TikTok has started inviting users in Australia, Mexico and Singapore to participate in a closed beta test for its new music streaming service, according to TechCrunch and CNBC. The short-form video hosting app initially launched beta testing for its fledgling streaming service in Brazil and Indonesia in early July. Now, it's expanding the scope of its music service's experimental phase and giving invited users in those regions a free three-month trial to be able to try it out. 

TikTok Music is a completely separate app that testers will be able to download from the Apple App Store or the Google Play Store. It does, however, connect to the main TikTok app, so users can find the full versions of songs that go viral on the video-sharing platform. The music streaming app reportedly offers personalized song recommendations, real-time lyrics, collaborative playlists and the ability to find songs through a lyrics search feature, as well. TechCrunch says it has a Shazam-like feature, which presumably means it can find songs by listening to it, and will let users download tracks for offline listening. 

The ByteDance-owned app told TechCrunch that once the testers' trial period is done, it will cost them AUD12 (US$8.16) per month in Australia, Mex$115 (US$6.86) in Mexico and S$9.90 (US$7.48) in Singapore to be able to keep using the service. TikTok already has a music streaming service called Resso available in India, Brazil and Indonesia, but it's shutting the app down in the last two countries in September. The company has yet to announce if and when its music app is also coming to the US, but it did file a trademark application for "TikTok Music" in the country back in May 2022. 

Fervo Energy says it has achieved a breakthrough in geothermal technology. It carried out a 30-day well test at its site in northern Nevada and says it was able to achieve a "flowrate of 63 liters per second at high temperature that enables 3.5 megawatts of electric production." The company says the test resulted in flow and power output records for an enhanced geothermal system (EGS) and that it was completed without incident.

A megawatt can power around 750 homes at once. Fervo is expected to connect its Project Red site to the grid this year. It will be used to power Google data centers and some of the company's other Nevada infrastructure. Google and Fervo signed an agreement in 2021 to develop a "next-generation geothermal power project."

This is the first time an energy company has shown that an EGS can work on a commercial scale, according to Bloomberg. It's been a long road to reach this point, as scientists have been trying to make EGS a reality since the 1970s.

For a natural geothermal system to produce electricity, it needs a combination of heat, fluid and rock permeability, as Bloomberg notes. In many areas, the rock has the required levels of heat, but not enough permeability for fluid to flow through it.

An EGS creates this permeability artificially by drilling deep underground and injecting fluid to create fractures in the rock. That approach can vastly increase the number of potential sites for a geothermal power plant.

Fervo says it's the first company to "successfully drill a horizontal well pair for commercial geothermal production, achieving lateral lengths of 3,250 feet, reaching a temperature of 191°C, and proving controlled flow through rigorous tracer testing."

One of the major advantages of geothermal power plants is that they're entirely carbon free — Google aims to run all of its offices and data centers on carbon-free energy by 2030. These plants can also operate at any time (unlike solar and wind), making geothermal energy a compelling source of renewable power. However, cost reductions and regulatory red tape are barriers to making EGS more broadly available, according to Fervo CEO Tim Latimer.

The company is hoping to replicate its success at a site in Utah. If Fervo sees similar results there and it successfully implements design upgrades to maximize output, the site is expected to generate enough electricity to power 300,000 homes simultaneously, Latimer said. That's around a quarter of all homes in Utah.

“Achieving our goal of operating on 24/7 carbon-free energy will require new sources of firm, clean power to complement variable renewables like wind and solar,” Michael Terrell, Google's senior director for energy and climate, said in a statement. “We partnered with Fervo in 2021 because we see significant potential for their geothermal technology to unlock a critical source of 24/7 carbon-free energy at scale, and we are thrilled to see Fervo reach this important technical milestone.”

Cricut, a company that makes popular cutting machines for crafting projects, announced a new model today. Although its devices are relatively affordable, the new Cricut Venture is a higher-end $1,000 machine aimed at “incredibly enthusiastic makers and those who sell their crafts.” In other words, this variant is less for occasional hobbyists and more for hardcore makers or small businesses selling their creations somewhere like Etsy.

Cricut Venture is a “large-format” cutting machine that can handle bigger projects — whether physical size or higher volume. The device, the company’s biggest yet, can cut up to 75 feet of repeated images and 12 ft of single images. Cricut says it’s also its fastest machine, cutting up to 25 inches per second. “Cricut Venture is optimized for size and speed,” said Michelle Fishberg, Cricut’s VP of Global Product Management. “You will find the beautiful design and quality Cricut is known for, with increased precision and attention to detail, all of which make cutting on this large-format machine an easy and productive experience.”

