OpenAI says ChatGPT is once again available in Italy after it addressed a series of conditions set out by regulators. The Garante data protection authority wanted OpenAI to resolve several issues by the end of this month in order to lift a temporary ban on the chatbot. 

“ChatGPT is available again to our users in Italy,” OpenAI told the Associated Press in a statement. “We are excited to welcome them back, and we remain dedicated to protecting their privacy.”

Italian regulators blocked ChatGPT in March over concerns that the AI's training methods and chatbot violated the European Union's General Data Protection Regulation (GDPR). Officials ordered OpenAI to stop processing the personal information of users in the country while they looked into a potential data breach.

Earlier this month, the Garante told OpenAI to make sure there was a privacy disclosure on the website that details how data is collected and used to train ChatGPT's algorithms. The watchdog also directed the company to offer a form through which EU users could object to having their data used for training purposes and to add an age verification system.

OpenAI added privacy controls to ChatGPT this week. Among other things, the settings allow users to switch off their chat histories, which stops OpenAI from using that data for training purposes. According to TechCrunch, users in Italy are now seeing a pop-up that asks them to confirm they are aged at least 18 or, if they're between 13 and 17, that they have a parent's or guardian's consent to use ChatGPT.

Italy's data protection authority said in a statement that it "welcomes the measures OpenAI implemented," though it urged the company to follow through on other demands that the Garante has made. "This applies in particular to implementing an age verification system and planning and conducting an information campaign to inform Italians of what happened as well as of their right to opt-out from the processing of their personal data for training algorithms."

The watchdog added that it hopes OpenAI will keep up efforts to comply with the GDPR and suggested it will be keeping an eye on ChatGPT. "The Italian [supervisory authority] will carry on its fact-finding activities regarding OpenAI also under the umbrella of the ad-hoc task force that was set up by the European Data Protection Board," the statement concluded.

Despite suffering a significant blow this week in its attempt to take over Activision Blizzard, Microsoft still believes it can get the job done. The company has signed a 10-year agreement with Spain-based cloud gaming provider Nware to make PC games it builds in-house available to stream on that platform, along with Activision Blizzard titles if and when the $68.7 billion deal goes through.

"While it's still early for the emerging cloud segment in gaming, this new partnership combined with our other recent commitments will make more popular games available on more cloud game streaming services than they are today," Microsoft president and vice-chair Brad Smith wrote on Twitter. "We are full speed ahead in our mission to bring players more ways to play their favorite games," Microsoft Gaming CEO Phil Spencer added.

In a similar fashion to NVIDIA's GeForce Now, Nware offers users access to a number of their game libraries via the cloud including those from Steam, Epic Games and Ubisoft. Users can use the service to stream their games on Windows PCs, Android smartphones, tablets and smart TVs.

The latest agreement follows similar deals Microsoft has struck with Nintendo, Steam, NVIDIA, Boosteroid, Ubitus and EE to make its games available on those companies' platforms. Microsoft says it has offered Sony, the biggest opponent of the proposed Activision merger, a 10-year agreement to keep the likes of Call of Duty on PlayStation. However, Sony hasn't taken up the offer.

These partnerships are part of a charm offensive Microsoft and Activision have been carrying out in an attempt to win over antitrust regulators. European Union officials will decide whether to rubber-stamp the blockbuster deal by May 22nd. Reports suggest that the European Commission will approve the acquisition.

However, Microsoft and Activision ran into a major obstruction this week when the UK's Competition and Markets Authority (CMA) blocked the deal. The watchdog voiced concerns over Microsoft's dominant position in the cloud gaming market.

Documents submitted to the CMA showed that Microsoft already had a cloud gaming market share of between 60 and 70 percent in 2022 — that was before Google killed off Stadia, which had under five percent of the market. Smith has attempted to counter the CMA's concerns by claiming that Microsoft's infrastructure only allows it to offer cloud gaming access to a maximum of 5,000 concurrent users in the UK.

Microsoft and Activision Blizzard plan to appeal the CMA's decision, a process that may take many months. The companies had hoped to close the deal by this summer. In December, the US Federal Trade Commission sued to block the takeover. The case is set to go to trial in August.

