The Federal Communication Commission is moving forward with its plan to ban AI robocalls. Commissioners voted unanimously on Wednesday in favor of a Declaratory Ruling that was proposed in late January. Under the measure, the FCC deems robocalls made using AI-generated voices to be "artificial" voices per the Telephone Consumer Protection Act (TCPA). That makes the practice illegal. The ruling takes effect immediately.

“Bad actors are using AI-generated voices in unsolicited robocalls to extort vulnerable family members, imitate celebrities and misinform voters. We’re putting the fraudsters behind these robocalls on notice,” FCC Chairwoman Jessica Rosenworcel said in a statement. “State Attorneys General will now have new tools to crack down on these scams and ensure the public is protected from fraud and misinformation.”

The TCPA is a 1991 law that bans artificial or recorded voices being used to call residences without the receivers' consent. It's up to the FCC to create rules to enforce that legislation, as Ars Technica notes. As the FCC pointed out last month, under the TCPA, telemarketers need "to obtain prior express written consent from consumers before robocalling them. If successfully enacted, this Declaratory Ruling would ensure AI-generated voice calls are also held to those same standards."

The FCC vote in favor of the ban comes at somewhat of an inflection point for AI. Not only have such technologies become vastly more widespread over the last year or so, an AI-generated version of President Joe Biden's voice was used in a recent robocall that urged Democrats not to vote in New Hampshire's Presidential primary. A criminal investigation into that incident is underway.

Given that we're in an election year and the volume of misinformation and disinformation is already likely to rise, clamping down on AI robocalls now seems like a wise move. While stage AGs can take action against robocallers, the FCC also has the ability to fine them under the TCPA. Last year, the agency issued its largest ever fine of $300 million last year against a company that made more than 5 billion robocalls in a three-month period.

Just as Microsoft renamed Bing Chat to Copilot to unify its generative AI branding, Google is doing the same thing with Bard and Duet AI. The services now bear the name Gemini, after Google's multimodal AI model. The name change leaked earlier this month. Google has also debuted a dedicated Gemini Android app alongside a paid version of the chatbot that has more enhanced capabilities.

"Bard has been the best way for people to directly experience our most capable models," Google CEO Sundar Pichai wrote in a blog post. "To reflect the advanced tech at its core, Bard will now simply be called Gemini. It’s available in 40 languages on the web and is coming to a new Gemini app on Android and on the Google app on iOS."

Those who download the Gemini Android app can actually replace Google Assistant as the default assistant on their device. So, when you long press the home button or utter "Hey Google," your phone or tablet can fire up Gemini instead of Assistant. You can also make this switch by opting in through Assistant.

Doing so will enable a new conversational overlay on your display. Along with swift access to Gemini, the overlay will offer contextual suggestions, such as the ability to generate a description for a photo you just took or ask for more information about an article that's on your screen.

You'll also be able to access commonly used Assistant features through the Gemini app, from making calls and setting timers to controlling smart home devices. Google said it will bring more Assistant functions to Gemini in the future. That certainly makes it sound as though Google is phasing out Assistant in favor of Gemini. The app also includes access to Gemini Advanced (more on that in a moment).

As for iOS, there won't be a separate Gemini app for now. Instead, you can access it through the Google app by tapping the Gemini toggle.

Gemini is available on Android and iOS in English in the US starting today. Next week, Google will start offering access to the chatbot in more locales in English, as well as in Japanese and Korean. As you might expect, Gemini is coming to more countries and languages down the line.

In addition, Google is opening up access to what it says is its largest and most capable AI model, Ultra 1.0, through Gemini Advanced. The company claims this is able to have longer and more in-depth conversations with the ability to recall context from previous chats. It says Gemini Advanced "is far more capable at highly complex tasks like coding, logical reasoning, following nuanced instructions and collaborating on creative projects."