Like the company’s other models, the Cricut Venture can handle materials like vinyl, iron-on, cardstock, bonded fabric, leather and poster boards. Engadget’s Terrence O’Brien reviewed the printer-sized Explore 3 in 2021 and found it worked well for projects like vinyl decals and cardstock animals (that he may or may not have ever fully assembled). However, these machines are more limited by the user’s experience with crafts than technology: The company lists banners, engraved metal, debased leather, custom cards, t-shirts and color stickers among the various tasks it can handle.

The Cricut Venture will be available from the company website and retail partners starting on July 25th. Its $1,000 sticker price is for the machine only; Cricut also offers bundles with a stand or materials to help you get started.

Meta has delivered the first significant update to Threads since the social network launched earlier this month, and you might appreciate it if you're still building your contact list. The company's Cameron Roth has detailed an upgrade to the iOS app that adds a Follows tab to the activity feed, making it easier to see who just followed you. You'll have an easier time following people back. Accordingly, you can open your Instagram follower list to see if you're missing anyone.

The update also adds translations for post text, so you'll have more incentive to follow people who speak unfamiliar languages. You can subscribe to unfollowed users to get notifications without crowding your timeline. There are a few basic interface tweaks as well, such as reposter labels you can tap. The app should be leaner and smoother, particularly when loading or scrolling through your activity feed.

Some of the features are enabled server-side, so don't be surprised if they're not all available immediately. Roth says they should be available by the end of today (July 18th). There's no mention of when Android will get an equivalent update, but we've asked Meta for comment and will let you know if we hear back. The Android beta program offers features before they reach the publicly available app, however.

There are still numerous missing features, and the Threads team is aware of it. You can't yet use a chronological feed, direct messages or hashtags. You can't completely remove yourself from Threads without also deleting your Instagram account. And without a full web version, it's not usually practical to use Threads on a computer. This first update shows that Meta is acting on at least some promises, even if it may take a while to address every issue.

Meta has motivation to act quickly. Threads use is declining after the initial spike, and the absence of some features (such as hashtags and a web app) may keep new users away. Greater parity could help sustain interest while Meta considers a European Union rollout and otherwise prepares for a significant expansion.

Remember Twitter Notes? It was supposed to be the feature that let Twitter users write whatever they want, blowing past the typical Twitter character limit. At the time, that limit was only 280 characters. After several updates this year, Twitter Blue subscribers can tweet up to 25,000 characters. Now, CEO Elon Musk has seemingly confirmed that the company is still working on the Notes feature, but is rebranding it to Twitter Articles.

In a reply to a tweet noting the name change, Musk confirms that Articles will be the place to post “long, complex articles with mixed media.” He goes on to say that “You could publish a book if you want.”

Officially announced last year, Notes was pitched as “a way to write longer on Twitter,” and was initially limited to select users in Canada, Ghana, the United Kingdom and the United States. Notably, this was announced before the Elon Musk takeover later that year, but after he announced his intention to buy the company. This is the first time Twitter has acknowledged the feature since then.

As for what this feature will look like, think blog posts but on Twitter. Articles, which will be separate from the main timeline, is an area where users can post long-form content without the typical limitations of a tweet. That means you’ll be able to embed photos, videos and other tweets within an Article. Users can share Articles in tweets, and published Articles will show up in your Twitter profile.

It’s unclear at this time when or if Articles will be available to the public. The feature is still in the experimental stage it has been in since last year, and is only available to a select number of users.

Framework’s modular Laptop 16 is available for preorder in the US, Canada, Germany, France, UK, Netherlands, Austria, Ireland and Australia. The company’s first gaming notebook starts at $1,399 for a DIY system (sans operating system). Pre-built variants start at $1,799.

The machine ships with an AMD Ryzen 7 7840HS or AMD Ryzen 9 7940HS processor; the DIY versions start at $1,399 and $1,699, respectively. You can use it with the processor’s integrated graphics or pay $1,799 or $2,099 for a ready-to-go model with Radeon RX 7700S discrete graphics.

