Microsoft's $68.7 billion acquisition of Activision Blizzard looked close to being dead not long ago, but it just took a big step toward clearing its last major obstacle. The UK's Competition and Markets Authority (CMA) has announced that Microsoft's revised agreement "substantially addresses previous concerns and opens the door to the deal being cleared." The agreement is still in consultation, but final approval now looks highly likely. 

"The CMA considers that the restructured deal makes important changes that substantially address the concerns it set out in relation to the original transaction earlier this year," the regulator wrote. "In particular, the sale of Activision’s cloud streaming rights to Ubisoft will prevent this important content — including games such as Call of Duty, Overwatch, and World of Warcraft — from coming under the control of Microsoft in relation to cloud gaming."

The UK regulator initially blocked the merger over fears it would hand Microsoft a 60 to 70 percent share of the cloud gaming market, making it a monopoly player. That in turn would give it "incentive to withhold games from competitors and substantially weaken competition in this important growing market." 

In response, Microsoft announced last month that it would sell Activision Blizzard streaming rights to Ubisoft in an attempt to win UK approval. It said that if the merger goes through, it would transfer "cloud streaming rights for all current and new Activision Blizzard PC and console games released over the next 15 years to Ubisoft Entertainment... in perpetuity." Ubisoft said in a separate release that the titles would be available across a range of services. 

The revised deal "substantially addresses most concerns," the CMA wrote, but it still wants to ensure that provisions in the sale of Activision's cloud streaming rights to Ubisoft can't be "circumvented, terminated or not enforced." It added that Microsoft has offered remedies to ensure that those rights are enforceable, and those should resolve any residual concerns. 

Microsoft managed to turn the deal around after taking a lot of blows from regulators. Late last year, the US Federal Trade Commission (FTC) sued to block the merger, but was later rebuffed by a federal court. The UK's CMA rejected the deal a few months later, but Microsoft appealed the decision and was later given more time to submit an amended deal. It made a major concession with the sale of streaming rights to Ubisoft — and that seems like it may have done the trick. We should know soon, as the CMA's consultation on Microsoft's proposed remedies closes on October 6. 

Apple TV+ has some intriguing TV series on hand to keep you occupied this fall. Along with the return of The Morning Show and the debut of a show that features Godzilla, the captivating For All Mankind is about to set sail on its fourth season. New episodes of the drama, which offers an alternate perspective on the space race, will start streaming on November 10.

Apple also released a teaser for the 10-episode season, which jumps forward eight years from the events of season three to 2003. Much of the action will take place on Mars, where astronauts stay busy by mining for resources. The teaser is in the form of a recruitment ad narrated by Ed Baldwin (Joel Kinnaman) that encourages people to sign up to work on the Red Planet.

As it happens, Apple just announced new iPhones that will soon be available to pre-order. If you snap up one of those — or another new iPhone, iPad, Apple TV, Mac or iPod touch — you'll get three months of Apple TV+ access at no extra cost.

Google Keep, the company's note-taking app, is getting a long-overdue feature that unfortunately doesn't seem fully baked. Google is adding a version history function, which could save you from having to manually retype a lot of text that you mistakenly deleted.

The tool allows you to download a text file with previous versions of your notes and lists, according to a support page. The help document states that Google is gradually rolling out the feature to everyone, so it may not be live for you yet. When it is, you can access it on the Keep web app, by clicking on the three-dot menu at the bottom of a note.

As Android Police points out, Keep's version history is only available on the web for now — you won't be able to see previous versions of your notes on the Android or iOS apps just yet. What's more, it doesn't cover images, so if you deleted a photo from a note, you won't be able to recover it using this option.

This is a fairly basic feature and it's somewhat baffling that Google hasn't offered it in Keep until now. After all, the company has long offered similar functions in Google Drive apps. The implementation is odd too. Rather than seeing the version history in the app and being able to revert to a previous incarnation of a note with a tap (like you can do in apps like Docs), having to download a file and copy text back in manually seems like a strange choice. That said, this is a step in the right direction for Keep.

