Enlarge/ You can have confirmation of one of our upcoming four phones, but you have to hear us talk about AI again. Deal?
Google
After every one of its house-brand phones, and even its new wall charger, have been meticulously photographed, sized, and rated for battery capacity, what should Google do to keep the anticipation up for the Pixel 9 series’ August 13 debut?
Lean into it, it seems, and Google is doing so with an eye toward further promoting its Gemini-based AI aims. In a video post on X (formerly Twitter), Google describes a “phone built for the Gemini era,” one that can, through the power of Gemini, “even let your old phone down easy” with a breakup letter. The camera pans out, and the shape of the Pixel 9 Pro appears and turns around to show off the now-standard Pixel camera bar across the upper back.
There’s also a disclaimer to this tongue-in-cheek request for a send-off to a phone that is “just the same old thing”: “Screen simulated. Limitations apply. Check responses for accuracy.”
Over at the Google Store, you can see a static image of the Pixel 9 Pro and sign up for alerts about its availability. The image confirms that the photos taken by Taiwanese regulatory authority NCC were legitimate, right down to the coloring on the back of the Pixel 9 Pro and the camera and flash placement.
Those NCC photos confirmed that Google intends to launch four different phone-ish devices at its August 13 “Made by Google” event. The Pixel 9 and Pixel 9 Pro are both roughly 6.1-inch devices, but the Pro will likely offer more robust Gemini AI integration due to increased RAM and other spec bumps. The Pixel 9 Pro XL should have similarly AI-ready specs, just in a larger size. And the Pixel 9 Pro Fold is an iteration on Google’s first Pixel Fold model, with seemingly taller dimensions and a daringly smaller battery.
Enlarge/ In goes the sensible characters, out goes a document for which you almost always have to adjust the sharing permissions.
Aurich Lawson | Getty Images
The best time to truly implement the Markdown markup language into Google Docs was in the early 2010s, but yesterday was a pretty good time, too.
Google Docs was born from the conjoined features of a series of software company acquisitions (Writely, DocVerse, and QuickOffice), plus the remains of Google Wave, smooshed together into Drive by 2012. By that point, Markdown, a project of web writer John Gruber with input from data activist Aaron Swartz, had been solidified and gathering steam for about eight years. Then, for another decade or so, writing in Markdown and writing in Google Docs were two different things, joined together only through browser extensions or onerous import/export tools. An uncountable number of cloud-syncing, collaboration-friendly but Markdown-focused writing tools flourished in that chasm.
In early 2022, the first connecting plank was placed: Docs could “Automatically detect Markdown,” if you enabled it. This expanded the cursory support for numbered and unordered lists and checkboxes to the big items, like headlines, italics, bold, strikethrough, and links. You could write in Markdown in Docs, but you could not paste, nor could you import or export between Docs and Markdown styling.
Now, or at some point in the next 14 days, real, actual Markdown work can be done in Google Docs. Docs can convert Markdown text to its equivalent Docs formatting on paste or when imported as a file, and it can export to Markdown from the copy menu or as a file. Google’s blog post notes that this is “particularly useful for technical content writers as they can now convert Docs content to/from Markdown,” so as to use Google’s always-on syncing and collaboration in the interim stages.
As someone who doesn’t work as a technical content writer (at least in proper job title fashion), but does write a lot, allow me to say that this is also particularly useful for people who adopted Markdown as a kind of One True Style. It is hard to avoid being invited to collaborate on Google Docs, even if you primarily work elsewhere. It is even harder to remember all the different shortcuts for headlines, bullet points, and other text elements across various apps, web apps, content management systems, IDEs, and other writing platforms.
There’s no indication of which flavor of Markdown Google’s import and export functions will hew to, and Ars was unable to test the new function as of July 17. Markdown is intentionally not fully standardized by its author, leading to some kerfuffles and, eventually, an understanding that each version, like GitHub Flavored Markdown, has its own additions and changes.
By allowing for import/export, but even better, “Copy as Markdown” and “Paste from Markdown,” Docs is now a place where I can be a Markdown grouch and still play reasonably nice with others. You should see “Enable Markdown” show up in your Docs’ Tools > Preferences menu within the next two weeks.