Of course, the company’s modular focus is what differentiates the Laptop 16 from the multitude of same-y notebooks at your local electronics store. The plan is for its expansion bay system to let you upgrade its graphics (and other hardware) over time without buying an entirely new machine. Its top deck, including the keyboard and numpad, is also user reconfigurable. It adds up to an ambitious plan for a gaming laptop, but the company has so far followed through on supporting its less powerful (but also modular) Laptop 13, the first version of which launched in 2021.

Framework

The Framework Laptop 16 machine has a 16-inch 2560 x 1600 display with a 165Hz refresh rate (with FreeSync). The screen can reach 500 nits brightness, and it’s rated for 1500:1 contrast and 100 percent of the DCI-P3 color gamut. The laptop has an 85Wh battery, which Framework says will retain 80 percent capacity after 1,000 charging cycles. (The company says it’s “easy to replace” if needed.) It supports WiFi 6E and Bluetooth 5.2 and has a 1080p webcam with dual mics and hardware privacy switches. It also includes a fingerprint reader, compatible with Windows and Linux.

The company will follow a batch ordering system, and it expects the first round to ship in Q4 2023. Framework says a limited supply of pre-orders that include an eligible AMD Radeon GPU will receive a free download code for the highly anticipatedStarfield: Premium Edition. (CEO Nirav Patel says game codes will arrive before its early access launch.) You can preorder the Laptop 16 now from the company website.

When Amazon introduced its 2nd-generation Echo Buds in 2021, the company redesigned them to address nearly every issue with its first attempt. They had true active noise cancellation (ANC), better sound and a smaller size. Amazon took a much different approach for its third version, opting to move from mid-range to a low-cost model with an all-new design that covers most of the basic features you’d expect. As always, the Alexa faithful will get the most out of the new $50 Echo Buds, but in a lot of ways you could argue a budget model that does well with the essentials is where Amazon should’ve been all along.

Design

While the first two iterations of the Echo Buds had similar designs, Amazon completely changed things up for the third generation. The 2023 model has an AirPods-like stem that extends down from your ear toward your jaw. The silicone ear tips are gone as well and the buds sit in your ear rather than completely closing off your ear canal. There are two sets of silicone “covers” included with the Echo Buds that Amazon says will “optimize your fit.” They don’t really change how the earbuds feel in your ears because they’re so thin. They just provide a tiny bit of grip to help them stay put.

The mostly open design is great for when you might not want to block out the world entirely. However, they’re only IPX2 rated for protection from water droplets. The 2nd-gen Echo Buds offered slightly better IPX4 resistance, so I wouldn’t rely on the 2023 model as your workout partner if you sweat a lot.

Amazon didn’t cut too many corners when it comes to controls to keep costs low, though. The upper, main part of the Echo Buds accept taps (single, double, triple and long press) for music playback, calls and to summon Alexa manually. The earbuds are also equipped with in-ear detection for automatic pausing. You can reconfigure the defaults any number of ways, including swapping out core controls for volume. You’ll need two spots for that though, as volume up is one gesture while down requires another.

The controls are mirrored on both sides out of the box, but you don’t have to keep them that way. You can also reassign the long press to mute the Echo Buds mics or activate your phone’s native assistant. Since these earbuds offer hands-free access to Alexa, you can technically get both. However, the Echo Buds won’t summon Siri or Google Assistant if the mics are muted, even if you’ve assigned them to the long press action. The bottom line here is that there is a lot of customization available and the earbuds reliably accept your taps without having to repeat a touch command.

Despite the longer overall design, the charging case with the new Echo Buds remains small. This one will easily tuck in a tiny pocket. The Bluetooth pairing button is on the outside up front, sitting just below an LED status indicator, and a USB-C charging port resides on the right side.

Software and setup

Photo by Billy Steele/Engadget

One of the biggest issues I have with the Echo Buds is that they require the Alexa app for a full setup. I’m well aware earbuds and headphones need apps to get the most out of their features, but the Alexa app is bloated as it handles all kinds of smart devices beyond just the Echo Buds. There’s a lot to contend with here just to tweak your settings. Amazon has put the earbuds info up top so you don’t have to go looking for it, but there’s still a lot to get lost in down below.

The Alexa app displays battery life percentages for both earbuds individually and the case right on the home screen, as well as an easily accessible mute button for the microphones. Tapping further into the settings shows you Bluetooth connections and gives you the option to disable Alexa hands-free. Note that this doesn’t mute the mics and you can still manually summon Amazon’s assistant with a long press on the Echo Buds. The earbuds also support multipoint connectivity for two devices at once, and in my experience switching is quick and reliable.