Newly unsealed court filings reveal how much data Xwitter has handed over to the January 6 investigation. This includes all tweets sent, drafted, liked and retweeted – even if they were subsequently deleted – by Donald Trump’s official account. This cache also included DMs sent, received or stored in draft form, as well as linked accounts used on the same device. Even more interesting is the company handed over records of all searches made by the account, too.

We already knew Xwitter had fought the order tooth-and-nail, leading to a court battle and a hefty fine. But the list of what was available should also serve as a warning to everyone else that the platform stores a lot more data on its users than you might expect. The fact it could serve up location history, deleted DMs and a list of searches might make you wonder what else it has on you.

– Dan Cooper

You can get these reports delivered daily, direct to your inbox. Subscribe right here!​​​​

The biggest stories you might have missed

With some tech savvy, you can disconnect your robot vacuum from the cloud

The best soundbars for 2023

8BitDo stuffed 16 buttons into its hand-crampingly small Micro controller

Amazon Music Unlimited raises subscription prices again

Solo Stove introduces a $349 gas-only version of its Pi pizza oven

Netflix's Scott Pilgrim Takes Off teaser hits all the right notes

Scientists recreate an iconic Pink Floyd song by scanning listeners' brains

It could eventually help people with speech challenges.

Researchers claim to have found a way to extract a song from a person’s brain by analyzing their neural activity. A group of test subjects with drug-resistant epilepsy, who already have implants in their brains, were played Pink Floyd’s Another Brick in the Wall, Part 1. Using the implants, the team monitored how the brains responded to the stimulus and used AI to recreate what they heard. The result is a bit like listening to a Pink Floyd cover band playing down the street while you’re swimming, but it’s recognizable enough. It’s hoped the discovery could be used as a jumping-off point to develop tech to help people with atrophied speech communicate.

Continue Reading.

Intel walks away from its $5.4 billion takeover of Tower Semiconductor

Rumors suggest Chinese regulators were slow to agree to the deal.

Intel Corporation

Intel has withdrawn its $5.4 billion offer to buy Israeli chip fabricator Tower Semiconductor after failing to get regulatory approval. Tower is not a bleeding-edge manufacturer; it makes chips for industrial and automotive applications using older processes. Intel wanted it as part of its plan not to just make its own chips but to manufacture third-party designs in its facilities. It’s thought the major roadblock was China, which refused to give its blessing in a “timely manner.” It’s not clear if China’s inaction was part of the current geopolitical brouhaha over the future of chip making, but it probably hasn’t helped.

Continue Reading.

GM’s latest investment could speed development of cheaper EV batteries

It is putting $60 million into a US-based battery company.

Karl Nielsen for General Motors

GM has a vested interest in making batteries both cheaper and more efficient, which is why it just dropped $60 million into Mitra Chem’s pocket. Mitra Chem is working on a US-made battery using cheaper, more common metals than are presently employed. It’s also looking to speed up research into new battery technologies using software that could eliminate much of the early gruntwork.

Continue Reading.

Google is working to improve Bard's soulless life advice

But not everyone on the project thinks it’s a good idea.

A Google contractor developing Bard, the search giant’s AI chatbot, has enlisted a score of experts to improve the system’s ability to dole out life advice. It’s in response to users asking the platform for help with intimate issues, like how to back out of a destination wedding. The work has caused alarm in some parts of the company, where workers feel people shouldn’t be asking an AI those sorts of questions.

Continue Reading.

Snapchat's My AI chatbot glitched so hard it started posting Stories

It shouldn’t be able to do that.

Snapchat’s in-app digital assistant experienced a malfunction so great it started posting its own stories. My AI is a chatbot you can access if you subscribe to Snapchat Plus for $3.99 a month, but it shouldn’t have access to your stories at all. In a statement, the company explained the system had a “temporary outage” and there’s nothing to worry about. For now, at least.

