Allison Johnson | The Verge

Here comes new Siri again

2026-06-05T14:40:17-04:00

Our first glimpse of the new AI Siri came all the way back at WWDC 2024.

Apple has been on its back foot, AI-wise, for the past few years. But in a strange way, playing from behind might not be such a bad move.

At WWDC on Monday, Apple appears to be getting ready to reintroduce us to the new Siri. Again. As a reminder, we met the new Siri in 2024 when Apple “launched” Apple Intelligence. Siri came with a new glowing border, different voice options, and the ability to punt questions to ChatGPT. The whole “Intelligence” bit of the Siri redesign was coming soon, Apple promised. It didn’t. In fact, its promotion around Apple Intelligence was so misleading that the company is settling a class-action lawsuit and has to pay iPhone owners for the features it never shipped. The funny thing is, by fumbling the ball so badly, Apple might have just fallen backward into an advantageous position.

Let’s be clear; if such a thing as a race to an AI assistant exists, Apple is losing badly. Gemini is already doing things like ordering Ubers and DoorDashing teriyaki. It can look at your calendar and figure out when you should leave for the airport. Gemini won the race, fair and square.

Gemini is already doing things like ordering Ubers and DoorDashing teriyaki

But there’s also a growing distrust of AI, particularly from young people, and the better Gemini gets, the creepier it is. It has to be if it’s going to deliver on the promise of a truly helpful assistant. But wanting your AI assistant to anticipate your next move and actually watching it happen? Those are very different things. I willingly gave Gemini permission to access my Google Photos and Gmail, but it always makes my skin crawl hearing Gemini say my son’s name out loud. I test out a lot of this stuff as it becomes available — hazard of the job — but the public reaction when these kinds of features start trickling down to the mainstream will be very telling.

New New Siri will be built on top of Gemini in some fashion. Apple is no doubt paying handsomely for the privilege, but there’s a potential upside to being one step removed in this way. You know what company doesn’t have its name attached to a big, unpopular data center project? Apple. Google isn’t winning friends and influencing people by rushing to start massive construction projects in backyards across the country. Apple gets to keep its hands clean, even if its payments to Google are presumably being funneled toward the great data center buildout.

Then there’s the Copilot of it all; the AI-buttons-everywhere factor. Siri’s attempts to summarize messages are amusing and often annoying, but at least Siri isn’t all up in every one of my work documents begging to summarize it for me. On the other hand, you can’t open a Google app without coming face-to-face with a Gemini sparkle these days, and it risks getting real old, real fast.

Don’t get me wrong; I think Apple would love to put Siri to work writing my emails, perfecting my photos into “memories,” and talking me through the next steps to rehabilitate the dying plants in my yard. It’s just that Siri can’t really do any of that yet. When we meet this new Gemini-enhanced Siri, it’ll be telling to see where and how aggressively it surfaces. According to Bloomberg’s reporting, it sure sounds like we’re going to see it in a lot of places: the Dynamic Island, Photos, maybe even its own dedicated Siri app for the first time. That’s a very different Siri from the timer-setting voice assistant we currently know, mostly hiding behind the scenes.

Remember this tagline from the first time Apple tried to launch AI Siri? Two years ago? Yeah, me neither." data-portal-copyright="" />

I suspect Apple is also going to play up the thing it already loves playing up: privacy. You can bet we’ll hear more about Private Cloud Compute, which supposedly keeps your data as secure as if it had never left your device. The updated Siri may also come with the option to automatically delete chats after a certain period of time, rather than holding onto that data by default. Promising a more private, secure AI experience might appeal to people who are squeamish about handing over even more personal info to Google. But it doesn’t do much for someone who’s just sick of having AI in their face all day in every piece of software.

An advantage, especially the kind you stumble on, can disappear as quickly as it arrived.

Apple could easily cast its slow AI rollout as the more responsible move. Google execs used to constantly talk about being “bold and responsible” with AI, but lately they’re too busy firing off new Gemini features and basking in the foothills of the singularity to dwell on that much. Passing off the delays as taking the time to do things right isn’t a bad bet, but the time for false starts is over. Siri’s going to have to pull it off for real this time; when a second chance like this comes around, you can’t count on it coming back.

This chunky little tablet got my kid to clean up his toys

2026-06-04T16:46:07-04:00

Just a cute little guy.

Never underestimate the power that a cheap tablet holds over a kid under six.

The Skylight Buddy is a device with one job: to be a cute little guy that helps your kid track routines and chores. It’s $139.99, plus an optional subscription. And to my surprise, even though it offers a pretty limited set of features without the $39-per-year “Plus” features, it actually worked.

Skylight recommends the Buddy for kids aged four up to 10. An adult has to set it up, naturally, which you do inside of the Skylight app. From there, you make a profile for your kid and assign it to the Buddy, which is strictly one-kid-per-device. That might be a non-starter for bigger families, but in my household of one kiddo it worked fine. In the app you can set up recurring or one-off tasks and group them into routines for the morning, afternoon, and evening. They appear on the Buddy’s screen as big cards with emoji labels, so it’s a viable option for a kid like mine who isn’t reading yet.

Skylight Buddy

Skylight’s kid-centric device guides little ones through daily routines with an easy interface and adorable design. Advanced features require a subscription, but the core functionality is free and surprisingly effective.

Score: 6

Pros	Cons
Basic functions available without subscription Well-considered, kid-centric design Adorable	A few appealing features require a subscription Ability to change routines is somewhat limited $140 feels steep for basic features

Where to Buy:

$139.99 at Skylight

The Buddy has a few other features available without a subscription, like the ability to use it as a night light and set a wake-up alarm. But a $39-per-year “Buddy Plus” subscription is required for extra features like reminders, the ability for kids to earn rewards for completing tasks, or setting visual timers for individual tasks. Buddy Plus features are included “for a limited time” if you already have the Calendar Plus subscription associated with its other products. We don’t, and we already have a white noise machine and a night / wake-up light, so that reduced the Buddy more or less to a daily checklist on a gussied-up Android tablet.

