Tech

Despite the upsetting news that Xbox is shutting down Hi-Fi Rush studio, Tango Gameworks, there is still some time before the lights go out for good. The developers at Tango are using that time to finish some last minute projects including working with Limited Run Games to release a physical version of the game and one final patch that’s scheduled for today.

Here’s a quick demo of Stem Splitter, a new feature added to Logic Pro for iPad and Mac, which takes a single track of mixed instruments and splits it into individual multitrack files for remixing or sampling.

This idea, using AI to separate music tracks, seemed so novel only 6 months ago when we got a new Beatles song using a similar process with John Lennon’s old tape recordings.

Personally, some rear seat riders in my Tesla Model 3 have told me I’m driving “too rough,” so I often turn down the regenerative braking settings, which has helped. Many aren’t yet used to that instant torque and the smooth, no-engine rumbling acceleration of an EV, either — but they will as personal transport continues to shift to electric.

We cut down the nearly two-hour presentation just for you, ICYMI. You can also read about everything that was announced if you prefer words. Happy Wednesday!

Does running custom shortcuts put the “Pro” in Apple Pencil Pro? I scoffed at the name when the new iPad stylus was announced last week (it doesn’t even have a built-in eraser function?) but I take it back knowing I can use it to turn off my LED candles like I’m waving a magic wand.

It’s Patch Tuesday, and Microsoft says a problem that caused VPN connections to fail after installing the April 2024 security update can be resolved by installing an update released today, May 14th.

According to the notes for KB5037771:

This update addresses a known issue that might cause your VPN connection to fail. This occurs after you install the update dated April 9, 2024, or later.

Google is distributing these little handbooks for prompting AI, which is kind of adorable? It has color-coded highlights breaking down the basic components of a prompt. There’s an early internet “How to use a search engine” vibe about it — I’m gonna hang on to this one for posterity.

Sergey posted up outside the area where Google was giving demos of Project Astra multi-modal chats. He said he thinks Sundar is doing a good job making hard decisions as CEO, said he mostly uses AI for coding tasks, and politely declined to answer a question from Bloomberg’s Shirin Ghaffary about Larry Page accusing Elon Musk of being a “speciesist.”

RetroArch, a long-running (and very good) emulator frontend, is waiting for Apple to green-light it for the App Store, according to one of its developers. PPSSPP, a well-known PSP emulator, is also in the approval queue, its developer told The Verge via email.

And that’s not all! Another multi-system emulator (Nintendo 3DS, DS, GBA) called Folium is also awaiting approval. There are yet others — see this list on Reddit.

At Google I/.O 2024 today, Google announced a multimodal version of Gemini Nano, allowing the on-device processing-powered AI model to recognize images, sounds, and spoken language in addition to text.

Those multimodal capabilities are also coming to the Android accessibility feature TalkBack, using AI to fill in missing information about unlabeled images, without requiring a connection to the internet.

I’m at mia@theverge.com. Thanks!

Obviously, someone noticed our video that clipped every single AI mention at I/O 2023 last year. Sundar Pichai closed the 2024 keynote by showing how AI can save us some work by using it to keep track. At the time, it was up to 121 AI mentions.

...by the time they were finished, it was probably more like 124.

Head over to Google’s Vertex AI Studio site and click “Try it in console” to goof around with some of the AI tools Google talked about at I/O today. The site is meant for developers who want to test the company’s models out while deciding what works best for their software, but anyone can play with it.

The scam detection feature Google just announced requires Android users to opt in, and Google claims it’s on-device only, but it’s still essentially listening to your every conversation to look for fraudulent-sounding language.

Are we really ready to swap scamming concerns with privacy-related ones?

Google teased the next generation of its small language model Gemma, including a larger version with 27 billion parameters (or how much a model understands).

The company also announced PaliGemma, an open-source model in the Gemma family for labeling photos and adding captions to images.