Gemini Advanced is available now in English in 150 countries and territories. To access it, you'll need to sign up for the new Google One AI Premium Plan. This costs $20 per month — the same price as Copilot Pro — after a two-month free trial. Along with Gemini Advanced, this subscription includes everything from the Google One Premium Plan, including 2TB of storage and a VPN. Subscribers will also be able to use Gemini in apps such as Gmail, Docs, Slides and Sheets in the near future (this is replacing Duet AI).

Of note, Google says it sought to mitigate concerns such as bias and unsafe content while building Gemini Advanced and other AI products. The company says it carried out "extensive trust and safety checks, including external red-teaming" (i.e. testing by third-party ethical hackers) on Gemini Advanced before refining the model with reinforcement learning and fine tuning.

Sony's Funimation purchased Crunchyroll from WarnerMedia for $1.175 billion in 2020, and they kicked off their transformation into a unified anime subscription service under the latter's name a year after the deal was announced. By 2022, Crunchyroll has already added more than 50 shows that were either exclusive to Funimation and weren't available with dubs to its library. Now, it sounds like they're almost done unifying their services: Funimation has revealed that it's going to shut down its old app and website on April 2. 

In Funimation's End of Services' page, it said most of its content has already been migrated to Crunchyroll. Those who've yet to leave Funimation, will automatically be transferred — all they need to do is use their existing credentials to log into Crunchyroll's website. After they do log in, they'll get a prompt telling them that their Watch and History lists are being migrated, as well. 

Viewers who have a Funimation and a Crunchyroll account will be prompted to merge their data from both services or to choose to use their data from one of them. And after April 2, their billing will go through Crunchyroll's and will follow its pricing, which starts at $8 a month. Unfortunately, Funimation customers who own digital copies complementing the DVDs or Blu-rays they purchased will lose access to them because Crunchyroll does not support them. "[W]e are continuously working to enhance our content offerings and provide you with an exceptional anime streaming experience," Funimation said. "We appreciate your understanding and encourage you to explore the extensive anime library available on Crunchyroll."

Meta and TikTok owner ByteDance are not keen on the idea of paying the European Union to regulate them. The companies have challenged a supervisory fee set forth by EU moderators, who are now required to monitor Meta, TikTok, and other major platforms under the Digital Services Act (DSA), Politico reports. Meta first announced its action, with ByteDance following suit a day later.

Under the current arrangement, all designated companies must split the €45.2 million ($48.7 million) that EU's regulators argue is necessary to properly supervise the 20 Very Large Online Platforms and two Very Large Online Search Engines (VLOSEs). Each regulated platform has 45 million or more users with its financial contribution based on the size of that number. They also can't owe more than 0.05 percent of its 2022 net profits. However, companies like Amazon and Pinterest that reported little to no profits won't owe anything. Meta, on the other hand, got a €11 million ($11.9 million) bill under the current arrangement. ByteDance has not publicly announced how much it owes.

Meta takes issue with the European Union regulators' methodology for choosing each company's fees. "Currently, companies that record a loss don't have to pay, even if they have a large user base or represent a greater regulatory burden, which means some companies pay nothing, leaving others to pay a disproportionate amount of the total," a Meta spokesperson remarked. Failure to comply with the fee could lead to a fine of up to six percent of a company's global revenue.

The DSA went into effect in 2023, with Meta's and ByteDance's VLOP designations came alongside other major sites, like Google and X, formerly known as Twitter. Along with the fee, VLOPs must comply with specific regulations like transparent advertising and content moderation, sharing data with the European Commission and cooperating with an annual independent audit.

According to The Information, Apple has created two clamshell-style foldable iPhone prototypes, though they’re all in the early stages of development. If Apple were to move forward with foldable iPhones, they likely wouldn’t come to market until at least 2026.

Companies have so far been unable to address the technical issues of foldables. The first Samsung Galaxy Fold notoriously had a gap between the two halves of the screen, which debris fell into. Second, Apple’s designers have found it tough to create a foldable iPhone compelling enough to exist beyond the iPhone’s current form factor. On top of those challenges, engineers wanted to make each side half as thin as a regular iPhone, so it would be roughly the same thickness when folded. But the tech isn’t there yet, given battery sizes and display constraints.