“Surely my child won’t care about checking things off a screen without a built-in reward system,” I thought. I figured I’d try the basic feature set and upgrade to Plus when that proved ineffective. As it turns out, the basic features were all I needed, so I never ended up testing the extras.

Kids who aren’t reading yet can still mark off the right tasks thanks to the big emoji." data-portal-copyright="" />

My kid loves checking things off “his screen.” There’s not much to it; I set up basic morning and evening routines with tasks like “eat breakfast” and “brush teeth.” He eagerly taps on the corresponding card when he’s done with each step, and when every task is finished at the end of the day there’s an onscreen celebration with a shower of emoji. The emoji changes, so it’s become kind of an event to see which emoji it’ll be each day: “MOM IT’S WAFFLES,” etc.

That’s basically it! I couldn’t believe the con I was running. Like, really? You’ll clean up your toys when a tablet tells you to, but not when I do? Without dangling the carrots of extra screentime or an ice cream outing? Once again: Do not underestimate the power of a screen.

I couldn’t believe the con I was running

The Buddy is adorable, which definitely helps. My review unit came with a green silicone case, which is an extra $20. It helps with the cute factor. But I think the Buddy is most appealing because it’s a screen that’s clearly for him and nobody else. As I was setting it up on the counter in our kitchen, I realized it wasn’t angled upward like an Echo Show, which would make it more comfortable for a taller person to use. It faces straight ahead, basically at eye level for a four-year-old. Clever.

I wish the app was as user-friendly. Setting up tasks as part of a routine is easy enough, but once you have a task set for a certain day, you can’t move it to another day. We generally do a bath every other night, but you know, some days call for an off-cycle bath. I’d like to be able to move the task forward a day and have the recurrences line up after that; instead I need to make it a daily part of the routine and remember to mark it as “skip” on non-bath nights.

A good chunk of features are locked behind a “Buddy Plus” subscription." data-portal-copyright="" /> I’d like a little more flexibility in creating routines." data-portal-copyright="" />

I’m also unable to reorder the individual tasks within a routine — you can do this with a Skylight Calendar, which I don’t have. Skylight’s VP of product, Anubhav Sarkar, tells me that this functionality will come to the app this month, but until then I’m stuck with the tasks lining up in the order that I created them. It’s a real pain to add an extra one-off step to the day’s routines.

There’s also no way to display a calendar view on the Buddy, even though you can add calendar events in the app. The Buddy seems more intended as a complement to other Skylight calendars in the house, which, again, we don’t have. But if you do, this probably won’t be a concern. Personally, I was hoping I could outsource delivering the bad news that there’s school today to a screen, but no luck.

The buttons are appropriately chunky and kid-friendly." data-portal-copyright="" /> The Skylight Buddy can play white noise and a wake-up alarm." data-portal-copyright="" />

Look, it sounds silly using a $140 screen to do the job of a $5 sticker chart, or whatever low-cost crafty visual aid you might be able to cook up. But there’s something pretty powerful about a kid being able to use their own device, and the hands-off-ness it affords a parent. For a while, we tried to implement a “pizza chart” system where our son earned little cardboard slices of pizza for each task he completed in his daily routines. At the end of the day, the pizza slices could be cashed in for screentime. It was pretty labor-intensive on our part to constantly hound him about the pizza he was or wasn’t earning, and only lasted a couple of weeks. Maybe it’s just that my kid is a little bigger now, but even the free version of the Buddy that we’ve been using has been more successful than our attempts at turning fake pizza into episodes of Paw Patrol.

I’d recommend the Buddy a little more freely if the app was just a bit more flexible. Being able to move tasks around without having to reconstruct a routine every time would go a long way. Even so, it’s been a useful addition in our house. We still hit snags in the morning and evening routines that no screen can solve, but I think it has genuinely helped give my kid a better sense of structure. It seems like a pretty easy addition if you have young kids and already orchestrate your household’s activities with the help of Skylight calendars. And lord knows I’ve spent more money on short-lived gadgets to try and smooth out parenting pain points. If you suspect that it’s going to help relieve some stress in your house, then you could definitely do worse than this cute $140 checklist.

Photography by Allison Johnson / The Verge

How one founder’s bet on ‘the old school web’ is paying off

2026-05-29T10:32:11-04:00

A good time with old maps. | Image: Past Maps

Craig Campbell walked away from the river of investor money flowing into AI to create, of all things, a website.

Sure, Campbell probably could have started an AI company. He’s a former engineer at Meta and an experienced tech founder who in 2022 sold his last venture — an e-commerce tool for businesses that use Shopify — right as the AI boom was booming. “I had my prior VC investors breathing down my neck, going ‘start something else. We’ll write you a blank check.’” He had other ideas.

People generally aren’t rushing to get into the website business, what with the Google Zero event horizon approaching. Campbell was undeterred and has grown his service — Past Maps — into a sustainable business. And he’s managed it in an increasingly unlikely way: via organic search.

Past Maps is true to its name. The site lets you view historical maps of a particular region with a modern-day map overlaid. You can adjust the opacity to fade between the two views. The maps come from publicly available sources like the US Geological Survey, but the tools to allow people to explore them in this way were developed by Campbell. He built them to help inform his metal detection hobby — by pinpointing the modern-day locations of old structures and trails, he’d identify new places to go looking for artifacts. He started sharing his map tooling on Reddit with other metal detection enthusiasts and found that other people wanted to get their hands on what he’d created. With that, his newest tech venture was charted.

