Airlines have been increasingly outsourcing repair and upkeep work to other countries, but experts and consumer groups disagree about its impact on safety.

Spain uses an algorithm to score how likely a domestic violence victim is to be abused again. A Times investigation last year identified flaws in the system.

The company’s Apple Intelligence system has erroneously characterized news stories, provoking a backlash from media companies.

• Hinge is employing AI to help with prompts
• It won't write them for you
• There are different levels of guidance

If the thought of crafting the perfect dating profile fills you with dread, Hinge has an AI solution. The popular dating app has rolled out a new feature called Prompt Feedback that uses AI to coach you in writing better responses to profile prompts. The company thinks AI assistance can make people's responses better for showcasing themselves or at least be less basic and boring.

According to Hinge, Prompts are 47% more likely to lead to dates than photo likes, but over half of surveyed Hinge daters admitted they don’t know how to craft good responses. Prompt Feedback's AI evaluates a dater’s profile responses and offers personalized advice on how to tweak them for maximum impact.

The AI is fine-tuned with insights from Hinge’s team of behavioral scientists. Notably, it doesn't rewrite your answer or make one up for you. It simply suggests ways to think about your answer and how to make it better, more detailed, and more likely to entice potential romantic partners. It's the supportive best friend in a movie who tells you, “Maybe skip the joke about your childhood pet hamster and mention your love of hiking instead.”

AI romance

(Image credit: Hinge)

The feature lives in the settings menu, where users can review feedback privately. Users will receive one of three levels of guidance: a simple “Great Answer” if the response is already on point, “Try a Small Change” for a bit of fine-tuning, or a “Go a Little Deeper,” which is the AI’s polite way of saying, “You can do better than ‘I like long walks on the beach.’”

Hinge isn't unique in applying AI to dating profiles. Tinder uses an AI feature that can help pick out profile photos, while Grindr is developing a digital wingman of its own. Meanwhile, Bumble offers photo help as well as AI to help users design their entire profile and get conversations between users off to a good start.

Writing a good profile takes effort, and dating app fatigue is real. For some users, the idea of AI critiquing their efforts might feel intimidating or unnecessary. For others, this could be the romantic nudge they need. Unlike the harsh criticism you might get from a sibling or brutally honest friend, this feedback is entirely optional. But if you’re open to a little constructive coaching, the tool is there to help.

You might also like...

• Bumble is giving you new AI tools to find love
• Tinder's new AI will pick out your best photos for your dating profile
• Can Grindr's AI Cupid help you find love?

• Luma Labs has released the new Ray 2 AI video generator
• Ray 2 turns text prompts into high-quality 5–10 second videos
• Ray 2 is available through the Dream Machine platform

Luma Labs has premiered a powerful, new AI model for generating videos on its Dream Machine platform called Ray 2. The new model can produce an array of realistic video clips of up to 10 seconds, from recreating a bee pollinating flowers to more surreal ideas like the typing anthropomorphic banana seen above.

The beauty of Ray 2 isn’t just its ability to render these wild scenarios but to do so with motion and physics that look shockingly natural. Unlike earlier video generation tools, which often struggled to produce anything faster than a leisurely stroll, you can see people really book it in a run. Ray 2 is capable of this level of production due in part to Luma training it on ten times more computational power than its predecessor, Ray 1. That means more realistic characters, faster rendering, smoother motion, and far fewer glitches.

Ray 2 is available through Luma’s Dream Machine platform, which offers both free and paid subscription tiers. The free plan lets users dabble with 720p resolutions, while paid plans unlock higher-quality 1080p visuals and unlimited usage if you’re willing to drop $66.49 a month.

AI Video Sprint

Surrealism: "a photorealistic anthropomorphic banana man doing taxes" pic.twitter.com/OQCFaSRRWEJanuary 15, 2025

Luma has plans to expand Ray 2’s capabilities with image-to-video, video-to-video, and editing tools. That could mean letting you turn a vacation photo into a short video or remixing a home movie into something cinematic.