A former Apple employee who pled guilty to defrauding the company out of over $17 million has been sentenced to three years in prison and ordered to repay his ill-gotten gains. Dhirendra Prasad, who primarily worked as a buyer in Apple's Global Service Supply Chain department during his time at the company, admitted to charges of conspiracy to commit mail fraud and conspiracy to defraud the United States in November. As iMore notes, two charges of conspiracy to commit money laundering were dismissed during sentencing.

Prasad joined Apple in 2008 and carried out his schemes between 2011 and 2018. According to the US Attorney's Office for the Northern District of California, he conspired with two vendors to conduct fraud against Apple by "taking kickbacks, stealing parts, inflating invoices and causing Apple to pay for items and services it never received." Prasad is said to have used his insider knowledge of Apple's fraud detection practices to avoid being caught for several years.

The government has already seized $5.5 million worth of assets from Prasad, who a judge ordered to pay an additional forfeiture money judgment of $8.1 million. On top of that sum and the $17.4 million Prasad has been told to give back to Apple, he was ordered to pay $1.9 million to the Internal Revenue Service — he conceded that he did not pay tax on the proceeds of his schemes. Following Prasad's time in prison, he will have to serve three years of supervised release.

Xbox gamers looking for a headset that both supports spatial audio and looks pretty sleek may be interested in checking out the Xbox Wireless Headset. Although there are a number of other options on the market, this model ties into the aesthetic of the Xbox Series X (and previous Xboxes, given the green and black color scheme). Best of all, the Xbox Wireless Headset is on sale right now. It has dropped by $15 to $85 at Amazon.

The headset can connect directly to your Xbox Series X/S or Xbox One without the need for a dongle. It's compatible with Windows 10 and 11 devices. There's the option to connect it to your phone and Xbox simultaneously, so you can take a call without interrupting your gaming session. Additionally, there's Bluetooth and USB-C support and you'll get up to 15 hours of use on a single charge, Microsoft says.

There's support for Windows Sonic, Dolby Atmos and DTS Headphone:X spatial audio protocols, though you'll need a $15 license to use Dolby Atmos beyond any free trial that's on offer. Spatial audio can, for instance, help you figure out which direction enemies are approaching from.

You'll be able to use rotating earcup dials to adjust the volume and chat levels. The headset uses dual beamforming microphone elements and voice isolation tech to optimize your in-game voice chat. There's an auto-mute option that kicks in when you aren't talking, as well as a manual mute button. Moreover, the headset has an adjustable headband with what Microsoft describes as a flexible and lightweight design.

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

Meta will soon start rolling out its latest Quest firmware, and there's a particularly handy feature for Meta Quest Pro owners. The company's highest-end VR headset is gaining WiFi 6E support. So, if you have a compatible router and a fast enough internet connection, your headset will supports speeds of up to 1.6Gbps.

Faster downloads could come in useful for those who try another new feature, which will be available on all models (except the first one, unfortunately). After you install the v53 firmware, you'll have the option to automatically update all of your apps before the headset shuts down. Meta says this should help avoid having to wait for an app update when you pick up your headset again. This might only be truly effective for more ardent Quest users, though. If you tend to go a few weeks between VR sessions, you might still end up having to update apps when you put your headset on again. You can opt out of this feature from the power off dialog. 

Elsewhere, there's another tool to help parents and guardians manage what teens can access in the Quest browser. The Quest Browser Website Category Filter is fairly self-explanatory. It can be used to block websites that fall under certain categories, such as gambling or hate and terrorism. The filters are tied to accounts, not devices, and can be set up from the parental supervision tab in the Quest mobile app.

Speaking of the Meta Quest Browser, it'll possible to watch Twitch there (unless perhaps a parent or guardian says otherwise). On top of that, Meta is moving some advanced camera settings it has been testing from the Experimental Settings panel into the core camera settings. Finally, users will at last have the option to record video with stereo audio instead of only mono.

Forza Motorsport developer Turn 10 Studios has previewed some of the game's accessibility features. First and foremost, the studio is hoping to make it easier for blind and low-vision players to enjoy the racing sim with the help of audio cues.

