Google on November 10, 2026 extended a beta feature that lets Gemini work through other Android apps on a user's behalf, preparing food orders, ride requests and restaurant bookings, to the UK, India and Japan. The same post added the Pixel 10 line and Motorola's first qualifying handset to Gemini Intelligence, the set of system-level AI features Google introduced for Android in May. It also described three features that pull personal information from Gmail, Calendar and the phone itself into home screen widgets, device settings and the car dashboard.

The post was written by Karthik Thirumalai, Product Director for Android AI, and went up on Google's blog on the same day as the company's November Pixel Drop. It is short, a three-minute read according to the page, and carries eight footnotes. Most of them limit the features to select countries and languages and to users aged 18 and over, and several warn that heavier use may require a subscription.

In Short

Google's assistant can now work through food, ride and reservation apps for people in the UK, India and Japan, and more phones, including the Pixel 10 line and the coming motorola signature 27, qualify for its Gemini Intelligence features. The change matters to any company whose app could be operated by an AI on a customer's behalf, and to anyone tracking how much Gmail and Calendar data Gemini draws on. Most features are limited to adults in select countries and to recent premium hardware, and the Motorola phone opens for US pre-order on November 19.

What Google announced on November 10

Gemini Intelligence is Google's name for Gemini features built into Android itself rather than confined to the Gemini app. "Gemini Intelligence brings the best of Gemini to Android devices, helping your devices work harder for you," Thirumalai wrote. Google first announced the program on May 12, 2026, in a post by Mindy Brooks, VP of product management, tied to The Android Show. That May post said the features would roll out in waves, starting with the latest Samsung Galaxy and Google Pixel phones in the summer, and named multi-step task automation, Rambler and Create My Widget among them.

According to the November 10 post, the first wave reached the Pixel 11 and Samsung's Galaxy Z Fold8 Ultra, Z Fold8, Z Flip8 and S26 series over the summer. "Today, we're introducing new features built for Gemini Intelligence, and expanding capabilities to more of Android's most advanced devices," Thirumalai wrote.

The announcement covers six items:

  • the motorola signature 27, Motorola's first Gemini Intelligence device, which opens for US pre-order on November 19
  • Gemini Intelligence features for the Pixel 10 line through the November Pixel Drop
  • Rambler, a dictation tool, on the Pixel 10 line and the Motorola phone
  • task automation in the UK, India and Japan
  • Create My Widget and Device Assistance, two features that draw on mail, calendar and device data
  • proactive suggestions on Android Auto, due "soon", and Create My Widget for Pixel Watch within weeks

Which devices qualify

Motorola joins Google and Samsung

Motorola becomes the third phone maker in the program, after Google and Samsung. "Motorola is introducing its first Gemini Intelligence device, motorola signature 27, which is available for pre-order in the U.S. on November 19," according to the post. Neither the post nor Motorola's US product page gives a price or a shipping date. Motorola's page lists a Snapdragon 8 Elite Extreme Gen 6 chipset, a 6.8-inch display, a 7,000mAh battery and 90W charging, and describes the phone as designed for Gemini Intelligence.

For Motorola, the arrangement sits alongside a different kind of AI deal on its cheaper phones. On October 28, Taboola said its DeeperDive answer engine would appear on the lockscreens of select moto g phones, with a phased rollout across 2026 and 2027 in markets including the UK, Germany, Spain, France, Italy, Mexico and Argentina. The two announcements show the same OEM placing Google's system-level assistant on its flagship and a third-party answer engine on the lockscreen of its mass-market line.

Pixel 10 follows the Pixel 11

"And with today's Pixel Drop, Pixel 10, Pixel 10 Pro XL, and Pixel 10 Pro Fold get the latest features for Gemini Intelligence," the post says. That sentence leaves out the Pixel 10 Pro, which appears two paragraphs later in the list of phones getting Rambler and is included on Google's Gemini Intelligence device page, so the omission appears to be editorial. The Rambler sentence itself names "Pixel 10" twice, which also looks like an editing slip.