According to Bloomberg, Google’s offer to the Cloud Infrastructure Services Providers in Europe (CISPE) required that the group maintain its EU antitrust complaint. It came “just days” before CISPE settled with Microsoft, and it was apparently not compelling enough to stop CISPE from inking a deal with the software giant that TechCrunch noted forced CISPE to accept several compromises.
Bloomberg uncovered Google’s attempted counteroffer after reviewing confidential documents and speaking to “people familiar with the matter.” Apparently, Google sought to sway CISPE with a package worth nearly $500 million for more than five years of software licenses and about $15 million in cash.
But CISPE did not take the bait, announcing last week that an agreement was reached with Microsoft, seemingly frustrating Google.
CISPE initially raised its complaint in 2022, alleging that Microsoft was “irreparably damaging the European cloud ecosystem and depriving European customers of choice in their cloud deployments” by spiking costs to run Microsoft’s software on rival cloud services. In February, CISPE said that “any remedies and resolution must apply across the sector and to be accessible to all cloud customers in Europe.” They also promised that “any agreements will be made public.”
But the settlement reached last week excluded major rivals, including Amazon, which is a CISPE member, and Google, which is not. And despite CISPE’s promise, the terms of the deal were not published, apart from a CISPE blog roughly outlining central features that it claimed resolved the group’s concerns over Microsoft’s allegedly anticompetitive behaviors.
What is clear is that CISPE agreed to drop their complaint by taking the deal, but no one knows exactly how much Microsoft paid in a “lump sum” to cover CISPE legal fees for three years, TechCrunch noted. However, “two people with direct knowledge of the matter” told Reuters that Microsoft offered about $22 million.
Google has been trying to catch up with Microsoft and Amazon in the cloud market and has recently begun gaining ground. Last year, Google’s cloud operation broke even for the first time, and the company earned a surprising $900 million in profits in the first quarter of 2024, which bested analysts’ projections by more than $200 million, Bloomberg reported. For Google, the global cloud market has become a key growth area, Bloomberg noted, as potential growth opportunities in search advertising slow. Seemingly increasing regulatory pressure on Microsoft while taking a chunk of its business in the EU was supposed to be one of Google’s next big moves.
A CISPE spokesperson, Ben Maynard, told Ars that its “members were presented with alternative options to accepting the Microsoft deal,” while not disclosing the terms of the other options. “However, the members voted by a significant majority to accept the Microsoft offer, which, in their view, presented the best opportunity for the European cloud sector,” Maynard told Ars.
Neither Microsoft nor Google has commented directly on the reported counteroffer. A Google spokesperson told Bloomberg that Google “has long supported the principles of fair software licensing and that the firm was having discussions about joining CISPE, to fight anticompetitive licensing practices.” A person familiar with the matter told Ars that Google did not necessarily make the counteroffer contingent on dropping the EU complaint, but had long been exploring joining CISPE and would only do so if CISPE upheld its mission to defend fair licensing deals. Microsoft reiterated a past statement from its president, Brad Smith, confirming that Microsoft was “pleased” to resolve CISPE’s antitrust complaint.
For CISPE, the resolution may not have been perfect, but it “will enable European cloud providers to offer Microsoft applications and services on their local cloud infrastructures, meeting the demand for sovereign cloud solutions.” In 2022, CISPE Secretary-General Francisco Mingorance told Ars that although CISPE had been clear that it intended to force Microsoft to make changes allowing all cloud rivals to compete, “a key reason behind filing the complaint was to support” two smaller cloud service providers, Aruba and OVH.
Google is making it easier for people to lock down their accounts with strong multifactor authentication by adding the option to store secure cryptographic keys in the form of passkeys rather than on physical token devices.
Google’s Advanced Protection Program, introduced in 2017, requires the strongest form of multifactor authentication (MFA). Whereas many forms of MFA rely on one-time passcodes sent through SMS or emails or generated by authenticator apps, accounts enrolled in advanced protection require MFA based on cryptographic keys stored on a secure physical device. Unlike one-time passcodes, security keys stored on physical devices are immune to credential phishing and can’t be copied or sniffed.