Turn 10 spent two years building Blind Driving Assists with the help of accessibility consultant Brandon Cole (who is blind) and other blind and low-vision players. The feature offers supplemental audio cues to help players navigate. They'll receive audible information on things like track position and orientation, and their approach and progress during turns. Players will also learn details about the car, including the level of deceleration needed and when to shift gears if they're using manual transmission.

In a video that shows off the feature and details its development, Cole notes that the steering guide option pans the engine noise and tire sounds to the left or right, depending on the direction players should turn. Beeping sounds will tell players when they're nearing the edge of the track. Cole says that, with the help of the cues, he was able to win a race against a bunch of AI-driven cars.

Blind and low-vision players can preview the audio cues in the accessibility menu. They'll hear in-depth descriptions of what each cue means, thanks to the inclusion of a customizable screen narrator. Players can individually switch each set of cues on and off, and they'll be able to tweak the pitch and volume of them. Although it will likely take a while for players to get used to the cues, they could be immensely useful for blind and low-vision folks who want to play Forza Motorsport.

Along with more common accessibility features such as full controller remapping, colorblindness filters and customizable subtitles, there's another called One Touch Driving. This enables players to customize braking, steering and throttle assists to minimize the number of simultaneous inputs needed. They can enable as many of the assists as they like. Turn 10 worked with players who have mobility and stamina disabilities to create One Touch Driving, which is intended for those who may find it difficult to hold down buttons or press several buttons at the same time.

Forza Motorsport is slated to arrive later this year. It will be available on Xbox Series X/S, Xbox Cloud Gaming and PC via the Microsoft Store and Steam.

After a string of delays and a scrubbed launch attempt, SpaceX finally conducted the first test flight of its Starship spacecraft earlier this month. While the vehicle got off the ground, it seems federal agencies will be dealing with the explosive fallout of the mission for quite some time.

Federal agencies say the launch led to a 3.5-acre fire on state park land. The blaze was extinguished. Debris from the rocket, which SpaceX said it had to blow up in the sky for safety reasons after a separation failure, was found across hundreds of acres of land. “Although no debris was documented on refuge fee-owned lands, staff documented approximately 385 acres of debris on SpaceX’s facility and at Boca Chica State Park,” the Texas arm of the US Fish and Wildlife Service told Bloomberg.

The agency noted it hasn’t found evidence of dead wildlife as a result of the incident. Still, it’s working with the Federal Aviation Administration on a site assessment and post-launch recommendations, while ensuring compliance with the Endangered Species Act.

Soon after the launch and Starship’s explosion, the FAA said it was carrying out a mishap investigation. Starship is grounded for now and its return to flight depends on the agency “determining that any system, process or procedure related to the mishap does not affect public safety.”

Starship’s approved launch plan included an anomaly response process, which the FAA says was triggered after the spacecraft blew up. As such, SpaceX is required to remove debris from sensitive habitats, carry out a survey of wildlife and vegetation and send reports to several federal agencies. “The FAA will ensure SpaceX complies with all required mitigations,” the agency told Bloomberg.

Even if SpaceX can sate federal agencies' concerns swiftly, it may be quite some time until the next Starship launch. The super heavy-lift space launch vehicle destroyed its launch pad, sending chunks of debris into the air. Footage showed the shrapnel landing on a nearby beach and even hitting a van hundreds of yards from the launch site. Fortunately, no one was hurt, according to the FAA.

Twitter users who have had the scarlet letter of a checkmark pinned to their profile against their wishes may soon be able to have it removed. They may soon see a page telling them how to cancel the subscription.

According to researcher Nima Owji, the company is set to add a page to its app that tells affected users they can contact Twitter support to cancel Blue. Twitter no longer has a communications team that can be reached for comment.

As it had long been promising, Twitter last week finally removed checkmarks from all of the accounts that were verified through the previous system because of their notability if they hadn't already subscribed to Blue. However, Twitter and its owner Elon Musk added blue checkmarks back to certain accounts anyway, namely ones that belong to celebrities (including some dead folks) or have more than a million users.

The likes of Lil Nas X, Bette Midler and Stephen King said they didn't pay for Blue even though they had a checkmark. The icon appeared on the accounts of several dead celebrities as well. Musk suggested that he was trolling some users by putting the checkmark back on their accounts.