The company is also hosting the Ray 2 Awards, offering creators a chance to win up to $7,000 in prizes. There’s a $5,000 prize for the most-viewed Ray 2 video on social media and a $3,000 raffle for anyone who shares their content and engages with Luma’s promotional posts. If nothing else, it’s a good excuse to finally bring your idea for “sloths competing in a high-stakes basketball game” to life.

Ray 2's limits mean it won't quite blow any competition away, however. The time limit makes it less capable than OpenAI's Sora model in some ways. Sora focuses on creating longer-form, cohesive video narratives. Then there's Runway's Gen-2, which brings users tools to tweak lighting, camera angles, and more, and Pika, which regularly drops new features like picture-to-video ability that Luma is still developing.

Still, Ray 2 has its charm and feels a bit like a streamlined alternative for those who prioritize speed and ease of use. The real significance of Ray 2 lies in how it helps lower the barrier to entry for anyone looking to make a video with AI. Even if that’s something as weird as a banana typing a note.

You might also like...

• Dream Machine 1.5 catches Sora and other rival AI video makers napping
• OpenAI Sora officially launches to change AI video – 5 things you need to know
• Pika challenges OpenAI and Sora with new AI video generator features

Dylan seemed to be in on the joke, posting an old black and white clip of himself saying “Good God, I must leave right away.”

The move is under discussion as the Chinese-owned app faces a Sunday deadline to find a new buyer or shut down in the United States.

An investigation by the Federal Trade Commission determined that consumers had not been aware that the automaker was providing their driving information to data brokers.

The successful flight to orbit of the Amazon founder’s powerful rocket suggests it could grow into a credible competitor with Elon Musk’s SpaceX.

Today, we're looking at a massive manufacturing deal that might have some life in it yet, despite an federal executive order opposing it; fresh return-to-office efforts at a banking titan and a new buyer stepping in to save a bankrupt retailer. First, however, we'll dig into the ways human-resource managers are already using artificial intelligence in the hiring process.

In their mass migration to the Chinese app RedNote, social media users make a gleeful mockery of the American government.

The cryptocurrency industry has pushed President-elect Donald J. Trump to establish one, creating real political momentum behind the plan.

Ahead of a Supreme Court ruling, they are mocking U.S. national security concerns about the Chinese-owned app.

• Microsoft's Windows 11 24H2 updates continue to affect multiple games on PC, notably Assassin's Creed titles.
• Ubisoft has been forced to fix Origins and Valhalla due to negative feedback from fans
• Odyssey remains unpatched with multiple crashes occurring

Microsoft's Windows 11 24H2 update issues are no secret at this point, affecting PC users across the board - and gamers are continuing to suffer, with Ubisoft's Assassin's Creed Origins and Valhalla only now receiving compatibility fixes.

As discussed in my article on the Asus ROG Ally using with Bazzite installed, Windows 11 24H2 has been a nightmare for gamers. In the case of Ubisoft's titles, Assassin's Creed Odyssey, Valhalla, and Origins have all been facing compatibility issues with constant crashes leaving them almost unplayable. While Ubisoft has released a patch for Origins and Valhalla, Odyssey still hasn't been fixed.

As reported by IGN, fans of the franchise have voiced their frustrations by sharing negative reviews across Ubisoft's games on Steam to force the developer to act, and while it may seem to some people that the fault lies with Ubisoft (especially with the bias of recent controversies surrounding Assassin's Creed Shadows), this isn't the case as the blame is on Microsoft.

Aside from games crashing, using Windows 11 has, for some people, become very frustrating with random bugs leaving audio inactive and the desktop unresponsive - it's getting from bad to worse for Microsoft, and it certainly won't do the company’s desire to get more people to switch to Windows 11 any good the longer it lasts.

Well, Bazzite and Valve's official SteamOS are both looking far more appealing now... (Image credit: Shutterstock) Don't point the finger at Ubisoft this time, it's Microsoft you should be frustrated with

I've always been a harsh critic when it comes to video game developers and poor optimization. Still, it seems Microsoft is responsible this time around - Windows 11 24H2 has caused problems for quite a few PC users, and doesn't look like it will stop any time soon. While a patch could have been released a little sooner by Ubisoft, the same could be said for other games that have been affected like Path of Exile 2.