Democratizing APP
APP, short for Advanced Protection Program, requires the key to be accompanied by a password whenever a user logs into an account on a new device. The protection prevents the types of account takeovers that allowed Kremlin-backed hackers to access the Gmail accounts of Democratic officials in 2016 and go on to leak stolen emails to interfere with the presidential election that year.
Until now, Google required people to have two physical security keys to enroll in APP. Now, the company is allowing people to instead use two passkeys or one passkey and one physical token. Those seeking further security can enroll using as many keys as they want.
“We’re expanding the aperture so people have more choice in how they enroll in this program,” Shuvo Chatterjee, the project lead for APP, told Ars. He said the move comes in response to comments Google has received from some users who either couldn’t afford to buy the physical keys or lived or worked in regions where they’re not available.
As always, users must still have two keys to enroll to prevent being locked out of accounts if one of them is lost or broken. While lockouts are always a problem, they can be much worse for APP users because the recovery process is much more rigorous and takes much longer than for accounts not enrolled in the program.
Passkeys are the creation of the FIDO Alliance, a cross-industry group comprised of hundreds of companies. They’re stored locally on a device and can also be stored in the same type of hardware token storing MFA keys. Passkeys can’t be extracted from the device and require either a PIN or a scan of a fingerprint or face. They provide two factors of authentication: something the user knows—the underlying password used when the passkey was first generated—and something the user has—in the form of the device storing the passkey.
Of course, the relaxed requirements only go so far since users still must have two devices. But by expanding the types of devices needed, APP becomes more accessible since many people already have a phone and computer, Chatterjee said.
“If you’re in a place where you can’t get security keys, it’s more convenient,” he explained. “This is a step toward democratizing how much access [users] get to this highest security tier Google offers.”
Despite the increased scrutiny involved in the recovery process for APP accounts, Google is renewing its recommendation that users provide a phone number and email address as backup.
“The most resilient thing to do is have multiple things on file, so if you lose that security key or the key blows up, you have a way to get back into your account,” Chatterjee said. He’s not providing the “secret sauce” details about how the process works, but he said it involves “tons of signals we look at to figure out what’s really happening.
“Even if you do have a recovery phone, a recovery phone by itself isn’t going to get you access to your account,” he said. “So if you get SIM swapped, it doesn’t mean someone gets access to your account. It’s a combination of various factors. It’s the summation of that that will help you on your path to recovery.”
Google users can enroll in APP by visiting this link.
Cooling pipes at a Google data center in Douglas County, Georgia.
Google’s greenhouse gas emissions have surged 48 percent in the past five years due to the expansion of its data centers that underpin artificial intelligence systems, leaving its commitment to get to “net zero” by 2030 in doubt.
The Silicon Valley company’s pollution amounted to 14.3 million tonnes of carbon equivalent in 2023, a 48 percent increase from its 2019 baseline and a 13 percent rise since last year, Google said in its annual environmental report on Tuesday.
Google said the jump highlighted “the challenge of reducing emissions” at the same time as it invests in the build-out of large language models and their associated applications and infrastructure, admitting that “the future environmental impact of AI” was “complex and difficult to predict.”
Chief Sustainability Officer Kate Brandt said the company remained committed to the 2030 target but stressed the “extremely ambitious” nature of the goal.
“We do still expect our emissions to continue to rise before dropping towards our goal,” said Brandt.
She added that Google was “working very hard” on reducing its emissions, including by signing deals for clean energy. There was also a “tremendous opportunity for climate solutions that are enabled by AI,” said Brandt.
As Big Tech giants including Google, Amazon, and Microsoft have outlined plans to invest tens of billions of dollars into AI, climate experts have raised concerns about the environmental impacts of the power-intensive tools and systems.
In May, Microsoft admitted that its emissions had risen by almost a third since 2020, in large part due to the construction of data centers. However, Microsoft co-founder Bill Gates last week also argued that AI would help propel climate solutions.
Meanwhile, energy generation and transmission constraints are already posing a challenge for the companies seeking to build out the new technology. Analysts at Bernstein said in June that AI would “double the rate of US electricity demand growth and total consumption could outstrip current supply in the next two years.”
In Tuesday’s report, Google said its 2023 energy-related emissions—which come primarily from data center electricity consumption—rose 37 percent year on year and overall represented a quarter of its total greenhouse gas emissions.