However, some users suggested that, in doing so, Twitter and Musk may have violated the Lanham Act, a US federal law that prohibits false endorsement. It could be argued that by having a checkmark on their account without paying for Blue, users may tacitly appear to be endorsing it.

Offering clear directions to users who never wanted Blue in the first place on how to cancel their subscription may come as small comfort, especially for those who've found themselves on the wrong side of the #BlockTheBlue trend. Still, it remains unclear how deceased celebrities like Norm Macdonald, Chadwick Boseman and Kobe Bryant are supposed to have the checkmark removed, unless their families check their Twitter account settings on the regular.

A bipartisan group of senators has introduced a bill that seeks to protect kids from the potentially harmful effects of social media. The Protecting Kids on Social Media Act would set 13 as the minimum age for using social media platforms, in line with the Children’s Online Privacy Protection Act (COPPA) and the current terms of many major services. Those aged between 13 and 17 would need parental consent to use social media apps, while platforms would not be allowed to use algorithms to recommend content to under 18s.

Democratic Sens. Brian Schatz and Chris Murphy introduced the bill alongside their Republican counterparts Tom Cotton and Katie Britt. The proposed legislation seeks to employ age verification technology, though it would prevent companies from using age verification data for other purposes. Were it to become law, the bill would establish a pilot program to offer platforms a government-provided age verification system. The Federal Trade Commission (FTC) and state attorneys general would have jurisdiction over the bill's provisions.

"Studies have shown a strong relationship between social media use and poor mental health, especially among children," the senators said in a press release. "With this clear evidence, the US Surgeon General has warned that 13 is too early for social media use and suggested that 16, 17, or 18 may be as well."

"The alarm bells about social media’s devastating impact on kids have been sounding for a long time, and yet time and time again, these companies have proven they care more about profit than preventing the well-documented harm they cause," Murphy said. "In particular, these algorithms are sending many down dangerous online rabbit holes, with little chance for parents to know what their kids are seeing online.”

This is hardly the first time that federal legislators have introduced bills in an attempt to protect children as they use online services. Sens. Marsha Blackburn and Richard Blumenthal last year introduced the Kids Online Safety Act, which aimed to block platform holders from offering algorithmic recommendations to under 16s. Such companies would have been required to disable "addictive" features and offer kids tools to help them protect their data. The bill didn't make it out of committee before the end of the last congressional session.

Blumenthal and Sen. Dick Durbin introduced a bill in February that would give people the right to delete details of their internet activity that companies collected before they turned 13. In a related case as part of an FTC settlement related to COPPA, Epic Games last year agreed to delete Fortnite players' personal data unless they claimed to be 13 or older.

The Protecting Kids on Social Media Act follows similar efforts at state level. Last month, Utah became the first state to require parental consent for under-18s to use social media. The legislation grants parents full access to their children'' accounts, including private messages. The laws are set to take effect next March, but it's not yet clear how they'll be enforced.

Philips Hue might command much of the attention when it comes to smart light bulbs, but there are other options that allow you to control your home lighting from your phone without having to spend quite as much. Govee, for instance, offers some more affordable solutions, and several of the company's products are currently on sale at Amazon. Among them is a two-pack of 33-foot, color-changing LED strips. After a $20 coupon is applied at checkout, the light strips can be all yours for $57. That's $25 off the typical price.

Govee's strip lights have a protective coating that's said to protect them from moisture and dust. The company says that, as opposed to RGB lighting, its RGBIC approach enables it to display "stunning segmented colors" on a line. 

You can connect the strips directly to your WiFi router and control them using Alexa or Google Assistant. Otherwise, you can customize the lighting through Govee's app. There's also the option to sync the lighting to music. It's worth noting that the bundle comes with one power adaptor that has two connection ports, so you'll need to keep the two LED strips next to each other.

Also in Govee's Amazon sale (which includes a set of string lights that has dropped in price by 41 percent), there's a solid deal on a floor lamp. This product works with Alexa and it has support for music sync as well. The lamp has multiple sections, each of which can display different colors. Curiously, the Amazon listing notes that buyers will either be sent an old or a new version at random. The old one is 55.5 inches, while the newer smart lamp measures 53.7 inches. In any case, you can currently snap one up for $80, which is $20 off the usual price.

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