Fortunately, Valve is bringing SteamOS to multiple handheld gaming PCs (not just its own Steam Deck) - but it isn't exactly clear whether we’ll be able to install the Linux-based gaming-focused operating system on desktop PCs (which would be a pleasant surprise), but Bazzite is already available. Since dual booting (where you choose between operating systems when your PC starts up) is possible it’s a great way to access games that can’t be played on SteamOS like Call of Duty: Black Ops 6 and other games that use anti-cheat systems.

There's no word on when we'll finally be rid of Windows 11 24H2's issues, and for many gamers, waiting for Microsoft to fix its latest update (or for developers to issue patches individually for their games) will be a frustrating experience. Installing an alternative operating system that’s less prone to problems and better optimized for gaming will start looking increasingly attractive for those gamers - and that’s something Microsoft should be worried about.

You may also like...

• I’m a big PC gamer – these were the 3 best gaming handhelds announced at CES 2025
• Windows 11 review
• Microsoft really wants you to update to Windows 11 in 2025

Shou Chew will join tech moguls like Mark Zuckerberg and Elon Musk at President-elect Donald J. Trump’s inauguration as the fate of the app hangs in the balance.

• Mercedes-Benz is adding Google Cloud’s Automotive AI Agent to its cars.
• The upgrade uses Google Gemini for real-time, personalized help.
• The new MBIX AI will appear first in new CLA models.

Google's dream of making its Gemini AI your road trip pal is becoming more real, at least if you own a Mercedes-Benz. Google Cloud’s Automotive AI Agent is making its debut in Mercedes-Benz’s updated MBUX Virtual Assistant. The assistant is built on Google’s Gemini AI to offer nuanced, conversational support.

The value is easy to understand. If you're on the road and hungry you could just ask your car, “Where’s the best Italian place around here?” Not only would you hear some options, but the AI would take the initiative to add, “The lasagna gets rave reviews, but avoid the tiramisu, it’s a little too soggy.”

The Automotive AI Agent is designed to handle multi-turn conversations and even remember the context of your previous chats. So, if hours after you ask about Italian restaurants, you inquire, "Are any of those places you mentioned before open late?" the AI won’t act like you’re starting from scratch. And the AI does know quite a lot partly because of Gemini's earlier integration with Google Maps. That means the AI can offer live traffic updates, nearby points of interest, and alternate routes. If you're stuck in a traffic jam, the car might suggest a quicker way home that keeps in mind your preference for avoiding toll roads.

MBUX Gemini

Google and Mercedes are betting that the ability of automotive AI agents to process complex queries and provide accurate, context-aware answers will make a difference when people decide what kind of car to buy. By making interactions smoother, more intuitive, and useful, this AI could make your drive easier. No more stopping on the shoulder to Google “nearest gas station,” and no more arguments about whose turn it was to check the directions.

That said, there’s a fine line between helpful and creepy. The assistant remembers conversations, which is great when it’s reminding you about that vegan café your partner wanted to try – but what happens when it casually recalls your midnight McDonald’s run from three weeks ago? The balance between utility and privacy will be a key factor in whether drivers fully embrace this technology.

You might also like

• Found code suggests that Google Gemini is coming to Android Auto
• Gemini AI starts riding shotgun on Google Maps
• ChatGPT gives Volkswagen cars the smarts to talk to you all day long

• Google Gemini will now offer up-to-date news
• Google has announced a partnership with The Associated Press
• You'll be able to get your daily news fix directly from the Gemini app

Google has announced that one of the world's biggest providers of news will now provide up-to-date bulletins directly in its AI chatbot, Gemini.

The major deal with The Associated Press was announced in a blog post on Wednesday, and is the first time Google has struck an AI news deal with a publisher.

This new addition to Gemini will give users access to AP's journalism right from within the chatbot, allowing you to quickly get news on the go without the need for a fact check.

The Associated Press Chief Revenue Officer, Kristin Heitmann said the deal looks to "provide timely, accurate news and information to global audiences.”

She added, "We are pleased Google recognizes the value of AP’s journalism as well as our commitment to nonpartisan reporting, in the development of its generative AI products,"

Google's blog announcement says "The AP will now deliver a feed of real-time information to help enhance the usefulness of results displayed in the Gemini app."