You don’t have to be looking for literal gold to enjoy Past Maps. For someone who’s just curious about what’s around them, it’s its own kind of treasure trove. I’ve used it to help grasp things like the shape of the Duwamish River before it was straightened out to help ships move through the waterway. Campbell’s customers use it for a wide range of reasons — from genealogy research to a daily user who maps old oil wells. It’s a research tool, but it’s also just plain fun.

Watch the Duwamish River in the lower portion of the frame go from squiggly to straight and back again." data-portal-copyright="" />

The growth trajectory has been steady. Campbell says traffic has grown from an average of 20,000 active users a month to now over 300,000 a month in year three. The income is good enough to sustain Campbell and his wife, who also helps with the business. But he can’t help but think about what the money might have been like if he had taken those VC investments to work on AI. “I’m making the same as when I was like, an E4 at Facebook, which is like a mid-level engineer.”

“This is how the web is supposed to work. This is actually the old school web.”

Past Maps’ biggest source of traffic is Google Search results. Campbell found early on that Past Maps was rising through the ranks of search when people went looking for historical information about locations of interest to them — a church their grandmother attended, or abandoned mine sites in a particular county.

By tagging his maps and webpages in a way that Google understands, he saw a cycle start to pick up. “As I started exploding out this data and making it finally available to Google and giving it a place on the web, traffic just started to build.”

“This is how the web is supposed to work. This is actually the old school web,” he says. “It is alive and well, but only in these really, really small niches.”

An old school web publisher of 10 or 15 years ago likely would have relied on display advertising for the bulk of their revenue. You can dabble with a free Past Maps account, but going deeper requires a $9 weekly pass or $52 per year for an annual subscription. Subscriptions protect Campbell from the whims of fluctuating marketing budgets and an ad tech industry largely controlled by Google — which the DOJ ruled as an illegal monopoly in 2025.

While AI may be eating the open web alive, Campbell has fully embraced AI tools to help run the business. Campbell says that he used to spend one or two hours a day handling every service request himself, writing lengthy emails rather than sending a form response and an FAQ. Now, he lets a local agent model on his desktop to handle the front-line triage. Its prescheduled task runs once an hour — assuming his laptop is powered on — and has access to his Gmail. It weeds out spam and marketing messages, identifies the things that need his attention, and drafts a response. He says this has cut down his customer service time to about 10 minutes a day.

“I do sometimes have angry customers,” Campbell says. “If they ask me for a refund, it cues up the refund and subscription cancellation request with Stripe. It does the whole thing, then it pings me.” At that point, he looks over the request, approves or denies it, and checks the message before hitting send.

Campbell is also using AI to help build an OCR tool — Optical Character Recognition — that will work with old maps. “Cartographers are assholes,” Campbell jokes. Historical maps are a particular challenge for existing OCR systems. Labels will curve along features like rivers, use inconsistent spacing, and are sometimes crowded in on top of each other. Campbell found that off-the-shelf tools would fail to parse these maps. He found more success with modern LLMs using reasoning, but it’s not a simple matter of prompting an agent to “OCR these maps,” he says.

“You have to still bring that human spark into the mix.”

Instead, he’s found success in combining a human sensibility for experimentation with the LLM’s capabilities, rather than relying solely on the tool. “It still doesn’t bring like that human-level reasoning spark, and creativity, and being able to stitch together decades of using tools like this,” he says. “You have to still bring that human spark into the mix.”

Campbell may have walked away from a supposed AI gold rush, but in doing so it seems he created a recipe for a successful business online in the age of Claude Code and AI summaries. When you start with something you’re passionate about, make something that’s useful, and share it with other people like you, that turns out to be a pretty good foundation. Campbell’s day-to-day looks awfully different from the way you’d build and run a website 10 years ago, but the things that have made the business a success today are thoroughly human.

The new Razr Ultra isn’t your average phone — for better and worse

2026-05-27T14:24:55-04:00

A fine-looking phone.

I had one ask for friends, colleagues, the lady checking me in for a meeting at a large software company’s headquarters, and everyone else who stopped to admire the phone I’ve been carrying around.

“Pet it.”

The Razr Ultra is not your average phone. I got the orient blue color option to test, which has a soft, woven back panel made of Alcantara fabric — which you’re more likely to find on the seats of a fancy car. I can’t stop petting it. I’m worried about how it’ll look after spending years in and out of dusty tote bags and my kid’s grubby hands, but after a couple of weeks of testing it hasn’t picked up any gunk or dirt that I haven’t been able to brush off.

Then there’s the above-average price. The Razr Ultra costs $1,499, which buys a well-equipped flip phone. I still think calling this phone an “Ultra” is a little too strong. You won’t get all of the trappings of a regular top-tier phone, like a telephoto camera, embedded magnets for Qi2 charging (just plain wireless charging), and full dust resistance. You pay for the privilege of the hinge and a seriously nice-looking (and -feeling) phone.

Motorola Razr Ultra (2026)

The 2026 Razr Ultra offers excellent battery life despite the constraints of the flip form factor. But an inconsistent camera and steep price make it hard to recommend.

Score: 6

Pros	Cons
Gorgeous design Excellent battery for a flip phone Cover screen is handy and delightful to use	Expensive Too much bloatware Photo processing is a little wonky

Where to Buy:

The pettable back panel isn’t even my favorite upgraded feature on the 2026 Ultra, which surprised me. It’s the battery. It has a 5,000mAh capacity, up from 4,700mAh on last year’s model. It manages to accommodate a battery with a capacity usually found on the biggest of big slab phones, even though it has to make room for a whole-ass hinge. Motorola can pull this off because it’s using silicon-carbon batteries, which provide higher capacity in the same amount of space a traditional lithium-ion battery takes up.

In practice, this meant I got comfortable committing battery crimes that I’d normally avoid with a flip phone. Using it as a hotspot while working outdoors on an 80-degree day? Forgoing a nightly recharge simply because the charger is across the room? Setting it up in tent mode and using the front screen as a Pomodoro timer? I did it all with the Razr Ultra. I never hit low power mode, and most days I didn’t even get down to 50 percent. This is very good performance from a phone with two physically small-ish batteries, and that’s a big win for the Razr.