Continue Reading.

My AI, the in-app digital assistant that rides herd on your Snapchat Plus experience, has suffered numerous breakdowns and technical malfunctions since its debut in February. Tuesday was more of the same as the chatbot took it upon itself to post single-second-long Stories to users' feeds and then go unresponsive for extended periods of time. Thing is, My AI doesn't have the capacity to post to Stories. And now it's got a bunch of people on Twitter wondering if we're at the dawning of the Singularity.

As first reported by TechCrunch, the My AI chatbot posted a two-tone image in Stories, which one user mistook for a shot of their ceiling. What's more, upon being asked about the mysterious post, the bot would either go dark or respond that it was suffering a technical issue. This explanation proved insufficient for many users, causing minor panic and jokes about the AI system's imminent awakening. In the end however, it really was a technical issue.

“My AI experienced a temporary outage that’s now resolved,” a spokesperson told TechCrunch, adding that "At this time, My AI does not have Stories feature."

The My AI bot is bundled as part of the company's $3.99/month Snapchat Plus package and offers users a variety of features. These include AR filter recommendations and suggestions for restaurant and activities based on currently popular places on the Snap Map. It also offers AI functionality in group chat, photo and video snaps, and text messages, as well as an AI persona. A text-to-image genAI is also reportedly in the works, though the ads have already arrived.

Activision is doing something different with Call of Duty: Modern Warfare III, this year's entry in the blockbuster military shooter series. Rather than starting from scratch and having to rebuild your collection of weapons and cosmetic items, Activision is letting players carry over nearly everything they unlock in Modern Warfare II.

This so-called Carry Forward initiative also applies to Call of Duty: Warzone, given that content is shared between the mainline games and the free-to-play battle royale. Warzone Mobile, which is slated to arrive later this year, will be integrated into all of this too.

For the most part, your unlocked operators, operator skins, bundles, all weapons, attachments and other rewards and cosmetic items will move forward from MW2 to MW3. What's more, if you continue to level up guns in MW2, that progress will be reflected in MW3.

This is a one-way street, though. Any MW3 progress or unlocks won't be replicated in MW2. There's no Carry Back feature.

The main things that won't progress from the 2022 game to this year's one are cosmetics for vehicles that aren't present in MW3. War Tracks, which are songs that can be played in vehicles, won't move over either. "Some Tactical and Lethal equipment may not be available depending on the removing of those items in MW3, to be replaced with MW3-only equipment," Activision noted in an extensive FAQ.

Even though MW2 and MW3 are handled by different internal studios (Infinity Ward and Sledgehammer Games, respectively), you won't have to jump through any hoops to transfer your weapons and cosmetics either. Activision will handle everything, though of course you'll need to be using the same account or profile for both games.

Activision will release Call of Duty: Modern Warfare III on November 10th. We'll learn much more about the upcoming game at a reveal event, which is set for August 17th.

Meanwhile, following the game's latest trailer, fans are speculating that MW3 will include an updated take on the hugely controversial No Russian mission from the original Modern Warfare II, which came out in 2009. That level wasn't in last year's rebooted version of MW2.

Elon Musk and X, the site formerly known as Twitter, are in more legal trouble. The Agence France-Presse (AFP) is suing X for not engaging in discussions about payment to the French publisher in exchange for its articles appearing on the platform. In 2019, France passed neighboring rights legislation, extending copyright law to content produced by news publishers, such as text and videos, for two years after release. The law requires any sites that share this work to negotiate with the publishers about remuneration instead of sharing it without compensation for its creators. 

In its press release, the AFP stated that it has "expressed its concerns over the clear refusal from Twitter (recently rebranded as 'X') to enter into discussions regarding the implementation of neighbouring rights for the press. These rights were established to enable news agencies and publishers to be remunerated by digital platforms which retain most of the monetary value generated by the distribution of news content."