The Pixel 10a is excluded. Google's Pixel Drop post says proactive suggestions require a Pixel 10, except the Pixel 10a, or a newer Pixel, and the help page for Create My Widget sets the same limit.

The reason is hardware. According to Google's Android site, Gemini Intelligence is limited to devices that meet a set of specification requirements, including at least 12GB of RAM, a qualified chipset, on-device AI models and Android 17 or later. The same page names eight Pixel phones across the Pixel 10 and Pixel 11 lines, six Samsung phones, two Samsung tablets, five Googlebook laptops and two Pixel Watch models, with Motorola listed as coming soon. That bar places the program at the premium end of Android, and Google has not said how many devices in use meet it.

Task automation in three more markets

The feature with the clearest commercial consequences is task automation. According to the post, it "lets you ask Gemini to handle logistics like ordering food, hailing rides, or booking reservations across your favorite apps", and it "is expanding to the UK, India, and Japan."

Google's help page for the feature, titled "Ask Gemini to handle your multi-step tasks in select Android apps", fills in what the post leaves out. It describes the feature as a beta. It lists five countries: India, Japan, South Korea, the UK and the US, which means the November 10 expansion added three markets to the US and South Korea. The page adds that Pixel 10 devices are not supported in South Korea. Languages are limited to English, Japanese and Korean, and only some apps are covered. The user must be 18 or over and signed in with a personal Google account; work, school and supervised accounts are excluded. No subscription is required, though the page refers to daily request limits, and the post's footnote says higher usage may require one.

None of the five markets is in the European Union.

How Gemini operates the apps

The mechanics explain why the feature is different from a chatbot answer. According to the help page, Gemini completes tasks inside an on-device virtual phone environment, where it opens the target app and interacts with its screen. That makes it a computer use agent running on the handset. The task continues in the background even if the user closes the chat.

The user can watch progress live and stop the task at any point. On other supported devices the user can also take control, but the help page says that is not possible on the Pixel 10 or the Galaxy S26, where the only option is to stop. Google says Gemini aims to ask for review and confirmation before certain actions, and that the user carries out the final step, such as confirming a ride booking or placing an order. Among the possible mistakes the page lists is completing a purchase or ordering something without the user's permission.

The device list on the help page, as read on November 11, included the Pixel 10, Pixel 10 Pro and Pixel 10 Pro XL, the Pixel 11 range, the Galaxy S26 series, Z Flip 8, Z Fold 8 and Tab S12 series. It did not include the Pixel 10 Pro Fold or the motorola signature 27, both of which the post names as Gemini Intelligence devices.

Two routes into other companies' apps

Screen automation is one of two ways Google is connecting Gemini to third-party services. The other runs through integrations that companies build themselves. On October 13, Google added a tasks mode to the Gemini app for paying subscribers, citing more than 40 connected apps including Instacart, OpenTable, Adobe and Shopify. On October 28, it opened a preview of its Connected Apps Platform, through which developers bring their services into Search and the Gemini app using a cloud-hosted MCP server or Android's AppFunctions APIs.

The difference matters for the companies on the other side. An integration is something an app maker opts into and shapes. Screen automation works through the interface built for people, whether or not the app maker has built anything for Gemini.

Agents that act inside other companies' services have already met resistance. Amazon blocked Meta's Muse agent from shopping on its site in September, citing among other things that the agent did not identify itself when browsing. Google's own desktop agent sits at the top of its price range: it released Gemini Spark for Mac on June 30 only to US AI Ultra subscribers paying $99.99 a month. The Android feature needs no subscription to start, which makes this form of agentic AI available to adults with a qualifying phone in five countries.