As it has in the past, the company allows you to activate and adjust Sidetone. That’s the amount of your own voice you want to hear during a call. VIP Filter returns as well, which lets you decide which contacts and apps you want to receive notifications from. EQ customization comes in the form of three sliders for bass, mids and treble, just above the option to reconfigure the tap controls. Amazon has included a Find My tool in the app and there are feature categories for Entertainment, Communication and Location & Maps that help to further personalize the Echo Buds Setup.

Sound quality

Photo by Billy Steele/Engadget

When it comes to open-type earbuds, sound quality can vary wildly from model to model. It’s understandable that some companies struggle to deliver the best audio without sealing off your ears and the wearer’s ear shape can dramatically impact the results. But Amazon tackled both of those challenges admirably and created a sound profile that’s quite good in a set of $50 buds.

The first thing that impressed me with the Echo Buds was the clarity and detail. Softer and acoustic genres like bluegrass, folk and lighter country sound great, with the warm strums of Zach Bryan’s guitars and vocals on American Heartbreak cutting through the mix. Switch over to the electro-pop-rock of PVRIS’ Evergreen and you’ll notice that these earbuds do lack a bit of low-end thump. That’s common with open designs, so it’s not surprising. If I press the buds in slightly, the bass is more present. Unfortunately, that means audio quality may vary based on how the Echo Buds sit in your ears.

Overall, the tuning is well-balanced across genres. At times, there’s a heavier dose of treble, but it’s not a constant issue. But in general the Echo Buds deliver better audio quality than much of the $50 competition. They pickup plenty of subtle reverb when it’s present and there’s nice texture to drums and distorted guitars across hip-hop and metal respectively. They even adequately render all of the layered and unique sounds of the KAYTRAMINÉ album.

Unlike the previous version of Echo Buds, there’s no active noise cancellation here. Due to the semi-open design it would be mostly wasted anyway. This means the new Echo Buds won’t do much to block a noisy coffee shop or air conditioner, but they should keep you from missing announcements, warnings or any other crucial info that might otherwise go unheard.

Battery life

Photo by Billy Steele/Engadget

Amazon says you can expect up to five hours of use on these Echo Buds. You can extend that figure by one hour if you opt to disable hands-free access to Alexa. There are three additional charges in the case and a quick-charge feature will give you up to two hours of listening time in 15 minutes. The Echo Buds are replenished via the USB-C port on the right side as there’s no wireless charging here. Amazon doesn’t include a cable either, citing its Climate Pledge Friendly commitment, but chances are you already have one. If not, the company offers the option to bundle it for $10 more.

During “regular” use, I still had eight percent left in the tank after five and a half hours. That’s a mix of playing music, taking calls and leaving the buds on my desk outside of the case a couple of times. I also had the wake word active, so I went beyond the company’s stated number. Five hours is near the industry average these days, especially for more affordable models, but it’s nice to see Amazon slightly overdelivering.

Call quality

To assist with calls, Amazon installed two mics and a voice accelerometer on each earbud. As always, companies’ claims about call performance are typically exaggerated to some degree with promises of “crystal clear” quality. That’s exactly what Amazon says about the Echo Buds, but overall clarity isn’t particularly great. Your voice comes across slightly muffled, though it’s perfectly fine for general video and voice calls. I wouldn’t do any recording with these – not that you would with most earbuds anyway. The Echo Buds do a decent job combatting background noise, including a loud fan that came through as soft white noise instead of its raucous rumble and a nearby game of Mario Kart 8 wasn’t picked up at all.

The competition

Photo by Billy Steele/Engadget

At $50, Amazon is significantly undercutting its competition from the biggest names in wireless earbuds. Sony’s cheapest option, the WF-C500s, are $100 and Samsung’s Galaxy Buds 2 are only about $10 less. Apple’s 2nd-generation “regular” AirPods are $129, but there’s a pricier, redesigned model of those that debuted in 2021.

For more apt comparisons, you’ll want to look to the likes of Jabra and Anker for the best budget wireless earbuds. The Jabra Elite 3 are my top affordable pick due to its impressive sound quality for the price ($80), good battery life, comfy fit and reliable touch controls. There’s no automatic pausing and EQ adjustments are limited to presets, but the audio is great out of the box. My colleague Jeff Dunn prefers Anker’s Soundcore Space A40. For $75, you get decent ANC, customizable sound, eight hours of battery life and almost all of the essentials.