X isn't the first tech company AFP has gone up against. In 2020, France's competition authority ordered Google to enter negotiations with publishers, and, while it reached an agreement in early 2021, the company was fined €500 million ($546 million) later that year for not reaching a fair agreement. In that case, part of the argument was that Google owns 90 percent of the search market, leaving them in a position where they could abuse their power if an equitable deal wasn't reached. Twitter's influence in this area of the internet isn't nearly as strong, so we'll have to wait and see if it will face the same fight. 

When Samsung released the original Galaxy Fold, it was about as durable as a Fabergé egg. But over the years, the company has made a number of changes to reduce the fragility of its flagship foldable phone. The Galaxy Z Fold 2 featured a redesigned hinge that prevented dirt from getting inside, while the Z Fold 3 added IPX8 water resistance and a stronger Armor Aluminum Chassis. And last year, the Z Fold 4 brought a more durable main screen and a new adhesive designed to keep its factory-installed screen protector more firmly in place.

That last one is a biggie because after owning a Z Fold 2 and a Z Fold 3, I found that the screen protectors on both phones started bubbling after six to eight months. This weakness is a concern for anyone thinking about buying an $1,800 foldable phone – especially when you consider that Samsung recommends that any repairs are done by an authorized service center. But as some who really likes foldable phones, I bought my own Z Fold 4 anyways and used it for a year. Here’s how well it held up.

Photo by Sam Rutherford/Engadget

I should mention that I’ve never put the phone in a case or used any other protective accessories like skins or sleeves. Despite being naked the whole time, the phone has done a decent job of withstanding typical daily abuse. Sure, there are some scratches and bare spots where paint has flaked off and a few dents from the phone being dropped or falling out of a pocket. But that’s sort of expected for a phone with no additional protection and both the front and back glass still look great.

More importantly, its flexible main screen looks practically as good as the day I got it. The screen protector is still sitting flat, there are no dead pixels or other blemishes and the hinge feels as sturdy as ever. All told, I’m pretty impressed considering some of the problems I encountered with previous generations. That said, while the pre-installed screen protector hasn’t started bubbling, there is one tiny spot along the top edge at the crease where you can see that it has started to (ever so slightly) separate from the display. So far, this hasn’t caused any issues. However, if past experience is any indication, this could cause the screen protector to start bubbling down the line.

Still, after claiming it switched to a new, more sticky adhesive to the Z Fold 4’s factory-installed screen protector in place, at least on my phone, Samsung’s tweak seems to have had at least some effect. Is the problem completely solved? No, not quite. Remember, this is just a single example, and it’s hard to account for things like the milder winter we’ve had this year, and chillier weather sometimes caused issues for Z Fold and Z Flip owners.

Also, while my Z Fold 4 has aged rather nicely, the screen protector on Engadget’s executive editor Aaron Souppouris’ Z Flip 4 has not fared nearly as well. He says the screen on his device was basically pristine for the first nine months. But after that, bubbles began to form and grew larger and larger until he removed the protector entirely and began using the phone with its naked flexible display.

It’s important to mention that Samsung instructs Z Flip and Z Fold owners not to use their devices without a screen protector. If you do remove it, you’re supposed to get it replaced as soon as possible. If you’re lucky, that can be as simple as finding a local Best Buy or uBreakiFix location and spending half an hour without your phone, and thankfully, Samsung offers one free screen protector replacement on both the Z Flip and Z Fold lines. Unfortunately, if you live in a remote area or just don’t have a nearby service center, you may need to rely on a mail-in repair, potentially leaving you without a phone for a couple of weeks or more. And for a lot of people, that’s not a reasonable option.

However, after talking to a number of Galaxy Z Flip and Z Fold owners who have removed their screen protectors, that seems to be merely a precaution. It’s totally possible to use a foldable phone without a screen protector just like you can on a regular handset. But given the more delicate nature of flexible displays (which are largely made of plastic instead of glass), the risk factor is higher. And with flexible screens costing a lot more to replace – up to $599 depending on the specific model – you don’t need a galaxy brain-sized noggin to understand why you might want to heed Samsung’s warnings. The counterpoint to that is because a foldable phone’s screen is protected by the rest of the device when closed, it’s only really vulnerable when you’re using it, as opposed to when it's simply resting in a pocket or bag.

