. And with that, TEAMGROUP has launched a new range of high-performance DDR5 memory specifically for laptop gaming with its new T-FORCE VULCAN SO-DIMM DDR5 Memory range compatible with 13th Gen Intel CPUs.

And these kits are fast, with max clock rates of 5,200 MHz with cooling delivered via an ultrathin graphene heat spreader. This alone makes installation a breeze (as much as it can be regarding gaming laptops), offering up improved speeds and cooling.

Capacities for the kits go up to an impressive 64GB too, which makes these an excellent option for high-end design and content creation in addition to playing the latest games. Kits will be available in 16GB and 32GB capacities.

The new ASRock DeskMini 4205 might be a Mini PC designed for commercial use thanks to its all-in-one fanless and efficient design powered by the Intel Celeron Processor 4205U. Still, the optional RGB add-ons make it a great home theatre option.

Though on that front, this Mini PC is limited to DisplayPort 1.2, but that's still capable of 4K 120 Hz with 10-bit color.

The ASRock DeskMini 4205 features dual LAN ports, support for three displays, and four bits of storage - two M.2 2280 PCIe slots and two 2.5" SATA 6Gbps. Three simultaneous displays, as per ASRock, make it a great fit for "digital signage and advertising."

OLED, it's what's on the menu - at least when it comes to new high-end gaming monitors like the new CORSAIR XENEON 27QHD240 OLED. A new 27-inch gaming monitor with a resolution of 2560x1440, a super-fast 240 Hz refresh rate, and all of the benefits of an OLED panel.

On that front, CORSAIR notes that the new XENEON 27QHD240 is built using a new third-generation OLED panel with META Technology that features a Micro Lens Array (MLA). A bunch of tech jargon for some, but the result is enhanced brightness via a focal lens placed in front of every pixel leading to a peak brightness of an impressive 1,000 nits.

On top of that, there's HDR 10 support, a 1,500,000:1 contrast ratio (OLED contrast is essentially infinite), and 98-% DCI-P3 color for a vibrant image. Throw in a 0.03ms response time, NVIDIA G-SYNC, and AMD FreeSync Premium compatibility, and this new CORSAIR gaming monitor is undoubtedly beast-level.

The new GeForce Game Ready 531.41 WHQL driver can now be installed via the GeForce Experience app or directly from NVIDIA. If you plan on jumping into this weekend's Diablo IV Open Beta, it's big.

With the Early Access beta from last weekend being one of the most talked about gaming experiences for 2023, Diablo IV is shaping up to be one of the year's biggest releases. The Open Beta kicks off on March 24 via Battle.net on PC, and the GeForce Game Ready 531.41 WHQL driver adds DLSS 2 enhancements for those rocking a GeForce RTX graphics card.

As per NVIDIA, with DLSS 2, players will "experience Diablo IV at the highest detail levels and resolutions," which is awesome to see, as normally, Betas for games don't warrant driver support. At launch on June 6, Diablo IV will also make use of DLSS 3 and Frame Generation, with ray-tracing support also on the cards.

NVIDIA has announced that its GeForce Now game streaming service now boasts over 1,500 games, with five new offerings having just been added this month.

That said, it has been something of a slow road to reach this milestone, when you consider that NVIDIA managed to hit the 1,000 games mark back in July 2021, which was a year and five months on from the official release of the streaming service.

That means it has taken the best part of two years to bring on board 500 extra games on top of the first thousand. Nevertheless, GeForce Now is clearly progressing, just not as fast as some gamers would ideally like.