Create My Widget pulls from Gmail and Calendar

Create My Widget, which the post says is rolling out over the next few weeks, lets users "design a personalized widget with a simple description in your own words", with "no coding required." The post's examples are a dashboard for an upcoming trip, a monthly spending tracker and a race day countdown. Create My Widget "can pull in information from Gmail, Calendar, and Search to create a custom widget just for you," according to the post.

Google's help page narrows the availability. The feature works for users aged 18 or over on the Pixel 10 and newer, except the Pixel 10a, in English in Australia, Canada, India, the United States and the United Kingdom. The page adds that Gemini Connected Apps integration is not available in the UK, and that users connect their Google Workspace when they set the feature up. Some features may require a subscription for higher usage.

Create My Widget was among the features Google named in May. It reaches Pixel 10 phones now and the Pixel Watch "in the next few weeks", which the Pixel Drop post specifies as the Pixel Watch 4 and Pixel Watch 5.

That reach intersects with a recent change in how Google treats such data. A privacy policy dated October 16 lets data from connected Gmail, Drive and Photos accounts shape ads for people who allow Search or Gemini Apps to access those services through Personal Intelligence, an exception to a 2021 pledge. Google's help page on that change says ads will not be based directly on the Gmail inbox, but may use summaries and inferences. The November 10 post does not say whether information that Create My Widget or the Android Auto suggestions read from Gmail falls under that policy, and it does not mention advertising at all.

Google launched Personal Intelligence in beta on January 14, 2026, connecting Gmail and Photos to Gemini with the feature off by default, and by late January its answers showed icons for the apps it had read. Gemini's handling of mail is also before a court. In Thele v. Google, a class action alleging that Gemini "smart features" were switched on in Gmail, Chat and Meet without consent, Google moved on September 29 to dismiss the refiled complaint after a judge had found no concrete harm in the earlier version.

Rambler and Device Assistance

Rambler "turns spoken thoughts into clean, organized text in real-time," according to the post, and is now available on the Pixel 10, Pixel 10 Pro, Pixel 10 Pro XL and Pixel 10 Pro Fold, as well as the motorola signature 27. Google's Pixel Drop post adds that Rambler can remove filler words such as "um". The post's footnote says it runs in Gboard, Google's keyboard, in select countries and languages, and is subject to usage limits.

Device Assistance is the other feature the post describes in detail. It "helps you control, change, or fix your device by just asking your phone for help," the post says. The examples are clearing storage space, checking which apps can access location, and muting distractions while letting important calls through. "Gemini deeply understands your device so it can act as your personal device expert," Thirumalai wrote. A footnote limits it to select Android devices running Android 17 or later, with capabilities that vary by device, language and country.

One of those examples is a privacy control: asking Gemini which apps can read location data puts the assistant between the user and Android's permission screens. The post does not describe what Gemini can change without confirmation.

Gemini's access to Android has widened step by step. In July 2025, the Gemini app gained automatic access to phone, messages and WhatsApp on Android, switched on by default and requiring users to turn it off manually. In December 2025, Google said its replacement of Assistant with Gemini on most mobile devices would run into 2026, with no completion date.

The car and the laptop

Android Auto

"Gemini Intelligence is designed to support you across whatever screen you're using," the post says. For Pixel owners with Gemini Intelligence, that will soon include proactive suggestions on Android Auto, "connecting the dots across your apps to surface helpful, personalized suggestions." The example given: when a friend texts to ask where a pilates class is, "Gemini Intelligence on Android Auto can find your confirmation email and instantly suggest sending the address with a single tap."

The footnote says the feature requires the Gemini app and setup, and is limited to users aged 18 or over in select countries and languages. No date is given beyond "soon".

The car is already a Gemini surface. On April 30, Google said Gemini was replacing Google Assistant in cars with Google built-in, starting in English in the US and reaching vehicles sold since 2020 through a software update, with Gmail and Calendar access promised later without a date. A day after the November 10 post, on November 11, Google said Gemini Live was rolling out globally on Apple's CarPlay to all Gemini users on iOS in every supported Gemini Live language, which puts Google's assistant on the dashboards of both major phone platforms.