Wrap-up

Amazon’s new approach to Echo Buds has produced interesting results. The company was only going to improve things so much at its previous mid-range price point, there’s a limit to how much tech you can add to a $150 set before the price has to go up. Going for the budget buyers instead of building a high-end set of $300 earbuds makes more sense considering its approach to other devices. Still, the company managed to create low-cost Echo Buds with sound quality that punches above the price point while still including all of the basic functionality. Alexa users even get a hands-free assistant, which isn’t all that common on $50 earbuds. There are certainly some omissions, but on the whole, Amazon’s choice to go budget friendly seems like the right call.

It was just another day in South Park. The kids were making fun of each other on the playground, while the parents were all doing their best to maintain their sanity in the small Colorado town. And then there was me, a tech journalist going door-to-door warning about the impending AI apocalypse. No, I wasn't actually guest starring on the long-running TV series — I was thrust into an episode entirely produced by the Showrunner AI model from The Simulation, the next iteration of the VR studio Fable.

All it took was some audio of my voice (recorded during a call with The Simulation's CEO Edward Saatchi), a picture and a two-sentence prompt to produce the episode. And while it wasn't the best South Park episode I've seen, I was shocked by how watchable it was. It begins with my AI character popping into Sharon and Randy Marsh's house in the morning, warning them about the AI uprising. Randy is intrigued, but Sharon is annoyed by my arrival (both of which were set up in one sentence of the initial prompt).

"They're [AI] infiltrating every aspect of our lives," my AI character says. "They're in our cars, our phones... even our toasters."

"Our toasters, really? I always knew that little bastard was up to something," Randy replies.

Sure, that's not exactly a tightly crafted joke, but it was enough to make me chuckle. And again, it didn't take much for Showrunner AI to piece that conversation together. Watching this episode made it clear that generative AI can actually produce watchable content (certainly more so than that AI Seinfeld project), but it also made me even more worried about the role of AI in media.

At this moment, writers in the WGA and performers in the SAG-AFTRA unions are striking for better residual pay and protections against potential AI exploitation. A tool like Showrunner AI, which can produce decent content without much effort, threatens creatives everywhere. The WGA strike and the fears around AI-generated content is also the main plot line in the demo South Park episode, "Westland Chronicles."

The Simulation

"Maybe it's a mistake to release it, I'm not sure," Saatchi said over email, when I asked if it's really the best time to launch Showrunner AI. "If our focus was becoming 'the AI TV studio' and gloating that we can make shows with no staff I'd feel very queasy — but we're trying to build a simulation and we need infinite story to make that work."

He added: "Now is the moment, in the biggest strike in 60 years, before AI has achieved takeoff, to negotiate the most aggressive protections possible for writers and actors from producers' use of AI — negotiations now so that these tools are in the hands of artists and creators only and not the hands of producers trying to become Griffin Mill Robert Altman's The Player."

(That character famously said, "I was just thinking what an interesting concept it is to eliminate the writer from the artistic process. If we could just get rid of these actors and directors, maybe we've got something here.")

The idea of building intelligent characters isn't new for Saatchi. In 2018, we discussed Fable Studio's VR adaptation of Neil Gaiman's The Wolves in the Wall, which centered on an interactive character named Lucy. Even then, he thought AI-powered beings were more intriguing than the notion of VR storytelling. Now, thanks to the proliferation of generative AI models, he can finally make that happen.

The Simulation

His goal with The Simulation is right there in the name: He wants to create simulations of characters living their lives in specific environments, similar to The Truman Show. That content could be cut down into episodic summaries with Showrunner AI. Like so many in the AI field, Saatchi eventually wants to build AGI, or Artificial General Intelligence, "an AI that reaches and surpasses human intelligence and is, ultimately, a new lifeform."

While many AI experts remain skeptical about the viability of AGI, Showrunner could still end up being a powerful tool. But of course, like any tool, it could end up being used for good or evil. Saatchi envisions it being helpful for creators to build their own shows without a huge budget, but he also admits it could be used by studios to "undermine artistic expression."