Google’s supply chain emissions—its largest chunk, representing 75 percent of its total emissions—also rose 8 percent. Google said they would “continue to rise in the near term” as a result in part of the build-out of the infrastructure needed to run AI systems.
Google has pledged to achieve net zero across its direct and indirect greenhouse gas emissions by 2030 and to run on carbon-free energy during every hour of every day within each grid it operates by the same date.
However, the company warned in Tuesday’s report that the “termination” of some clean energy projects during 2023 had pushed down the amount of renewables it had access to.
Meanwhile, the company’s data center electricity consumption had “outpaced” Google’s ability to bring more clean power projects online in the US and Asia-Pacific regions.
Google’s data center electricity consumption increased 17 percent in 2023, and amounted to approximately 7-10 percent of global data center electricity consumption, the company estimated. Its data centers also consumed 17 percent more water in 2023 than during the previous year, Google said.
Enlarge/ The logo for PaLM 2, a Google large language model.
Google
Google announced today that it has added support for 110 new languages to Google Translate, nearly doubling the number of languages that can be translated.
In a blog post, Google Senior Software Engineer Isaac Caswell claimed that the newly added languages are spoken by more than 614 million people, or about 8 percent of the global population.
He noted that about a quarter of the languages originate in Africa, “representing our largest expansion of African languages to date.”
The blog post also went into some light detail about Google’s philosophy for choosing languages and for deciding which dialects to support:
Languages have an immense amount of variation: regional varieties, dialects, different spelling standards. In fact, many languages have no one standard form, so it’s impossible to pick a “right” variety. Our approach has been to prioritize the most commonly used varieties of each language. For example, Romani is a language that has many dialects all throughout Europe. Our models produce text that is closest to Southern Vlax Romani, a commonly used variety online. But it also mixes in elements from others, like Northern Vlax and Balkan Romani.
This update brings the total number of languages supported by Google Translate to 243, which is just the beginning of its publicized initiative to ultimately support 1,000 languages through the use of AI. You can see the full list of languages added in a help page published by Google.
By contrast, Apple Translate supports 21 languages, though that number includes both US and UK English as distinct options. Apple recently announced plans to add Hindi to its Translate app. Of course, Apple and Google take very different approaches to—and have different levels of investment in—these tools.
Enlarge/ A screen shows a virtual meeting with Microsoft Teams at a conference on January 30, 2024 in Barcelona, Spain.
Microsoft may be hit with a massive fine in the European Union for “possibly abusively” bundling Teams with its Office 365 and Microsoft 365 software suites for businesses.
On Tuesday, the European Commission (EC) announced preliminary findings of an investigation into whether Microsoft’s “suite-centric business model combining multiple types of software in a single offering” unfairly shut out rivals in the “software as a service” (SaaS) market.
“Since at least April 2019,” the EC found, Microsoft’s practice of “tying Teams with its core SaaS productivity applications” potentially restricted competition in the “market for communication and collaboration products.”
The EC is also “concerned” that the practice may have helped Microsoft defend its dominant market position by shutting out “competing suppliers of individual software” like Slack and German video-conferencing software Alfaview. Makers of those rival products had complained to the EC last year, setting off the ongoing probe into Microsoft’s bundling.
Customers should have choices, the EC said, and seemingly at every step, Microsoft sought instead to lock customers into using only its software.
“Microsoft may have granted Teams a distribution advantage by not giving customers the choice whether or not to acquire access to Teams when they subscribe to their SaaS productivity applications,” the EC wrote. This alleged abusive practice “may have been further exacerbated by interoperability limitations between Teams’ competitors and Microsoft’s offerings.”
For Microsoft, the EC’s findings are likely not entirely unexpected, although Tuesday’s announcement must be disappointing. The company had been hoping to avoid further scrutiny by introducing some major changes last year. Most drastically, Microsoft began “offering some suites without Teams,” the EC said, but even that wasn’t enough to appease EU regulators.
“The Commission preliminarily finds that these changes are insufficient to address its concerns and that more changes to Microsoft’s conduct are necessary to restore competition,” the EC said, concluding that “the conduct may have prevented Teams’ rivals from competing, and in turn innovating, to the detriment of customers in the European Economic Area.”