Don’t worry, I didn’t use any of these suggested replies." data-portal-copyright="" />

Every time I revisit the flip form factor, I remember how great it is. You know what rules? Answering texts on the cover screen, typing single-handed while holding a coffee in the other hand. I used that front screen to catch up on Slack notifications, present my boarding pass to a gate agent, and entertain my four-year-old for a few minutes with a gyroscope-guided marble maze game. Doing these things without opening up the whole phone and inviting in that chaos feels like a cheat code. Motorola’s cover screen software makes it easy to access any app you want, and it handily beats Samsung’s outer screen UI.

I wish I had better news to report elsewhere. I had high hopes for the main camera, which uses a new physically larger 50-megapixel sensor than the previous gen. But Motorola is still making some weird photo processing choices. Colors are very saturated, bordering on artificial, and on an overcast day my photos turn out looking overly bright and flat. Someone who’s not too picky about these things probably won’t mind the look. I am picky, and I was hoping Motorola would have steered its processing back toward the real world with this new sensor. Sadly, that does not seem to be the case.

Then there’s the bloatware. This is a recent Motorola tendency that used to be worse on its budget phones, but has since leveled out to a bearable, if still gross, amount of preinstalled stuff on its whole lineup. You might think that spending $1,500 on a state-of-the-art foldable phone would spare you from looking at a third-party “Newsfeed” in the app drawer filled with “content that’s tailored to your interests,” aka targeted advertising, but you’d be wrong!

You can uninstall and disable a lot of the bloatware, and I highly recommend that you do. In fact, Motorola confirmed a recent incident in which a bad actor took advantage of one of these third-party apps to route some users to an affiliate link when when opening the Amazon app. Motorola’s executive director of product management Allison Yi tells The Verge that the company “promptly corrected the routing configuration.” I didn’t see this behavior on my Razr review unit, and it doesn’t appear that anybody’s personal data has been compromised. But it’s very concerning, and seems entirely avoidable by not loading up the phone with third-party software to begin with.

Long live tent mode." data-portal-copyright="" />

Despite its good looks, the Razr Ultra isn’t the best option for someone seeking the thrills of a modern flip phone. I’d feel much more comfortable recommending it for like, $1,200, or if Moto had figured out how to seal up that hinge for full dust resistance. Or maybe if it cost as much as it does now, but didn’t come with a bunch of third-party apps that may be susceptible to bad actor shenanigans. It’s not your average phone — that much is obvious just looking at it. But maybe being a little more average in some ways isn’t such a bad thing.

Photography by Allison Johnson / The Verge

Update, May 27th: Updated to include a statement from Motorola regarding and incident in which a bad actor took advantage of pre-loaded software to attach an affiliate link when some Razr users opened the Amazon app.

Agree to continue: Motorola Razr Ultra (2026)

Every smart device now requires you to agree to a series of terms and conditions before you can use it — contracts that no one actually reads. It’s impossible for us to read and analyze every single one of these agreements. But we started counting exactly how many times you have to hit “agree” to use devices when we review them since these are agreements most people don’t read and definitely can’t negotiate.

To actually use the 2026 Motorola Razr Ultra, you must accept the following:

Motorola’s Privacy and Software Updates

But you also get to decide how Motorola’s support works on your phone:

Help improve Motorola products (optional)
Enhanced device support (optional)
Smart updates (optional)

After entering your Google account, you are asked to:

Add a phone number to your Google account (optional)

And you must agree to the following from Google:

You’ll also need to agree to the following on Google Services:

Install updates and apps: “You agree this device may also automatically download and install updates and apps from Google, your carrier, and your device’s manufacturer, possibly using cellular data. Some of these apps may offer in-app purchases.”
Use basic device backup (optional)
Use location (optional)
Allow scanning (optional)
Send usage and diagnostic data (optional)

Google Assistant:

You can set up Google Assistant (optional)
Activate Voice Match for Hey Google (optional)
Access Assistant without unlocking your device (optional)

To use Moto AI:

Motorola AI Terms and Conditions (optional)

Lastly, you have the option to join Motorola’s user community:

Give permission to Motorola to send push notifications about its services and benefits (optional)
Provide your email to Motorola (optional)

In total, you have to accept five main agreements and can bypass 14 when setting up the Motorola Razr Ultra.

Google’s new anything-to-anything AI model is wild

2026-05-22T19:49:26-04:00

Just a stuffed deer having the time of his life. | Image: Gemini / The Verge

Last year I deepfaked my kid’s stuffed animal to make it look like his plush deer was on vacation.

It was an experiment to see if I could re-create the events depicted in a Gemini ad Google was running, and I never showed the videos of Buddy the deer on his adventures to my four-year-old. But it was a revealing exercise that made me think a lot about the difference between some harmless fun with generative AI and full-on slop. Maybe that Venn diagram is a perfect circle! Maybe not. But what I know for sure is that the tools to make realistic videos are surprisingly good, requiring surprisingly little effort and know-how. And that trend is continuing hot into Gemini’s Omni era.

Omni is a new family of generative models that will allegedly one day be able to turn any kind of input — photo, video, text — into anything else. But for starters, it’s just creating video. Omni Flash is the first of these models Google has released, now available in the company’s AI video generation and editing platform, Flow. You can still use the previous model, Veo, if you want, but Omni improves on Veo in a few ways.

With Omni, you can upload a video and use that along with a text prompt as the starting point for your AI-generated creation. Google also claims Omni incorporates more real-world knowledge when producing videos and can do a better job of keeping characters consistent throughout a video as a result. There was only one way to really know if those claims are true: I brought back AI Buddy to pack his little AI-generated bags for another adventure.