The Simulation

After watching myself starring in a South Park episode, though, I'm more worried than hopeful. The tech isn't perfect yet — my voice sometimes sounds overly robotic, and characters pronounce my name differently almost every time — but it's close enough. Saatchi and his team, including the AI research Philipp Maas, who developed the South Park simulation, still preview every episode to ensure they make sense. But with no end in strike for the Hollywood union strikes, it's hard not to imagine some studio looking at this tool as a potential savior.

“We aren’t releasing the Showrunner commercially and aren’t in talks with the South Park people (though we are with several studios about their IP, as well as creators to make original IP) but we used South Park only so that people have a comparison point between a super high quality human show and an AI show," Saatchi said. "If the showrunner was ever released with the permission of an IP (letting fans create their own episodes) we believe all the revenue should go to the IP holders — these are just remixes after all!"

Currently, The Simulation is developing showrunners around other animated concepts, like a space exploration series, and a Silicon Valley satire dubbed "Exit Valley." The technology can't produce live action content yet. The company is also releasing a research paper explaining how its showrunner AI technology works, with the hopes of encouraging more AI workers to build on it.

The most striking thing about my South Park episode is that it actually feels like an episode of the show. Over the course of five minutes, I visited the local school, where all of the kids proceeded to make fun of me. One wisely lectured me about overhyping claims about the AI apocalypse, while I was also warning them to watch out for clickbait. By the end, I was eager to see more. And that was the most worrying thing of all.

Streaming service Peacock is raising the prices of its subscription plans, if you have an inclination to rewatch The Office for the hundredth time. The ad-supported Premium subscription goes up by a buck, to $6 per month, while the ad-free Premium Plus option shoots up two bucks to $12 per month. An annual Premium subscription is going up $10, so $60 per year, while the Premium Plus subscription increases $20, making it $120 per year. This all goes into effect on August 17th, according to an email sent to current subscribers.

This is actually the first price hike for the NBCUniversal-owned streaming service since launching back in 2020. The increase impacts both current and incoming subscribers. The streamer nixed its free tier earlier this year, so you have to pay up to learn if Jim and Pam ever share that paper-drenched kiss.

Why the price hike? Peacock is hemorrhaging money, despite a rapidly-increasing subscriber base. As of April, the streamer boasted 22 million paying subscribers, but increased programming costs led to a loss of more than $700 million. NBC’s golden child was the last holdout, as every other major streaming platform already slapped subscribers with price bumps. Paramount+ prices shot up in February and the same goes for Max, Disney+, Apple+ and all the rest.

This could all be moot if those dual Hollywood strikes aren’t resolved in the short-term, as the streamers will run out of content quicker than you can say “go outside and touch grass.” Shows premiering now were filmed last year, thereabouts, so expect a serious dry spell next year around this time.

As for Peacock, there’s a reason it reached over 20 million subscribers in just a couple of years. It’s not just home for The Office and other beloved, though ancient, sitcoms. The service streams tons of live sports, including NFL football, Premier League soccer and WWE wrestling. It’s also gotten itself a stable of decent original programming, from the absolute lunacy of Mrs. Davis to the critically-acclaimed old-school sleuthing of Poker Face.

The rumors of a commercially-oriented Meta AI model were true. Meta and Microsoft have teamed up to unveil Llama 2, a next-generation large language (very generalized) AI model intended for both commercial and research purposes. The upgraded open source code places a greater emphasis on responsibility. Developers "red-teamed" models (that is, tested them for safety) and created a transparency schematic to detail potential issues. They also include a responsible use guide, and there's an acceptable use policy to prevent abuses like criminal activity, misleading representations and spam.

Meta is releasing both pre-trained and conversation-oriented versions of Llama 2 for free. Microsoft is making it available through the Azure AI catalog to use with cloud tools, such as content filtering. The tool can also run directly on Windows PCs, and will be available through outside providers like Amazon Web Services and Hugging Face.

The original AI model was open source, but intended strictly for academics and researchers. Llama 2 lets companies customize the technology for their own purposes, such as chatbots and image generators. Ideally, this lets companies benefit from the modelling while providing outsiders a way to check for biases, inaccuracies and other flaws. Open source code isn't new in the AI world — Stability's Stable Diffusion is a notable example. However, major rivals like OpenAI's GPT-4 tend to be locked down for the sake of subscription or licensing revenue. There are also concerns that hackers and other bad actors may ignore open source licenses and use those tools for dangerous purposes.

This is a developing story. Please check back for updates.