Microsoft will now be given an opportunity to defend its practices. If the company is unsuccessful, it risks a potential fine up to 10 percent of its annual worldwide turnover and an order possibly impacting how the leading global company conducts business.
In a statement to Ars, Microsoft President Brad Smith confirmed that the tech giant would work with the commission to figure out a better solution.
“Having unbundled Teams and taken initial interoperability steps, we appreciate the additional clarity provided today and will work to find solutions to address the commission’s remaining concerns,” Smith said.
The EC’s executive vice-president in charge of competition policy, Margrethe Vestager, explained in a statement why the commission refuses to back down from closely scrutinizing Microsoft’s alleged unfair practices.
“We are concerned that Microsoft may be giving its own communication product Teams an undue advantage over competitors by tying it to its popular productivity suites for businesses,” Vestager said. “And preserving competition for remote communication and collaboration tools is essential as it also fosters innovation” in these markets.
Changes coming to EU antitrust law in 2025
The EC initially launched its investigation into Microsoft’s allegedly abusive Teams bundling last July. Its probe came after Slack and Alfaview makers complained that Microsoft may be violating Article 102 of the Treaty on the Functioning of the European Union (TFEU), “which prohibits the abuse of a dominant market position.”
Nearly one year later, there’s no telling when the EC’s inquiry into Microsoft Teams will end. Microsoft will have a chance to review all evidence of infringement gathered by EU regulators to form its response. After that, the EC will review any additional evidence before making its decision, and there is no legal deadline to complete the antitrust inquiry, the EC said.
It’s possible that the EC’s decision may come next year when the EU is preparing to release new guidance to more “vigorously” and effectively enforce TFEU.
Last March, the EC called for stakeholder feedback after rolling out “the first major policy initiative in the area of abuse of dominance rules.” The initiative sought to update TFEU for the first time since 2008 based on reviewing relevant case law.
“A robust enforcement of rules on abuse of dominance benefits both consumers and a stronger European economy,” Vestager said at that time. “We have carefully analyzed numerous EU court judgments on the application of Article 102, and it is time for us to start working on guidelines reflecting this case law.”
Artificial intelligence-generated “deepfakes” that impersonate politicians and celebrities are far more prevalent than efforts to use AI to assist cyber attacks, according to the first research by Google’s DeepMind division into the most common malicious uses of the cutting-edge technology.
The study said the creation of realistic but fake images, video, and audio of people was almost twice as common as the next highest misuse of generative AI tools: the falsifying of information using text-based tools, such as chatbots, to generate misinformation to post online.
The most common goal of actors misusing generative AI was to shape or influence public opinion, the analysis, conducted with the search group’s research and development unit Jigsaw, found. That accounted for 27 percent of uses, feeding into fears over how deepfakes might influence elections globally this year.
Deepfakes of UK Prime Minister Rishi Sunak, as well as other global leaders, have appeared on TikTok, X, and Instagram in recent months. UK voters go to the polls next week in a general election.
Concern is widespread that, despite social media platforms’ efforts to label or remove such content, audiences may not recognize these as fake, and dissemination of the content could sway voters.
Ardi Janjeva, research associate at The Alan Turing Institute, called “especially pertinent” the paper’s finding that the contamination of publicly accessible information with AI-generated content could “distort our collective understanding of sociopolitical reality.”
Janjeva added: “Even if we are uncertain about the impact that deepfakes have on voting behavior, this distortion may be harder to spot in the immediate term and poses long-term risks to our democracies.”
The study is the first of its kind by DeepMind, Google’s AI unit led by Sir Demis Hassabis, and is an attempt to quantify the risks from the use of generative AI tools, which the world’s biggest technology companies have rushed out to the public in search of huge profits.
As generative products such as OpenAI’s ChatGPT and Google’s Gemini become more widely used, AI companies are beginning to monitor the flood of misinformation and other potentially harmful or unethical content created by their tools.
In May, OpenAI released research revealing operations linked to Russia, China, Iran, and Israel had been using its tools to create and spread disinformation.