Photo by Sam Rutherford/Engadget

So what’s the big takeaway? I think Samsung’s new adhesive has made a bit of a difference because, even in the case of Aaron’s Z Flip, it lasted longer than both of my previous Z Folds before the screen protector started bubbling. Even so, the screen protectors on Samsung’s foldable still require a bit more babying than a standard glass brick. This sort of fragility may be a deal-breaker for some, and understandably so. Thankfully, I live near multiple repair centers and I’m prepared to use my foldable without a screen protector – even though that’s not advised. 

For me, the ability to have a screen that expands when I want to watch a movie or multitask is worth the slightly reduced durability. But either way, this is something you need to consider before buying a foldable phone. In some ways, it’s like owning a car with a convertible roof, because while they're a bit more delicate and costly to repair, there’s nothing like driving around with the top off – or in this case a phone that can transform into a small tablet at a moment’s notice.

Just remember to do the sensible thing and put your expensive foldable phone in a case.

Nothing has finally unveiled the Phone 2 after plenty of teasers, and it's likely what you're looking for if you thought the Phone 1 was underpowered — or if you simply couldn't buy the earlier model where you live. The new device offers performance much closer to a flagship thanks to a Snapdragon 8+ Gen 1 chip versus the mid-tier 778G+ from last year's hardware. While that's still not cutting edge, the company claims it's 80 percent faster. It enables 4K video at 60 frames per second, too, and RAW HDR photography captures eight frames (and thus more overall scene detail) instead of three frames like its predecessor.

Accordingly, Nothing says it has upgraded the Phone 2's camera quality. The updated 50MP primary and 50MP ultra-wide rear cams now have 2X "super-res" digital zoom, object tracking and other imaging updates. The front camera, meanwhile, jumps from a 16MP sensor to 32MP. As with some competitors, there's now an "Action Mode" to deliver extra-stable video recording.

There are some more conspicuous changes. You can expect a larger 6.7-inch, 120Hz LTPO OLED (if still 1080p) screen with a higher 1,600-nit peak brightness and thinner bezels. There's a tapered "2.5D" glass back. And yes, the signature Glyph lighting on the back is more advanced. In addition to more LED segments, you can create different lighting sequences for every contact and notification type. You can also have persistent lights for must-see notifications, and some lights now double as progress trackers for delivery and ride hailing services like Uber.

Nothing

Software plays a considerably more important role. Where the first model only had a few modest customizations, Nothing OS 2.0 on the Phone 2 lets you tweak considerably more. You can now have multiple home screens with custom color themes, grid sizes and app labels. You'll likewise find customizable folders, and a more advanced widget set includes shortcuts to quick settings. Those widgets are available on the lock screen as well.

The Phone 2 is billed as longer-lasting thanks to its 4,700mAh battery, and you'll get a complete charge in 55 minutes. The 15W wireless charging and 5W reverse wireless charging aren't surprising, but they're not always present in this upper-midrange phone segment.

Crucially, the Nothing Phone 2 will be priced right when it arrives in North America. It will be available in the US and Canada on July 17th at 4AM Eastern starting at $599 (and an oddly high $929 CAD) for a version with 8GB of RAM and 128GB of storage. Pay $699 ($999 CAD) and you'll get 12GB of RAM with 256GB of storage, while the top-end 12GB/512GB configuration sells for $799 ($1,099 CAD). Pre-orders are available now, and there will be early sales on July 13th through physical "Nothing Drops" in New York City (69 Gansevoort Street) and London (4 Peter Street).

