Google I/O 2025: New Android 16, Gemini AI and everything else to expect at Tuesday's keynote
Google I/O, the search giant's annual developer conference, kicks off on Tuesday, May 20. The event is arguably the most important on the company's annual calendar, offering the opportunity for the company to share a glimpse at everything it has been working on over the past year — and contextualize its biggest priorities for the next twelve months.
The dance card for Google I/O was apparently so packed that the company spun off a dedicated Android showcase a whole week earlier. (See everything that was announced at the Android Show or go to our liveblog to get a feel for how things played out.) With that event now behind us, Google can stay focused on its most important core competency: AI.
Google's presentation will come on the heels of announcements from three big rivals in recent days. Further up the Pacific coast, Microsoft is hosting its Build developer conference, where it's already unveiled an updated Copilot AI app. Meanwhile, at the Computex show in Taiwan, NVIDIA CEO Jensen Huang highlighted a partnership with Foxconn to develop an "AI factory supercomputer" powered by 10,000 Blackwell AI chips. And Meta held its debut LlamaCon AI conference last month, but CEO Mark Zuckerberg's plans for AI dominance have reportedly since hit some snags. (Apple will share its updated AI roadmap on June 9 when its WWDC developers conference kicks off.)
If you'd like to tune in from home and follow along as Google makes its announcements, check out our article on how to watch the Google I/O 2025 keynote. We'll also be liveblogging the event, so you can just come to Engadget for the breaking news.
The presentation featured Android Ecosystem President Sameer Samat, who took over for Burke in 2024. We saw Samat and his colleagues show off years, Android hasn't had much of a spotlight at Google's annual developer conference. Thankfully, last week's Android Show breakout let Google's mobile operating system take the spotlight for at least a day.
The presentation featured Android Ecosystem President Sameer Samat, who took over for Burke in 2024. We saw Samat and his colleagues show off the new Material 3 Expressive design, and what we learned confirmed some of the features that were previously leaked, like the "Ongoing notifications" bar. Material 3 Expressive is also coming to Wear OS 6, and the company is expanding the reach of Gemini by bringing it to its smartwatch platform, Android Auto and Google TV. Android is also amping up its scam-detection features and a refined Find Hub that will see support for satellite connectivity later in the year.
Speaking of timing, Google has already confirmed the new operating system will arrive sometime before the second half of the year. Though it did not release a stable build of Android 16 today, Samat shared during the show that Android 16 (or at least part of it) is coming next month to Pixel devices. And though the company did cover some new features coming to Android XR, senior director for Android Product and UX Guemmy Kim said during the presentation that "we'll share more on Android XR at I/O next week."
It clearly seems like more is still to come, and not just for Android XR. We didn't get confirmation on the Android Authority report that Google could add a more robust photo picker, with support for cloud storage solutions. That doesn't mean it won't be in Android 16, it might just be something the company didn't get to mention in its 30-minute showcase. Plus, Google has been releasing new Android features in a quarterly cadence lately, rather than wait till an annual update window to make updates available. It's possible we see more added to Android 16 as the year progresses.
One of the best places to get an idea for what's to come in Android 16 is in its beta version, which has already been available to developers and is currently in its fourth iteration. For example, we learned in March that Android 16 will bring Auracast support, which could make it easier to listen to and switch between multiple Bluetooth devices. This could also enable people to receive Bluetooth audio on hearing aids they have paired with their phones or tablets.
Remember Google Glass? No? How about Daydream? Maybe Cardboard? After sending (at least) three XR projects to the graveyard, you would think even Google would say enough is enough. Instead, the company is preparing to release Android XR after previewing the platform at the end of last year. This time around, the company says the power of its Gemini AI models will make things different. We know Google is working with Samsung on a headset codenamed Project Moohan. Last fall, Samsung hinted that the device could arrive sometime this year.
Whether Google and Samsung demo Project Moohan at I/O, I imagine the search giant will have more to say about Android XR and the ecosystem partners it has worked to bring to its side for the initiative. This falls in line with what Kim said about more on Android XR being shared at I/O.
If Google felt the need to split off Android into its own showcase, we're likely to get more AI-related announcements at I/O than ever before. The company hasn't provided many hints about what we can expect on that front, but if I had to guess, features like AI Overviews and AI Mode are likely to get substantive updates. I suspect Google will also have something to say about Project Mariner, the web-surfing agent it demoed at I/O 2024. Either way, Google is an AI company now, and every I/O moving forward will reflect that.
Speaking of AI, Project Astra was one of the more impressive demos Google showed off at I/O 2024. The technology made the most of the latest multi-modal capabilities of Google's Gemini models to offer something we hadn't seen before from the company. It's a voice assistant with advanced image recognition features that allows it to converse about the things it sees. Google envisions Project Astra one day providing a truly useful artificial assistant.
However, after seeing an in-person demo of Astra, the Engadget crew felt the tech needed a lot more work. Given the splash Project Astra made last year, there's a good chance we could get an update on it at I/O 2025.