“There had been a lot of understandable concern around quite sophisticated cyber attacks facilitated by these tools,” said Nahema Marchal, lead author of the study and researcher at Google DeepMind. “Whereas what we saw were fairly common misuses of GenAI [such as deepfakes that] might go under the radar a little bit more.”
Google DeepMind and Jigsaw’s researchers analyzed around 200 observed incidents of misuse between January 2023 and March 2024, taken from social media platforms X and Reddit, as well as online blogs and media reports of misuse.
Ars Technica
The second most common motivation behind misuse was to make money, whether offering services to create deepfakes, including generating naked depictions of real people, or using generative AI to create swaths of content, such as fake news articles.
The research found that most incidents use easily accessible tools, “requiring minimal technical expertise,” meaning more bad actors can misuse generative AI.
Google DeepMind’s research will influence how it improves its evaluations to test models for safety, and it hopes it will also affect how its competitors and other stakeholders view how “harms are manifesting.”
Enlarge/ Michael Jackson in concert, 1986. Sony Music owns a large portion of publishing rights to Jackson’s music.
Universal Music Group, Sony Music, and Warner Records have sued AI music-synthesis companies Udio and Suno for allegedly committing mass copyright infringement by using recordings owned by the labels to train music-generating AI models, reports Reuters. Udio and Suno can generate novel song recordings based on text-based descriptions of music (i.e., “a dubstep song about Linus Torvalds”).
The lawsuits, filed in federal courts in New York and Massachusetts, claim that the AI companies’ use of copyrighted material to train their systems could lead to AI-generated music that directly competes with and potentially devalues the work of human artists.
Like other generative AI models, both Udio and Suno (which we coveredseparately in April) rely on a broad selection of existing human-created artworks that teach a neural network the relationship between words in a written prompt and styles of music. The record labels correctly note that these companies have been deliberately vague about the sources of their training data.
Until generative AI models hit the mainstream in 2022, it was common practice in machine learning to scrape and use copyrighted information without seeking permission to do so. But now that the applications of those technologies have become commercial products themselves, rightsholders have come knocking to collect. In the case of Udio and Suno, the record labels are seeking statutory damages of up to $150,000 per song used in training.
In the lawsuit, the record labels cite specific examples of AI-generated content that allegedly re-creates elements of well-known songs, including The Temptations’ “My Girl,” Mariah Carey’s “All I Want for Christmas Is You,” and James Brown’s “I Got You (I Feel Good).” It also claims the music-synthesis models can produce vocals resembling those of famous artists, such as Michael Jackson and Bruce Springsteen.
Reuters claims it’s the first instance of lawsuits specifically targeting music-generating AI, but music companies and artists alike have been gearing up to deal with challenges the technology may pose for some time.
In May, Sony Music sent warning letters to over 700 AI companies (including OpenAI, Microsoft, Google, Suno, and Udio) and music-streaming services that prohibited any AI researchers from using its music to train AI models. In April, over 200 musical artists signed an open letter that called on AI companies to stop using AI to “devalue the rights of human artists.” And last November, Universal Music filed a copyright infringement lawsuit against Anthropic for allegedly including artists’ lyrics in its Claude LLM training data.
Similar to The New York Times’ lawsuit against OpenAI over the use of training data, the outcome of the record labels’ new suit could have deep implications for the future development of generative AI in creative fields, including requiring companies to license all musical training data used in creating music-synthesis models.
Compulsory licenses for AI training data could make AI model development economically impractical for small startups like Udio and Suno—and judging by the aforementioned open letter, many musical artists may applaud that potential outcome. But such a development would not preclude major labels from eventually developing their own AI music generators themselves, allowing only large corporations with deep pockets to control generative music tools for the foreseeable future.
Microsoft is all-in on its Copilot+ PC push right now, but the fact is that they’ll be an extremely small minority among the PC install base for the foreseeable future. The program’s stringent hardware requirements—16GB of RAM, at least 256GB of storage, and a fast neural processing unit (NPU)—disqualify all but new PCs, keeping features like Recall from running on all current Windows 11 PCs.