There's no mention of North American carrier deals as of this writing, so this sequel might not be as easy to find as more mainstream offerings. However, the launch in the region remains a big deal. The Phone 2 significantly expands the audience for Nothing's handsets, and provides fresh competition to bang-for-the-buck phones like Google's similarly-priced Pixel 7.

“Learn to code.” That three-word pejorative is perpetually on the lips and at the fingertips of internet trolls and tech bros whenever media layoffs are announced. A useless sentiment in its own right, but with the recent advent of code generating AIs, knowing the ins and outs of a programming language like Python could soon be about as useful as knowing how to fluently speak a dead language like Sanskrit. In fact, these genAIs are already helping professional software developers code faster and more effectively by handling much of the programming grunt work.

How coding works

Two of today’s most widely distributed and written coding languages are Java and Python. The former almost single handedly revolutionized cross-platform operation when it was released in the mid-’90s and now drives “everything from smartcards to space vehicles,” according to Java Magazine in 2020 — not to mention Wikipedia’s search function and all of Minecraft. The latter actually predates Java by a few years and serves as the code basis for many modern apps like Dropbox, Spotify and Instagram.

They differ significantly in their operation in that Java needs to be compiled (having its human-readable code translated into computer-executable machine code) before it can run, while Python is an interpreted language which means that its human code is converted into machine code line-by-line as the program executes, enabling it to run without first being compiled. The interpretation method allows code to be more easily written for multiple platforms while compiled code tends to be focused to a specific processor type. Regardless of how they run, the actual code-writing process is nearly identical between the two: somebody has to sit down, crack open a text editor or Integrated Development Environment (IDE) and actually write out all those lines of instruction. And up until recently, that somebody typically was a human.

The “classical programming” writing process of today isn’t that different from the process those of ENIAC, with a software engineer taking a problem, breaking it down into a series of sub-problems, writing code to solve each of those sub-problems in order, and then repeatedly debugging and recompiling the code until it runs. “Automatic programming,” on the other hand, removes the programmer by a degree of separation. Instead of a human writing each line of code individually, the person creates a high-level abstraction of the task for the computer to then generate low level code to address. This differs from “interactive” programming, which allows you to code a program while it is already running.

Today’s conversational AI coding systems, like what we see in Github’s Copilot or OpenAI’s ChatGPT, remove the programmer even further by hiding the coding process behind a veneer of natural language. The programmer tells the AI what they want programmed and how, and the machine can automatically generate the required code.

Building the tools to build the tools allowing any tool to build tools

Among the first of this new breed of conversational coding AIs was Codex, which was developed by OpenAI and released in late 2021. OpenAI had already implemented GPT-3 (precursor to GPT-3.5 that powers BingChat public) by this point, the large language model remarkably adept at mimicking human speech and writing after being trained on billions of words from the public web. The company then fine-tuned that model using 100-plus gigabytes of GitHub data to create Codex. It is capable of generating code in 12 different languages and can translate existing programs between them.

Codex is adept at generating small, simple or repeatable assets, like “a big red button that briefly shakes the screen when clicked” or regular functions like the email address validator on a Google Web Form. But no matter how prolific your prose, you won’t be using it for complex projects like coding a server-side load balancing program — it’s just too complicated an ask.

Google’s DeepMind developed AlphaCode specifically to address such challenges. Like Codex, AlphaCode was first trained on multiple gigabytes of existing GitHub code archives, but was then fed thousands of coding challenges pulled from online programming competitions, like figuring out how many binary strings with a given length don’t contain consecutive zeroes.

To do this, AlphaCode will generate as many as a million code candidates, then reject all but the top 1 percent to pass its test cases. The system will then group the remaining programs based on the similarity of their outputs and sequentially test them until it finds a candidate that successfully solves the given problem. Per a 2022 study published in Science, AlphaCode managed to correctly answer those challenge questions 34 percent of the time (compared to Codex’s single-digit success on the same benchmarks, that’s not bad). DeepMind even entered AlphaCode in a 5,000-competitor online programming contest, where it surpassed nearly 46 percent of the human competitors.