At the time of writing, when I ask Gemini for the news in its iOS app, the chatbot replies "Unfortunately, I cannot provide you with real-time, up-to-the-minute news updates. My knowledge of current events is limited."

Google hasn't detailed a timeline of when we can expect to see AP news in Gemini, but considering the language of the post, it shouldn't be too long.

A trustworthy news source

(Image credit: Google)

Google says: "For years, we’ve worked with The Associated Press (AP) to provide up-to-date and accurate information for features in Google Search. To build on that collaboration, the AP will now deliver a feed of real-time information to help further enhance the usefulness of results displayed in the Gemini app. This will be particularly helpful to our users looking for up-to-date information."

This major change to Gemini could see more users flocking to the chatbot to get their daily news fix. In the past, I've been disappointed with the lack of trustworthy news sources in AI chatbots like Gemini or OpenAI's chatbot, so the shift to include more publishers allows the end user to read information in these apps without worrying about their validity.

Would you use Gemini to read the news? Let us know in the comments section below.

You may also like

• Gemini is taking over some more Google Assistant tasks
• Samsung Galaxy Watch 6 owners poised to get Wear OS 5 soon – as Google Pixel Watch still struggles
• Google Gemini Explained

The company said its Columbus plant could eventually produce tens of thousands of autonomous systems and weapons each year.

Rules for a $7,500 tax break for electric vehicle purchases and leases recently changed, but more far-reaching changes are expected when President-elect Donald J. Trump takes office.

• Google Chrome's PermissionsAI test anticipates user permissions.
• PermissionsAI could reduce intrusive pop-ups by analyzing user behavior.
• It is currently in Chrome Canary for early testing now.

Google’s Chrome browser is testing a new feature dubbed PermissionsAI, which is designed to make those pesky pop-ups asking for access to your location or permission to send notifications much quieter.

The tool, first found by Windows Report, uses Google's “Permission Predictions Service” and Gemini Nano v2 to guess whether users are likely to grant a website’s request. If the answer is probably no, the feature tucks the request into a less intrusive UI instead of flashing it front and center like it currently does.

The idea is for Chrome to use AI to make browsing more pleasant by quietly handling the minor annoyances that can pile up when online. PermissionsAI analyzes your previous interactions with similar requests to predict your response. If you’re the kind of person who reflexively denies every notification pop-up, PermissionsAI won’t even bother you with a loud, in-your-face request. Instead, it quietly logs the request in a subtle UI where you can engage with it later.

PermissionsAI is currently being tested in Chrome Canary – the experimental version of the browser – and isn’t available to the public yet. It pairs well with the Safe Browsing security tool, which shields users from harmful websites and malicious downloads, meaning a wrong guess won't wreak havoc on your computer.

Browsing AI

This marks the latest move by Google to embed AI in Chrome. Gemini is now part of many of the browser's features, with the AI organizing open tabs, offering product comparisons, and helping to compose text. PermissionsAI, though, is less flashy than other features and may turn out to be one of the more appreciated improvements simply because it removes a common irritation.

While the concept is straightforward, the details of how PermissionsAI works remain vague. Google hasn’t disclosed exactly how its AI calculates the likelihood of you clicking “Allow” versus “Deny,” but it’s safe to assume the system leans heavily on machine learning.

By studying patterns in user behavior, Chrome could cut down on interruptions and make life a little easier for web developers who hear complaints about their pop-ups annoying people.

It’s worth asking whether PermissionsAI and other Gemini-powered features will strike the right balance between helpful and intrusive. While reducing the noise of pop-ups is universally appealing, AI-powered predictions aren’t infallible.

What happens if PermissionsAI incorrectly predicts that you wouldn’t want to approve a request and you miss an important pop-up entirely? Still, if PermissionsAI can reliably filter out the noise while giving users control over important decisions, it could become one of Chrome’s most welcome features yet.

You might also like

• Chrome could get a massive AI upgrade if this rumor is true
• Google’s AI Overviews are now available to help a billion people avoid reading full articles
• Google is making Gemini AI part of everything you do with your smartphone – here's how