The results are such a mixed bag they’re baffling. Some were very good — much more consistent and true to my prompt than when I was testing out Veo five months ago. But even the best clips Omni cooked up for me still have certain AI jump scares, like when Buddy suddenly switches orientation while he’s skydiving.

For another video, I gave Omni some artistic freedom. “Create a montage of Buddy packing for a vacation and embarking on a cruise ship for a tropical vacation. The mood is cute and playful. Buddy packs something funny in his suitcase that comes into play later in the clip.” It had Buddy pack a jar of honey; later in the clip he reaches for it as if it’s a bottle of sunscreen. “Uh oh,” the character says as he squirts honey onto his hoof.

Honestly, not a bad bit. Except that the bottle of honey constantly changes throughout the video, from a jar, to a clear squirt bottle filled with water, then back to a squeeze bottle filled with honey. And I can’t even begin to describe how the model came up with the final frame of the video — almost as if it just barfed up a bunch of elements of the sequence it just made.

You can use text-based prompts to suggest edits to your videos, and I’ll give Google credit: This works better with Omni than it did when I tested Veo 3. But the results were bad with Veo — so bad that I found it way easier to just prompt a new video from scratch every time I wanted something changed. Omni will actually take your edits on board, but the results don’t always hit.

I had it emphasize Buddy’s facial reactions in his vacation clips, and the results just wound up looking strange. It would also give Buddy antlers from time to time, which he does not have. Buddy is a baby, thank you very much. When I prompted it to remove the antlers that appeared in one scene, it obliged — and then added antlers in all the other ones.

The thing is, none of this is free. Generating videos costs credits, varying from 15 to 40 credits based on the length of the scene and the “ingredients” you start with. One round of edits costs 40 credits. I have the $20-per-month AI Pro plan that comes with 1,000 credits each month. After around 20 clips generated with a few edits on some, I’m down to 145. If you have specific ideas about the video you want Omni to generate, you might be looking at a lot of costly back-and-forth with the model to get a video that’s close to your vision.

I can genuinely say I wasn’t prepared for what I saw

One of Omni’s purported strengths is adding AI-generated stuff to real videos, so I gave Buddy a break and deepfaked myself. Starting with a selfie video with a neutral expression, I prompted Omni to generate videos of me eating a plate of spaghetti, sitting in an airplane seat, and standing in front of the Eiffel Tower taking a bite out of a baguette. And I can genuinely say I wasn’t prepared for what I saw.

There are AI tells in my deepfake videos. The clink of the fork hitting the bowl of pasta is a little too manufactured. There’s a woman in the background of the airplane video who shows up twice. But aside from those little glitches and a vaguely uncanny sense about them, they’re convincing as hell.

I showed my husband the pasta clip; he knew I was testing an AI video tool but I didn’t tell him what in the scene had been generated by AI. Without knowing what was AI-generated about it, he bought that I was sitting in front of a camera eating pasta, and said that his only clue something was up was that the bowl looked unfamiliar. The pasta-eating itself looked real enough to convince my husband. A man who has looked at me in real life basically every single day for the last decade.

My other deepfakes are varying levels of “good enough to fool people on social media.” A couple of the Eiffel Tower clips look slightly cartoonish, but one of them is convincing enough that you might need to rewatch it a few times to clock that it’s AI. I know it’s not me when the AI me turns her head and reveals her hair pulled back in a ponytail. But I’m not sure anyone else would know the difference, and that makes me feel weird.

We’re definitely deep in the uncanny valley

I’m a little exhausted by it all, to be honest. I was shocked when I tested Veo 3 at the realism it could produce. I’ve been shocked at how easy it is to make fake people in fake photos again and again over the past few years. I should probably be shocked by Omni too, and I guess I am, but the edge has worn off.

It’s still not quite as easy to make an AI-generated cinematic masterpiece as Google would like you to believe. But Omni does improve on Veo in some recognizable ways. If you have a Google account and a credit card, then you can take a video of yourself sitting at home and make it look like you’re on a flight to Maui with a trivial amount of effort. I don’t think we’re at the “foothills of the singularity” exactly, but we’re definitely deep in the uncanny valley.

All images and videos in this story were generated by Google Gemini.

Vibe coding is coming to your phone

2026-05-20T13:37:21-04:00

Coming to your homescreen soon: your own app. | Photo: Allison Johnson / The Verge

“There’s an app for that” was the promise of the App Store from the very beginning. The app that will get your phone to do the thing you want it to? It’s just a few taps away. The tagline wasn’t strictly true — I’m still waiting for that one perfect grocery list app. Still, apps shaped the modern smartphone into what it is today. We spend all day, every day inside of apps — scrolling, listening, and tapping until we find what we want. But your next favorite app might just be one that you made yourself.

If you weren’t familiar with the concept of “vibe coding” at the beginning of 2026, you probably are now. As AI coding tools have become better and more accessible, more and more non-developers are finding success creating apps that fulfill a niche need. Vibe coders are mostly working with desktop software, but signals from this Google I/O and beyond indicate that mobile will be the next frontier.

Prompt your way to a new, native Android app. | Image: Google" data-portal-copyright="Image: Google" />

For starters, Google is making it easier to just straight up vibe-code a whole Android app. At I/O the company announced an update to its AI Studio vibe-coding tool, allowing you to create a native Android app and export it to a phone in a matter of minutes. The feature is limited to “personal utility” apps to start with, and the rules for putting an app on the Play Store remain the same. But if you’re the kind of person looking for a particular feature from a habit tracking app that none of them seem to offer, you might just be able to build it yourself.

If a whole app feels too ambitious, then maybe a widget is more your speed. At last week’s Android Show, Google announced an upcoming feature to create your own widgets with a prompt — Google’s examples include widgets that highlight certain weather metrics or suggest new recipes to try.

BYO widget.