According to a report from The Information , Google might be planning to unveil its own take on Pinterest at I/O. That characterization is courtesy of The Information, but based on the features described in the article, Engadget team members found it more reminiscent of Cosmos instead. Cosmos is a pared-down version of Pinterest, letting people save and curate anything they see on the internet. It also allows you to share your saved pages with others.
Google's version, meanwhile, will reportedly show image results based on your queries, and you can save the pictures in different folders based on your own preferences. So say you're putting together a lookbook based on Jennie from Blackpink. You can search for her outfits and save your favorites in a folder you can title "Lewks," perhaps.
Whether this is simply built into Search or exists as a standalone product is unclear, and we'll have to wait till I/O to see whether the report was accurate and what the feature really is like.
Last year, Wear OS didn't get a mention during the company's main keynote, but Google did preview Wear OS 5 during the developer sessions that followed. The company only began rolling out Wear OS 5.1 to Pixel devices in March. This year, we've already learned at the Android Show that Wear OS 6 is coming, with Material 3 Expressive gracing its interface. Will we learn more at I/O? It's unclear, but it wouldn't be a shock if that was all the air time Wear OS gets this year.
Google has jumped the gun and already launched a standalone NotebookLM app ahead of I/O. The machine-learning note-taking app, available in desktop browsers since 2023, can summarize documents and even synthesize full-on NPR-style podcast summaries to boot.
Google has a terrible track record when it comes to preventing leaks within its internal ranks, so the likelihood the company could surprise us is low. Still, Google could announce something we don't expect. As always, your best bet is to visit Engadget on May 20 and 21. We'll have all the latest from Google then along with our liveblog and analysis.
Update, May 5 2025, 7:08PM ET: This story has been updated to include details on a leaked blog post discussing "Material 3 Expressive."
Update, May 6 2025, 5:29PM ET: This story has been updated to include details on the Android 16 beta, as well as Auracast support.
Update, May 8 2025, 3:20PM ET: This story has been updated to include details on how to watch the Android Show and the Google I/O keynote, as well as tweak the intro for freshness.
Update, May 13 2025, 3:22PM ET: This story has been updated to include all the announcements from the Android Show and a new report from The Information about a possible image search feature debuting at I/O. The intro was also edited to accurately reflect what has happened since the last time this article was updated.
Update, May 14 2025, 4:32PM ET: This story has been updated to include details about other events happening at the same time as Google I/O, including Microsoft Build 2025 and Computex 2025.
Update, May 19 2025, 5:13PM ET: Updated competing AI news from Microsoft, Meta and NVIDIA, and contextualized final rumors and reports ahead of I/O. If you buy something through a link in this article, we may earn commission.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Business Insider
an hour ago
- Business Insider
AI leaders have a new term for the fact that their models are not always so intelligent
As academics, independent developers, and the biggest tech companies in the world drive us closer to artificial general intelligence — a still hypothetical form of intelligence that matches human capabilities — they've hit some roadblocks. Many emerging models are prone to hallucinating, misinformation, and simple errors. Google CEO Sundar Pichai referred to this phase of AI as AJI, or "artificial jagged intelligence," on a recent episode of Lex Fridman's podcast. "I don't know who used it first, maybe Karpathy did," Pichai said, referring to deep learning and computer vision specialist Andrej Karpathy, who cofounded OpenAI before leaving last year. AJI is a bit of a metaphor for the trajectory of AI development — jagged, marked at once by sparks of genius and basic mistakes. In a 2024 X post titled "Jagged Intelligence," Karpathy described the term as a "word I came up with to describe the (strange, unintuitive) fact that state of the art LLMs can both perform extremely impressive tasks (e.g. solve complex math problems) while simultaneously struggle with some very dumb problems." He then posted examples of state of the art large language models failing to understand that 9.9 is bigger than 9.11, making "non-sensical decisions" in a game of tic-tac-toe, and struggling to count. The issue is that unlike humans, "where a lot of knowledge and problem-solving capabilities are all highly correlated and improve linearly all together, from birth to adulthood," the jagged edges of AI are not always clear or predictable, Karpathy said. Pichai echoed the idea. "You see what they can do and then you can trivially find they make numerical errors or counting R's in strawberry or something, which seems to trip up most models," Pichai said. "I feel like we are in the AJI phase where dramatic progress, some things don't work well, but overall, you're seeing lots of progress." In 2010, when Google DeepMind launched, its team would talk about a 20-year timeline for AGI, Pichai said. Google subsequently acquired DeepMind in 2014. Pichai thinks it'll take a little longer than that, but by 2030, "I would stress it doesn't matter what that definition is because you will have mind-blowing progress on many dimensions." By then the world will also need a clear system for labeling AI-generated content to "distinguish reality," he said. "Progress" is a vague term, but Pichai has spoken at length about the benefits we'll see from AI development. At the UN's Summit of the Future in September 2024, he outlined four specific ways that AI would advance humanity — improving access to knowledge in native languages, accelerating scientific discovery, mitigating climate disaster, and contributing to economic progress.
