Post

wired.comHow to Watch Google I/O 2025The annual developer conference kicks off with a keynote address on Tuesday. Look out for updates on Android, Gemini, Search, and Google’s latest forays into XR.

The keynote address will kick off at 10 am PT, and it’ll be livestreamed. There will also be a developer-focused keynote starting at 1:30 pm PT. Here’s how to watch both: www.wired.com/story/how-to...

What are we expecting to see at I/O this year? AI. So much AI. www.wired.com/story/how-to...

wired.com‘You Can’t Lick a Badger Twice’: Google Failures Highlight a Fundamental AI FlawGoogle’s AI Overviews feature credible-sounding explanations for completely made-up idioms.

It was just one year ago that we first heard about AI Overviews, and since then the feature has fundamentally changed the Google search experience in ways both good and… weird. Really weird.

wired.comAndroid 16 Is Getting a Facelift, and Gemini Is Rolling Onto More Google PlatformsGoogle teased all the new features coming to its mobile operating system ahead of next week’s I/O conference.

We’re also sure to get updates about Gemini and how the latest model is being used on Android phones. (We already know a little something about this since Google teased a bunch of new features last week.)

Google already shared the design changes in Android 16, which will include a new design language called Material 3 Expressive. Crucially, Google claims that “users are more likely to switch to products that use M3 Expressive components and techniques.” www.wired.com/story/androi...

20/5/25, 4:23 pm0 reposts16 likes

The Dash

@thedash.bsky.social

They need a dedicated widget for the "Ok, got it" feature they always insist you use. Another one for "oops, something went wrong" might be good too. #ux

The Dash

@thedash.bsky.social

Some AI agent that can try again for you perhaps?

wired.comGoogle Is Using On-Device AI to Spot Scam Texts and Investment FraudAndroid’s “Scam Detection” protection in Google Messages will now be able to flag even more types of digital fraud.

We also already know about the new scam-blocking capabilities of Android phones, which will now be able to flag more types of digital fraud.

L'esprit de l'escalier

@mitchellcolbert.bsky.social

Oh you mean the scam text blocker that was just upgraded into my android phone will actually do that? Right now it is doing fuck and all.

In a demo of Project Astra built into Gemini on phones, Google did a clever little inversion in which the human using it was hallucinating fake shit and the AI model was correcting the human. AI will ruin reality before it ultimately proposes a solution to fix it, writes @laurengoode.bsky.social.

wired.com‘You Can’t Lick a Badger Twice’: Google Failures Highlight a Fundamental AI FlawGoogle’s AI Overviews feature credible-sounding explanations for completely made-up idioms.

Google's CEO Sundar Pichai just said their AI doesn't make mistakes all the time... Interesting, considering:

noYOUhush

@noyouhush.bsky.social

first & only time used AI it was wrong 👿🤡👿🤡👿🤡👿🤡 inferior data = inferior result #FactsMatter

wired.comWith AI Mode, Google Search Is About to Get Even ChattierGoogle's new chatbot-style AI Mode search experience, previously an experiment, is launching for US users. Publishers and marketers will have to adjust their search strategies once again.

Big update: Google is rolling out its AI Mode chatbot-style search experience to everyone in the US starting today. 🫠 It’s the next step in the evolution of AI Overviews, which were introduced at I/O last year and haven’t exactly been super popular.

SanguinePar

@sanguinepar.bsky.social