But the Copilot chatbot remains supported on all Windows 11 PCs (and most Windows 10 PCs), and a change Microsoft has made to recent Windows 11 Insider Preview builds is actually making the feature less useful and accessible than it is in the current publicly available versions of Windows. Copilot is being changed from a persistent sidebar into an app window that can be resized, minimized, and pinned and unpinned from the taskbar, just like any other app. But at least as of this writing, this version of Copilot can no longer adjust Windows’ settings, and it’s no longer possible to call it up with the Windows+C keyboard shortcut. Only newer keyboards with the dedicated Copilot key will have an easy built-in keyboard shortcut for summoning Copilot.
If Microsoft keeps these changes intact, they’ll hit Windows 11 PCs when the 24H2 update is released to the general public later this year; the changes are already present on Copilot+ PCs, which are running a version of Window 11 24H2 out of the box.
Changing how Copilot works is all well and good—despite how quickly Microsoft has pushed it out to every Windows PC in existence, it has been labeled a “preview” up until the 24H2 update, and some amount of change is to be expected. But discontinuing the just-introduced Win+C keyboard shortcut to launch Copilot feels pointless, especially since the Win+C shortcut isn’t being reassigned.
The Copilot assistant exists on the taskbar, so it’s not as though it’s difficult to access, but the feature is apparently important enough to merit the first major change to Windows keyboards in three decades. Surely it also justifies retaining a keyboard shortcut for the vast majority of PC keyboards without a dedicated Copilot key.
People who want to continue to use Win+C as a launch key for Copilot can do so with custom keyboard remappers like Microsoft’s own Keyboard Manager PowerToy. Simply set Win+C as a shortcut for the obscure Win+Shift+F23 shortcut that the hardware Copilot key is already mapped to and you’ll be back in business.
Win+C has a complicated history
Enlarge/ Win+C always seems to get associated with transient, unsuccessful Windows features like Charms and Cortana.
Andrew Cunningham
The Win+C keyboard shortcut actually has a bit of a checkered history, having been reassigned over the years to multiple less-than-successful Windows initiatives. In Windows 8, it made its debut as a shortcut for the “Charms” menu, part of the operating system’s tablet-oriented user interface that was designed to partially replace the old Start menu. But Windows 10 retreated from this new tablet UI, and the Charms bar was discontinued.
In Windows 10, Win+C was assigned to the Cortana voice assistant instead, Microsoft’s contribution to the early-2010s voice assistant boom kicked off by Apple’s Siri and refined by competitors like Amazon’s Alexa. But Cortana, like the Charms bar, never really took off, and Microsoft switched the voice assistant off in 2023 after a few years of steadily deprioritizing it in Windows 10 (and mostly hiding it in Windows 11).
Most older versions of Windows didn’t do anything with the Win+C, but if you go all the way back to the Windows 95 era, users of Microsoft Natural Keyboards who installed Microsoft’s IntelliType software could use Win+C to open the Control Panel. This shortcut apparently never made it into Windows itself, even as the Windows key became standard equipment on PCs in the late ’90s and early 2000s.
So pour one out for Win+C, the keyboard shortcut that is always trying to do something new and not quite catching on. We can’t wait to see what it does next.
Enlarge/ Screen capture of a Runway Gen-3 Alpha video generated with the prompt “A giant humanoid, made of fluffy blue cotton candy, stomping on the ground, and roaring to the sky, clear blue sky behind them.”
On Sunday, Runway announced a new AI video synthesis model called Gen-3 Alpha that’s still under development, but it appears to create video of similar quality to OpenAI’s Sora, which debuted earlier this year (and has also not yet been released). It can generate novel, high-definition video from text prompts that range from realistic humans to surrealistic monsters stomping the countryside.
Unlike Runway’s previous best model from June 2023, which could only create two-second-long clips, Gen-3 Alpha can reportedly create 10-second-long video segments of people, places, and things that have a consistency and coherency that easily surpasses Gen-2. If 10 seconds sounds short compared to Sora’s full minute of video, consider that the company is working with a shoestring budget of compute compared to more lavishly funded OpenAI—and actually has a history of shipping video generation capability to commercial users.
Gen-3 Alpha does not generate audio to accompany the video clips, and it’s highly likely that temporally coherent generations (those that keep a character consistent over time) are dependent on similar high-quality training material. But Runway’s improvement in visual fidelity over the past year is difficult to ignore.