Now even the AI has notes

Just as GPT-3.5 serves as a foundational model for ChatGPT, Codex serves as the basis for GitHub’s Copilot AI. Trained on billions of lines of code assembled from the public web, Copilot offers cloud-based AI-assisted coding autocomplete features through a subscription plugin for the Visual Studio Code, Visual Studio, Neovim, and JetBrains integrated development environments (IDEs).

Initially released as a developer’s preview in June of 2021, Copilot was among the very first coding capable AIs to reach the market. More than a million devs have leveraged the system in the two years since, GitHub's VP of Product Ryan J Salva, told Engadget during a recent interview. With Copilot, users can generate runnable code from natural language text inputs as well as autocomplete commonly repeated code sections and programming functions.

Salva notes that prior to Copilot’s release, GitHub’s previous machine-generated coding suggestions were only accepted by users 14 - 17 percent of the time, “which is fine. It means it was helping developers along.” In the two years since Copilot’s debut, that figure has grown to 35 percent, “and that's netting out to just under half of the amount of code being written [on GitHub] — 46 percent by AI to be exact.”

“[It’s] not a matter of just percentage of code written,” Salva clarified. “It's really about the productivity, the focus, the satisfaction of the developers who are creating.”

As with the outputs of natural language generators like ChatGPT, the code coming from Copilot is largely legible, but like any large language model trained on the open internet, GitHub made sure to incorporate additional safeguards against the system unintentionally producing exploitable code.

“Between when the model produces a suggestion and when that suggestion is presented to the developer,” Salva said, “we at runtime perform … a code quality analysis for the developer, looking for common errors or vulnerabilities in the code like cross-site scripting or path injection.”

That auditing step is meant to improve the quality of recommended code over time rather than monitor or police what the code might be used for. Copilot can help developers create the code that makes up malware, the system won’t prevent it. “We've taken the position that Copilot is there as a tool to help developers produce code,” Salva said, pointing to the numerous White Hat applications for such a system. “Putting a tool like Copilot in their hands … makes them more capable security researchers,” he continued.

As the technology continues to develop, Salva sees generative AI coding to expand far beyond its current technological bounds. That includes “taking a big bet” on conversational AI. “We also see AI-assisted development really percolating up into other parts of the software development life cycle,” he said, like using AI to autonomously repair a CI/CD build errors, patch security vulnerabilities, or have the AI review human-written code.

“Just as we use compilers to produce machine-level code today, I do think they'll eventually get to another layer of abstraction with AI that allows developers to express themselves in a different language,” Salva said. “Maybe it's natural language like English or French, or Korean. And that then gets ‘compiled down’ to something that the machines can understand,” freeing up engineers and developers to focus on the overall growth of the project rather than the nuts and bolts of its construction.

From coders to gabbers

With human decision-making still firmly wedged within the AI programming loop, at least for now, we have little to fear from having software writing software. As Salva noted, computers already do this to a degree when compiling code, and digital gray goos have yet to take over because of it. Instead, the most immediate challenges facing programming AI mirror those of generative AI in general: inherent biases skewing training data, model outputs that violate copyright, and concerns surrounding user data privacy when it comes to training large language models.

GitHub is far from alone in its efforts to build an AI programming buddy. OpenAI’s ChatGPT is capable of generating code — as are the already countless indie variants being built atop the GPT platform. So too is Amazon’s AWS CodeWhisperer system, which provides much of the same autocomplete functionality as Copilot, but optimized for use within the AWS framework. After multiple requests from users, Google incorporated code generation and debugging capabilities into Bard this past April as well, ahead of its ecosystem-wide pivot to embrace AI at I/O 2023 and the release of Codey, Alphabet’s answer to Copilot. We can’t be sure yet what generative coding systems will eventually become or how it might impact the tech industry — we could be looking at the earliest iterations of a transformative democratizing technology, or it could be Clippy for a new generation.