These tools draw on Gemini’s knowledge base, so the possibilities are pretty wide open. Naturally it all depends on this feature actually, you know, working. But the idea of putting the very specific information I want where I want it on my phone is awfully compelling. Not to make too much out of some widgets, but stuff like this has kind of been the whole premise of personal computing for the past couple of decades. If it works as advertised — again, emphasis on if — it kind of unlocks a whole new level of personalization for your phone.

Google calls the AI-generated widgets a first step toward something called a “generative UI,” where your phone creates an interface and apps on the fly based on what you need in the moment. Sounds great in theory! But it also sounds like it could get messy fast. Android president Sameer Samat acknowledges that there’s a pretty obvious way to take the concept too far. “While I don’t think we want to wake up every morning and have our devices have different UI, I do think there’s a level of personalization and customization to the user that could be delightful,” he tells me.

It seems like Apple might be taking steps toward a more personal iPhone, too. Bloomberg’s Mark Gurman reports that the company is working on a way to create shortcuts based on prompts. Shortcuts are automations you can program within the dedicated Shortcuts app, either by putting them together from preassembled bits or figuring it out on your own. They seem simple in theory but get complicated fast, which has deterred me from ever seriously getting into shortcuts. But the prospect of prompting my way into a shortcut that opens the transit app when I get to the bus stop, or sets a particular focus mode when I connect to my home Wi-Fi, is pretty appealing.

I’ve heard a lot of promises over the past few years from tech company execs about how AI will fundamentally change how we interact with mobile devices. So far, we have an upgraded voice assistant in Gemini, a Siri that will go ask ChatGPT for you, and… basically the same phones we’ve been using for the past decade. Being able to prompt an app, widget, or automation into existence isn’t exactly a platform shift, but it could actually help our phones become a little more personal.

Google I/O 2026 live blog: On the ground at Google’s keynote

2026-05-19T12:38:02-04:00

We’re back at the Shoreline Amphitheatre in Mountain View, California, for this year’s edition of Google I/O. These days, Silicon Valley is buzzing about the future of AI search, agents, vibe coding, and e-commerce, so you can bet we’re expecting to hear tons of news on these fronts. And who knows, we might get a peek at some smart glasses demos and concept projects, too.

That said, we aren’t expecting much hardware. After all, Google jumped the gun last week, announcing the Googlebook and Fitbit Air. We also already heard much of what’s in store for Android during the Android Show.

The AI landscape has changed quite a bit over the past few years, so the pressure is on for Google to distinguish Gemini from rivals like OpenAI’s ChatGPT and Anthropic’s Claude. If you want the latest on what Google has to say, follow along below. We’ll be in the audience delivering beat-by-beat updates on the show. The keynote starts at 10AM PT / 1PM ET.

',c(t.firstChild,i)))})(window);

Gemini is in danger of going full Copilot

2026-05-19T11:39:37-04:00

I actually use the Gemini app quite a bit on my phone, but let’s not get carried away.

Gemini has a creep problem.

A few years ago, that little sparkle icon started showing up in all of our Google apps. Gemini in your inbox! Gemini in your Google Drive! It was slow at first, and easy enough to tune out, but something has changed in the past few months. Gemini is creeping. It’s showing up in all kinds of places at a relentless pace, and personally, it’s starting to really cheese me off.

The AI-everywhere fatigue is familiar to anyone who has ever used Windows 11. Microsoft went absolutely bananas putting Copilot shortcuts onto every surface it could find, to the extreme irritation of many users. Likewise, we will doubtlessly hear about all kinds of new Gemini features at this week’s Google I/O conference, and I’m praying that Google has learned from Microsoft’s mistakes as it unleashes them on our Workspace apps. Nobody likes a creep.

I’m actually kind of a Gemini enjoyer, too. I used it to vibe-code an app to figure out which chores I have time for in a given day. I chat with Gemini on every Android phone I test, and I’ve started downloading the app on iPhones, too. That might put me in, like, the top 10 percent of Gemini users who don’t work at Google. I’ve even come around to the AI overviews Google sticks on top of every search result these days. Sure, there were the early glue-on-your-pizza days. And they’re probably contributing to the death of the open web. But lately I’m finding them reliable enough when the stakes are low. I’ll Google how often to water my lavender plants, or how long to bake potato wedges at 400 degrees; so far AI overviews haven’t killed my lavender or undercooked my potatoes.

But everyone has their limit, and I think the newest Gemini intrusion into Google Docs is when I reached mine. It’s a persistent sparkle icon at the bottom of the window, and if you make the mistake of mousing over it, you’ll get a full-on toolbar with suggested prompts to get Gemini to write for you. Blogging is my craft, thank you very much, so I shut that shit right down. Now, even the Gemini icons that I’d been able to tune out before are starting to bother me. I guess at some point I gave Chrome permission to put a Gemini shortcut in the menu bar at the top of my MacBook homescreen, because there’s a little sparkle up there, staring at me all the time. When did that happen? Was I tricked? It’s all a bit Haley Joel Osment in The Sixth Sense. They’re everywhere.

If 2025’s I/O keynote is any indication, we’re going to hear Gemini a lot." data-portal-copyright="" />

I’m definitely not alone in this reaction to Gemini creep. Recent studies have indicated that young people are less and less enthused about AI, and that they dislike it more the more they use the tools. Constantly nagging people to use a thing that they don’t like generally doesn’t go well. Just ask Microsoft, the company that spent two years stuffing Copilot into every nook and cranny it could find. The backlash has been loud, and the company is now walking some of that back.

And then there’s the matter of AI as a threat to the developer community — you know, the people Google addresses at I/O. Tech companies are laying off software engineers left and right, saying they don’t need as many warm bodies as AI coding tools have gotten better. I’m not sure that Gemini offering to help write your cover letter is much comfort when you’re applying for jobs currently being decimated by AI.