Googlebook

"Last month we launched Googlebook, the first laptop designed for Gemini Intelligence," Thirumalai wrote. Google unveiled Googlebook on May 12 as a laptop platform combining Android and ChromeOS, with Acer, ASUS, Dell, HP and Lenovo as the first manufacturers. Google opened pre-orders on September 21 at prices starting at $899, including 12 months of Google AI Pro, according to its pre-order announcement. The laptops reached US shelves on October 4 and arrived on October 5 in Canada, the UK, Ireland, France, Germany and Australia.

What the post leaves unanswered

Several facts relevant to businesses are missing from the announcement:

  • Apps. Neither the post nor the help page lists which food, ride or reservation apps task automation supports in the new markets, whether app makers can opt in or out, or how Gemini chooses between competing apps when a request names none.
  • Price and dates. Motorola has not published a price for the signature 27, and the Android Auto suggestions carry no date beyond "soon".
  • Limits. Footnotes say higher usage may require a subscription, but the post publishes no thresholds.
  • Scale. The post gives no device or user counts.

Why this matters for marketers

The audience behind these features is large even if the qualifying hardware is not. Alphabet said on October 28 that the Gemini app had passed one billion monthly users, up from 650 million a year earlier. Gemini Intelligence is how that assistant moves from an app into the operating system of the most expensive Android phones.

For delivery platforms, ride-hailing companies and restaurants, task automation raises a question the post does not address. When Gemini rather than a person moves through an app, what happens to the promotions, sponsored listings and upsell screens those businesses pay for or rely on? The user confirms the final step, according to the help page, so a person still sees the end of the flow, but not necessarily the screens before it. For advertisers outside those categories, the open question is the one raised by Create My Widget: how Gmail and Calendar data read by Gemini relates to the October 16 policy on ads.

Apple sets the competitive frame. It shipped Siri AI on September 14 as an English-only beta, unavailable in the EU or China, and said at WWDC in June that its models were custom-built in collaboration with Google's Gemini family. Gemini's model family thus has a role on both major phone platforms, while Google's own Android assistant gains the ability to operate third-party apps.

That position is drawing scrutiny. On November 9, the day before Google's post, the Open Markets Institute published a report naming Google, Amazon, Microsoft and Meta as the companies best placed to dominate AI agents and proposing duties of loyalty and interoperability. The report contains no new data. The November 10 announcement adds three countries to the territory where one of those companies' agents can work inside other businesses' apps.

Timeline

Summary

Who: Google, through a post by Karthik Thirumalai, Product Director for Android AI, with Motorola as a new hardware partner alongside Samsung. The features reach adult users of qualifying Pixel, Samsung and Motorola devices, and affect app makers whose services Gemini can now operate.

What: Google extended Gemini's beta task automation, which works through other apps' screens to prepare food orders, rides and reservations, to the UK, India and Japan. It brought Gemini Intelligence to the Pixel 10 line, named the motorola signature 27 as Motorola's first qualifying phone, and introduced Create My Widget, which draws on Gmail, Calendar and Search, and Device Assistance. Widgets on Pixel Watch are due within weeks, and proactive suggestions on Android Auto "soon".

When: The post and the November Pixel Drop were published on November 10, 2026. The motorola signature 27 opens for US pre-order on November 19, and Create My Widget rolls out over the following weeks.

Where: Task automation now covers India, Japan, South Korea, the UK and the US, according to Google's help page. Create My Widget works in English in Australia, Canada, India, the US and the UK. None of the task automation markets is in the European Union.

Why: Google is moving Gemini from a standalone app into Android's operating system on premium hardware, where it can read personal data and act inside other companies' apps. For marketers, that changes who or what moves through an app's purchase flow and adds Gmail and Calendar data to the information Gemini uses, at a time when Google's privacy policy allows connected Gmail data to shape ads.