AI video heats up
It’s been a busy couple of weeks for AI video synthesis in the AI research community, including the launch of the Chinese model Kling, created by Beijing-based Kuaishou Technology (sometimes called “Kwai”). Kling can generate two minutes of 1080p HD video at 30 frames per second with a level of detail and coherency that reportedly matches Sora.
Gen-3 Alpha prompt: “Subtle reflections of a woman on the window of a train moving at hyper-speed in a Japanese city.”
Not long after Kling debuted, people on social media began creating surreal AI videos using Luma AI’s Luma Dream Machine. These videos were novel and weird but generally lacked coherency; we tested out Dream Machine and were not impressed by anything we saw.
Meanwhile, one of the original text-to-video pioneers, New York City-based Runway—founded in 2018—recently found itself the butt of memes that showed its Gen-2 tech falling out of favor compared to newer video synthesis models. That may have spurred the announcement of Gen-3 Alpha.
Gen-3 Alpha prompt: “An astronaut running through an alley in Rio de Janeiro.”
Generating realistic humans has always been tricky for video synthesis models, so Runway specifically shows off Gen-3 Alpha’s ability to create what its developers call “expressive” human characters with a range of actions, gestures, and emotions. However, the company’s provided examples weren’t particularly expressive—mostly people just slowly staring and blinking—but they do look realistic.
Provided human examples include generated videos of a woman on a train, an astronaut running through a street, a man with his face lit by the glow of a TV set, a woman driving a car, and a woman running, among others.
Gen-3 Alpha prompt: “A close-up shot of a young woman driving a car, looking thoughtful, blurred green forest visible through the rainy car window.”
The generated demo videos also include more surreal video synthesis examples, including a giant creature walking in a rundown city, a man made of rocks walking in a forest, and the giant cotton candy monster seen below, which is probably the best video on the entire page.
Gen-3 Alpha prompt: “A giant humanoid, made of fluffy blue cotton candy, stomping on the ground, and roaring to the sky, clear blue sky behind them.”
Gen-3 will power various Runway AI editing tools (one of the company’s most notable claims to fame), including Multi Motion Brush, Advanced Camera Controls, and Director Mode. It can create videos from text or image prompts.
Runway says that Gen-3 Alpha is the first in a series of models trained on a new infrastructure designed for large-scale multimodal training, taking a step toward the development of what it calls “General World Models,” which are hypothetical AI systems that build internal representations of environments and use them to simulate future events within those environments.
Google’s continued abuse of the Fitbit brand is continuing with the shutdown of the web dashboard. Fitbit.com used to be both a storefront and a way for users to get a big-screen UI to sift through reams of fitness data. The store closed up shop in April, and now the web dashboard is dying in July.
In a post on the “Fitbit Community” forums, the company said: “Next month, we’re consolidating the Fitbit.com dashboard into the Fitbit app. The web browser will no longer offer access to the Fitbit.com dashboard after July 8, 2024.” That’s it. There’s no replacement or new fitness thing Google is more interested in; web functionality is just being removed. Google, we’ll remind you, used to be a web company. Now it’s a phone app or nothing. Google did the same thing to its Google Fit product in 2019, killing off the more powerful website in favor of an app focus.
Dumping the web app leaves a few holes in Fitbit’s ecosystem. The Fitbit app doesn’t support big screens like tablet devices, so this is removing the only large-format interface for data. Fitbit’s competitors all have big-screen interfaces. Garmin has a very similar website, and the Apple Watch has an iPad health app. This isn’t an improvement. To make matters worse, the app does not have the features of the web dashboard, with many of the livid comments in the forums on Reddit calling out the app’s deficiencies in graphing, achievement statistics, calorie counting, and logs.
The web dashboard.
Fitbit
Google bought Fitbit back in 2021 and has spent most of its time shutting down Fitbit features and making the products worse. Migrations to Google Accounts started in 2022. The Google Assistant was removedfrom Fitbit’s 2022 product line, the Sense 2 and Versa 4, when support existed on the previous models. Social features—a key part of fitness motivation for many—were killed off in 2023. Google has mostly focused on making Fitbit an app for the Pixel Watch.