It seems more likely that Apple’s first foldable device may be an iPad, which it’s reportedly been working on since 2020. The company is trying to avoid the crease that emerges in the center of foldable displays after repeated folds. Apple is also apparently trying to make sure the screen is totally flat when it’s opened, with no bump in the middle.

— Mat Smith

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

​​

The biggest stories you might have missed

Apple Vision Pro teardown reveals pixels the size of red blood cells

Ring announces a new battery-powered doorbell with 3D motion detection and improved visuals

Layoffs are sucking the joy out of video games

Bluesky has added almost a million users one day after opening to the public

YouTube TV adds enhanced 1080p option

The company says it’s the ‘highest video quality’ it can offer.

Google is rolling out a new setting called 1080p Enhanced, for YouTube TV and Primetime Channels subscribers. The 1080p Enhanced setting improves on the existing 1080p60 resolution with its upgraded bitrate. While it’s the same resolution, the better bitrate will give better images.

Google confirmed the update after a Reddit user posted about it on the platform. The 1080p Enhanced setting “delivers our highest video quality,” said the response. Reddit users who already have access to 1080p Enhanced report the resolution is available for all of the same channels as 1080p60, such as Paramount and Syfy.

Continue reading.

OnePlus 12 review

A no-nonsense flagship for a great price.

Engadget

The OnePlus 12 has the same speedy performance and better battery life than the Galaxy S24+, along with solid cameras and a great screen for $200 less. There are few if any AI tricks, but that doesn’t detract much from great battery life, high-res cameras and that $800 price. Check out our full review.

Continue reading.

ChatGPT will digitally tag images generated by DALL-E 3

To battle misinformation.

OpenAI is now including provenance metadata in images generated with ChatGPT on the web and DALL-E 3 API. The metadata follows the C2PA (Coalition for Content Provenance and Authenticity) open standard, so when an image is uploaded to the Content Credentials Verify tool, you can trace its source. Unfortunately, it’s all rather easily bypassed: It only works when the metadata is intact. It doesn’t work if you upload an AI-generated image sans metadata — like any screenshot or images pulled from social media.

Continue reading.

Homeworld 3, the much-anticipated sequel to 20-year-old real-time strategy game Homeworld 2, has been delayed once again, Gearbox Publishing and Blackbird Interactive announced. It's now set to come out on May 13, 2024, a couple of months after the previously announced March 8 debut.

The decision to delay was made following a playthrough by users from outside the company. "This resulted in additional insights and perspective the will be incorporated to make Homeworld 3 the best experience possible," the team wrote. A public demo also likely weighed on the decision, showing some issues like balky camera and other controls.

Gearbox Publishing/Blackbird Interactive

Homeworld 3 was originally pegged for a 2022 release, but it was postponed until 2023 and later February 2024, then again slightly until March 8. Though public beta users found some elements balky, the game generated positive comments as well, so the new May 13 date may be realistic. 

Homeworld 3 is an ambitious sequel to its popular predecessor, attempting to capture the vibe that people loved while introducing new modes including PvP and co-op. The latter (War Games) has a roguelike aspect emphasizing coordination with fellow players and may be the new title's ace in the hole. In any case, May is a generally a quiet month for game releases, so if the latest date works out, it would be a fine time to get into a complex RTS. 

Some devices easily stand out in their category, and in the case of headphones, that's Sony's WH-1000XM5 model. If you've been coveting our favorite wireless headphones, there's good news: Sony's WH-1000XM5 headphones are currently down to $280 from $400 — a 30 percent discount. Woot is offering them at this low price for the next four days or until they sell out. Though Amazon owns Woot, it doesn't have the same return policy as its parent company.

Sony's WH-1000XM5 are great for many reasons — we gave them a 95 in our review for a reason — but one of their best features is how incredibly comfy they are when worn. They only weigh 0.14 ounces less than the WH-1000XM4s, but improved weight distribution and a slimmer headband mean they feel much lighter on your head and ears. 