Yahoo
4 hours ago
- Yahoo
Scammers targeting NYC parking meters: How to protect your money
The Brief Scammers are placing fraudulent QR code stickers on NYC parking meters, leading drivers to phishing websites to steal payment information. The Department of Transportation acted swiftly by notifying users, inspecting meters, and working with Microsoft to remove the phishing site. Cybersecurity expert Robert Siciliano advises using official apps for parking payments and monitoring credit card statements, while reporting any suspicious QR codes. NEW YORK - Drivers beware! Scammers are targeting New York City parking meters by placing fraudulent QR code stickers on machines, directing drivers to third-party websites to steal payment information. What we know The Department of Transportation warns that scanning these QR codes leads to a phishing site where users are asked for payment details, which are then stolen by scammers. At least one fake sticker was found on a meter, prompting immediate action from the agency, including notifications to users, inspections of all meters, and collaboration with Microsoft to remove the phishing site. What they're saying Cybersecurity expert Robert Siciliano advises vigilance, noting that lone QR codes are often fraudulent. "A lone QR code that is plastered on is usually going to be fraud," he said. "Most municipalities and cities know at this point that there are three to five different apps that consumers might use to log in to pay for their parking, and as long as you have that app on your phone, generally you should be good." He also emphasizes monitoring credit card statements for unauthorized charges. "The problem with this particular crime is if you are not paying attention to your credit card statements in real time and pay the bill, you are responsible for that fraud," he said. What you can do The official way to pay for metered parking in the city is through the ParkNYC app or by inserting a credit card directly into the meter. Reporting any suspicious QR codes is crucial. The Source This article uses information from the New York City DOT and cybersecurity expert Robert Siciliano.
Yahoo
5 hours ago
- Yahoo
Why Smart People Make Dumb Money Decisions, According to Humphrey Yang
According to the TIAA Institute-GFLEC Personal Finance Index, about half of American adults lack financial literacy, and even more fall short when it comes to decisions regarding risk. According to financial YouTuber Humphrey Yang, being smart can put you at a greater risk of making poor choices. Read More: Find Out: In a recent YouTube video, Yang covered three biases that often trap smart people into making money decisions that leave them poor. But even if you consider yourself intelligent and financially literate, that doesn't guarantee you'll do the best things with your money. Here are the signs to watch out for if you're making dumb money decisions, and tips to avoid falling for them. Authority bias is when you believe what a person — like a CEO, celebrity or financial advisor — says because of their high influence or position. This can get you in trouble since their advice might be completely wrong or not based on the reality of your situation. Yang gave the example of quantum computing stock prices. In December 2024, a Google Willow announcement led many investors to buy these stocks, which boosted their prices. But in January 2025, Nvidia's CEO said the tech had many years to go, and stock prices fell a lot. 'The truth is that many people probably didn't do any due diligence when it came to these stocks, and they probably bought them on a speculative future after the Willow announcement, and then they sold them on a whim after a negative comment,' Yang said. To protect yourself from this bias, don't rely solely on what a single person says to do with your money. Yang said you should also forget whatever is special about that person to improve your objectivity, see what other people say differently about the topic, and trust your instincts. Discover More: If you often look only for information that aligns with your beliefs about money and brush off anything that says differently, you've fallen for confirmation bias. Besides leading to bad money moves, this bias can make you an easier person to scam, according to the Ohio Attorney General. Yang explained, 'It's especially dangerous for those that are super logical because if you're a super methodical thinker, you can actually build a logical sounding argument to defend your pre-existing opinion.' He gave an example of how this can play out with tech stocks. If you favor those stocks, you might watch for positive news reports, listen to influencers who are fans of tech, and focus on friends who profited big. You might not consider any bad earnings projections or the investors who went broke. According to Yang, asking 'why' several times helps avoid bad decisions due to confirmation bias. This lets you dig into your motivation and reasoning for making the money move. He also suggested writing down the decisions you make so you can later look back on why you did certain things and what you expected. 'This is arguably the most dangerous cognitive bias for smart people, and that's basically when people overestimate their knowledge, abilities and their predictions,' said Yang. Overconfidence bias can cause you to not consider risks since you mistakenly think you have an advantage with money over other people, and that could even be due to expertise in an unrelated area. Yang explained that this mistake played a role in various financial crises over the last few decades. Being overconfident might also lead you to not diversify your money enough and risk major losses. Yang gave examples of copying Warren Buffett's portfolio with limited investment choices or investing substantially in your own employer's stock due to familiarity. To avoid letting overconfidence damage your finances, consider that some successes might have come from pure luck rather than a wise choice you made. Yang said you should also regularly compare your predictions to reality and stick to simple investing strategies, like using index funds instead of betting on the next big individual stock. More From GOBankingRates 3 Luxury SUVs That Will Have Massive Price Drops in Summer 2025 These Cars May Seem Expensive, but They Rarely Need Repairs Clever Ways To Save Money That Actually Work in 2025 This article originally appeared on Why Smart People Make Dumb Money Decisions, According to Humphrey Yang