This is all before considering that companies like Google aren’t winning themselves any popularity contests as they push to build massive data centers around the country. But without even getting into all that, it’s just a bad user experience to constantly badger people into adopting tools they don’t want. I expect that kind of behavior from a Meta app, not a piece of software I use for work. I don’t want to “ask Gmail” when I open my inbox, I want to type in three keywords and find the email I’m looking for. I don’t want to chat with Gemini about my Chrome tabs. I don’t want to “learn the highs and lows” of a folder in my Google Drive. I want AI tools when I find them useful. Otherwise, I just want this stuff out of my face, and I don’t think I’m alone.

Photography by Allison Johnson / The Verge

The Razr Fold is stuck in the middle

2026-05-14T12:06:17-04:00

Can’t beat Moto on style.

For a phone that gets a lot right, the Motorola Razr Fold is frustratingly hard to recommend.

The Razr Fold is the company’s first book-style foldable, and it enters the US market with something not currently available on the competition: truly excellent battery life that rivals the best slab-style phones. No need to worry if you spend a long session gaming or working in a Google doc on the inner screen. The Razr Fold is going to easily get you through your day and then some. Samsung’s and Google’s foldables are much more likely to generate battery anxiety on a day of heavy use.

Also, the Razr Fold looks nice. The rounded corners, soft-touch back panel, and well-considered color options are very welcome and very Motorola. Nobody is making better-looking phones than this company right now. As a reminder, it’s okay to care about how your phone looks.

But the Razr Fold sets a very high bar for itself by virtue of costing $1,900. And at that price, it falls short in a few ways. This is a phone that puts all of its polish front and center — a great design, a big battery, and a clever multitasking system — but then drops the act around the edges. And there are a few too many rough edges for this high-end of a device.

Motorola Razr Fold

Motorola’s first book-style foldable is stylish, with a battery that goes all day and a little extra. But it’s a pricey entrant to a category with some well-established competition.

Score: 6

Pros	Cons
Excellent battery life Thoughtful multitasking interface Attractive styling	It costs $1,900 Some photo processing inconsistencies No magnets Bloatware bloatware bloatware

Where to Buy:

The Razr Fold’s 6,000mAh battery really is its biggest asset. It’s able to offer so much capacity thanks to its silicon-carbon technology, allowing it to store more energy than traditional lithium-ion batteries. Chinese phone makers are quickly adopting it while Apple, Samsung, and Google have been reluctant — likely because of some concerns about the battery tech degrading faster than straight-up lithium-ion. Motorola and its China-based parent company Lenovo seem not to share the same concerns, since this year’s entire Razr lineup uses silicon-carbon tech.

It’s hard to say how justified the concerns are around longevity since silicon-carbon hasn’t been used in phones for that long. In my extremely unscientific opinion and limited short-term testing, I can say that silicon-carbon absolutely kicks ass in a phone. I never even came close to triggering low power mode on the Razr Fold. Even on a day with a healthy amount of camera use and an hour using the inner screen, the battery was still above 50 percent by bedtime.

I never even came close to triggering low power mode on the Razr Fold

Motorola has also figured out a multitasking solution on the inner screen that rivals could learn from. It feels like a good compromise between Samsung’s “anything goes” approach and Google’s three app maximum. You can open two apps in split-screen or have one app occupy most of the screen with a second in the background just a quick tap away.

On top of that, you can embrace chaos and open multiple apps in small “freeform” windows you can resize and move around the screen. I got up to four apps running in this mode on top of two apps operating in split-screen before I decided it was too much stimulation. But I like this meet-in-the-middle approach that allows me to invite as much or as little chaos as I want.

I’m not saying you should run four apps at once, but I’m not not saying it." data-portal-copyright="" />

Battery and multitasking are pretty important things to get right on a folding phone. And yet, the Razr Fold disappoints in other ways. For starters, there are entirely too many preloaded apps on a device of this stature. Many, but not all of them, can be uninstalled. But just on principle, I don’t think anyone should have to spend time clearing out bloatware on a phone that costs this much.

The Fold is also missing some small but life-improving hardware upgrades. There are no Qi2 magnets built into the back of the phone, and I increasingly believe every phone should have magnets. The Fold supports wireless charging, but you won’t be able to thwack it onto a MagSafe-style charger like you can with the Pixel 10 Pro Fold.

And while I appreciate Motorola making a stylus for the Razr Fold, there’s no way to attach it to the phone and no case for the Fold with a place to store it. I forgot to bring it with me on more than one trip to a coffee shop where I intended to use it. The stylus is sold separately for $99 anyway, so maybe it’s a fair assumption that if you’re going out of the way to get one, you’re committed enough to figuring out the carry situation on your own.

Two phones for the price of two phones." data-portal-copyright="" />

Aside from the silicon-carbon battery, the Razr Fold doesn’t break any new ground hardware-wise. It’s less heavy and bulky than the Pixel 10 Pro Fold, but lacks that phone’s full dust resistance. And although the Razr Fold is leaner than the Pixel, it still constantly reminds me that I’m carrying it when it’s in my bag or pocket. The Z Fold 7, which costs $100 more than the Razr Fold, is appreciably smaller and lighter. That’s its whole deal. The Fold is kind of in no-man’s land between the incumbents, and that’s an uncomfortable place to be if you’re going to charge a premium.

Recent Motorola phones have been pretty weak in the camera department — forgivable on a $400 phone but not one that costs $1,900. There’s some good news: Image processing looks better on the Razr Fold than it has on past Motorola phones. I see less tendency toward overly bright, washed-out images. Photos in bright daylight show pleasant contrast and saturated blues. The 50-megapixel ultrawide is a cut above the usual 12-megapixel chip, producing plenty of sharp detail.

Consistency is the thing, though, and the Razr Fold’s camera system occasionally missteps. I noticed a jarring change in color cast a couple of times when switching between the wide and telephoto. The tendency to boost shadows too aggressively also shows up here and there, particularly when I’m using the 2x crop zoom, or photographing a gray, overcast scene outdoors.