Tech-wise, the M5s offer double the processors and microphones devoted to noise canceling compared to the M4. Then there are 30mm carbon fiber drivers, which give music greater depth, and DSEE Extreme, which boosts the sound quality. The only aspect of the headphones that's frustrating compared to their predecessor is storage. The M5s don't fold, so their storage case is bulky to carry around. 

Follow @EngadgetDeals on Twitter and subscribe to the Engadget Deals newsletter for the latest tech deals and buying advice.

A key feature of Apple's Vision Pro VR, er, spatial computing headset is Personas that that lets people see a digital version of themselves during calls, Zoom meetings, etc. At launch, they looked a bit creepy, but Apple has improved them considerably in the latest release, according to posts on X spotted by MacRumors. They're now more realistic, so users look less like impressionist paintings and more like humans. 

Once the visionOS 1.1 update is installed, you'll be prompted to recapture your Persona to get the "latest appearance updates" — this is apparently done in part with the headset off and pointing at your face. Most users feel the updated Personas are better, and visually, they look less blurry and a touch more realistic, plus the proportions seem better.

As a reminder, Personas are a digital representing of the users created by data from the headset's sensors and a new machine learning algorithm. They're designed to be an "authentic spatial representation" of someone that shows their facial expressions and hand movements — so you won't be a dork in a mask on group calls.

If you need to do that, though, a similar feature called EyeSight shows other people a digital version of your eyes when you're wearing the Vision Pro. Apple still calls Personas a beta feature and provides detailed instructions to get the best results.

While Toyota helped lead the hybrid charge with its Prius, the company has been less active in EV production. But that might be in the past, as Toyota has announced another $1.3 billion for its Kentucky facility, with a focus on EVs. In particular, the company is building a new three-row electric SUV that will be available for US customers. 

Toyota's Kentucky operation is its flagship facility, with nearly 9,400 employees. "Today's announcement reflects our commitment to vehicle electrification and further reinvesting in our US operations," Kerry Creech, president of Toyota Kentucky, said in a release. "Generations of our team members helped prepare for this opportunity, and we will continue leading the charge into the future by remaining true to who we are as a company and putting our people first for generations to come."

Little is known so far about the three-row electric SUV, as Toyota only mentions it briefly in relation to the investment. However, it follows the company's prior announcement of its Urban SUV concept slated to launch in Europe this year. 

The $1.3 billion investment will also go towards creating a battery pack assembly line at its Kentucky factory (a separate manufacturing plant in North Carolina produces the batteries). Toyota has big plans for its battery production, previously announcing three battery ranges focused on higher performance and low cost slated for 2026 and onward. The company is also working on its first solid-state batteries, which have the potential to charge an EV from 10 to 80 percent in 10 minutes. Toyota's bZ4X EV currently takes 32 minutes to do so. 

Apple isn't one of the top players in the AI game today, but the company's new open source AI model for image editing shows what it's capable of contributing to the space. The model called MLLM-Guided Image Editing (MGIE), which uses multimodal large language models (MLLMs) to interpret text-based commands when manipulating images. In other words, the tool has the ability to edit photos based on the text the user types in. While it's not the first tool that can do so, "human instructions are sometimes too brief for current methods to capture and follow," the project's paper (PDF) reads. 

The company developed MGIE with researchers from the University of California, Santa Barbara. MLLMs have the power to transform simple or ambiguous text prompts into more detailed and clear instructions the photo editor itself can follow. For instance, if a user wants to edit a photo of a pepperoni pizza to "make it more healthy," MLLMs can interpret it as "add vegetable toppings" and edit the photo as such.

Apple

In addition to changing making major changes to images, MGIE can also crop, resize and rotate photos, as well as improve its brightness, contrast and color balance, all through text prompts. It can also edit specific areas of a photo and can, for instance, modify the hair, eyes and clothes of a person in it, or remove elements in the background. 

As VentureBeat notes, Apple released the model through GitHub, but those interested can also try out a demo that's currently hosted on Hugging Face Spaces. Apple has yet to say whether it plans to use what it learns from this project into a tool or a feature that it can incorporate into any of its products.