I’m also not a fan of the generative AI-augmented digital zoom that kicks in above 20x. It introduces some artifacts when trying to clean up text — classic AI. There’s also an uncanniness to the way it processes small details in a scene, like distant features in the side of a building. They look too clean, and features are sort of smoothed out and blended in a way that just looks unnatural. Thankfully, there’s some text that pops up on the screen when it kicks in, and you can turn the feature totally off in settings.

Tent mode is still a pretty neat trick." data-portal-copyright="" />

If Motorola had launched the Razr Fold four years ago, I might feel differently about it. The Galaxy Z Fold was still growing out of its awkward too-tall shape, and the Pixel Fold didn’t exist at all yet. A folding phone with sophisticated styling? With a battery that lasts all day? That might have been worth $1,900 back then, considering there was nothing like that at the time.

But it’s 2026, which is an unfortunate time to be launching a pricey phone. The memory crisis surely played a part in the Razr Fold’s high price. Samsung and Google will likely launch successors to their folding phones later this summer, and I won’t be surprised if their prices go up. But here and now, the Razr Fold feels like a bit too little, for too much, too late.

Photography by Allison Johnson / The Verge

Agree to continue: Motorola Razr Fold

To actually use the Motorola Razr Fold, you must accept the following:

Motorola’s Privacy and Software Updates

But you also get to decide how Motorola’s support works on your phone:

Help improve Motorola products (optional)
Enhanced device support (optional)
Smart updates (optional)

After entering your Google account, you are asked to:

Add a phone number to your Google account (optional)

And you must agree to the following from Google:

You’ll also need to agree to the following on Google Services:

Install updates and apps: “You agree this device may also automatically download and install updates and apps from Google, your carrier, and your device’s manufacturer, possibly using cellular data. Some of these apps may offer in-app purchases.”
Use basic device backup (optional)
Use location (optional)
Allow scanning (optional)
Send usage and diagnostic data (optional)

Google Assistant:

You can set up Google Assistant (optional)
Activate Voice Match for Hey Google (optional)
Access Assistant without unlocking your device (optional)

To use Moto AI:

Motorola AI Terms and Conditions (optional)

Lastly, you have the option to join Motorola’s user community:

Give permission to Motorola to send push notifications about its services and benefits (optional)
Provide your email to Motorola (optional)

In total, you have to accept five main agreements and can bypass 14 when setting up the Motorola Razr Fold.

Gemini’s latest updates are all about controlling your phone

2026-05-12T14:07:37-04:00

Gemini Intelligence comes with a Liquid Glass-ish visual treatment. | Image: Google

It is, once again, Gemini season. Google is announcing a host of new Gemini features during its pre-I/O Android showcase, many of which aim to help use your phone for you. You’ll find Gemini in more places, like Chrome on Android, in your autofill suggestions, and all up in your apps — if you want.

Google also has a new name for us to remember, because it just can’t help itself: Gemini Intelligence. It “brings the very best of Gemini to our most advanced Android devices,” according to Google’s director of Android experiences, Ben Greenwood. Google is bundling some existing and new Gemini features under this name, and seems to be reserving them for premium Android phones like the Galaxy S26 series. Write that one down on your I/O bingo card.

Task automation is apparently one of those “best of Gemini” features. It’s already on some recent Pixel and Samsung Galaxy phones, and it enables Gemini to use certain apps on your behalf. It’s been limited to a handful of rideshare and food delivery apps until now. That’s changing “soon,” says Google, when task automation will open up to a wider range of apps.

It will also add multimodality; previously, Gemini could only use voice or text prompts to inform its actions. Now you can throw a screengrab or a photo into the mix, which kind of seems like something you should have been able to do from the start. You’ll be able to give Gemini a screenshot of a grocery list in your notes app and it will add those items to your cart. You know, provided you have an Android phone that supports Gemini Intelligence.

Another feature under the Gemini Intelligence umbrella that’s all new is Create My Widget. Google’s blog post calls it a first step toward “generative UI,” and it allows you to describe the functionality you want in natural language and let AI create a custom widget. Google’s examples include a custom weather widget for a cyclist who wants to see wind speed and precipitation at a glance, and a dashboard to surface specific recipe suggestions, like “three high-protein meal prep recipes every week.” These widgets will also carry over to Wear OS, so they’ll be available on watches, too.

I’m going to make so many unhinged widgets.

It’s a pretty simple idea on the face of it, but if you think about widgets as tiny apps you can vibe-code right onto your phone’s homescreen, it gets a little more interesting. Maybe this really is a small step toward an interface that just creates itself on the fly. Or maybe that’s a lot of pressure to put on a humble widget feature. Either way, I’ll be curious if we hear more during the I/O keynote about “generative UI.”

Google is also bringing Gemini features that exist on the desktop version of Chrome to its Android app. That means you’ll be seeing a Gemini button in Chrome where you can share the contents of a webpage and ask Gemini questions directly inside the browser. If you’re a subscriber to Google’s AI Pro or Ultra plans, you’ll also get auto browse to help complete tasks for you, like booking appointments. That will start rolling out in late June.

Gemini will be able to get more personal with its autofill suggestions. | Image: Google" data-portal-copyright="Image: Google" />

Gemini will also show up — optionally — in autofill on Android. You’ll be able to choose to connect Gemini to help fill out forms. This means Gemini can use its Personal Intelligence connection to things like your Google Photos and Gmail to look for the right information. In theory, that could mean things like pulling in your license plate number from Photos. Helpful? Creepy? Some mix of both? That’s Gemini season, baby. Gemini Intelligence features will “roll out in waves as they become ready throughout this year,” says Greenwood, with Galaxy and Pixel phones first in line to start getting updates this summer.