Apple’s latest iPhone software, iOS 26, is now on general release. iOS 18.7, also just released. Apple usually runs the two versions in parallel for maximum two months, to give the cautious time to get used to what’s new. Staying off iOS 26 for now may also be appealing if you don’t care for the signature update: Liquid Glass. This has been warmly received but not universally liked. Still, be warned, you will have to update sooner or later if you want full security protections. iOS 18.7 is for all iPhones from 2018 onwards. That means the iPhone Xs, iPhone Xs Max and iPhone XR and all iPhones since, including the iPhone SE in both its second- and third-generation models. And all members of the iPhone 16 series are included, right up to the iPhone 16e. Note that iOS 26 is available to all of these phones except the first three. It’s compatible with iOS 11 and onwards. Apple is tight-lipped about what’s in iOS 18.7, merely saying “This update provides important bug fixes and security updates and is recommended for all users.” It hasn’t yet updated with full details of exactly what bugs have been squished and usually releases details of security concerns later, when users have had a chance to update to protect themselves.
Google brings Gemini in Chrome; with agentic browsing capabilities Gemini can now work across multiple tabs, allowing users to quickly compare and summarize information across multiple websites
Google is rolling out Gemini in Chrome to all Mac and Windows desktop users in the U.S. after previously limiting the capability to Google AI Pro and Google AI Ultra subscribers. The tech giant also announced that it’s bringing agentic capabilities to Chrome in the future, adding its AI Mode search feature to the address bar, launching new Gemini features, using AI to combat AI-generated scams, rolling out automatic password resets, and more. U.S. users who have their language set to English can now ask Gemini to clarify complex information on any web page they’re reading using the Gemini icon in the top-right corner of their Chrome window. Gemini can now work across multiple tabs, allowing users to quickly compare and summarize information across multiple websites. Gemini will soon be able to retrieve web pages you’ve previously visited, making it easier to return to past browsing sessions without sifting through your browser history. Additionally, Google is launching a deeper integration between Gemini in Chrome and other Google apps, like Calendar, YouTube, and Maps. Google says this will allow users to do things like schedule meetings, see location details, and more without having to leave the page they’re on. Google notes that the AI assistant will be able to complete tedious tasks, like booking a haircut or ordering weekly groceries. Gemini will navigate to the site, add things to your cart, and let you take the final action by checking out with your payment option. Google is also bringing AI Mode, its advanced search feature, directly into the Chrome address bar. With AI Mode, users can ask complex questions with follow-ups to dig deeper into topics. Chrome can now suggest relevant questions based on the context of the page to kickstart your search in the address bar. Google says users will get a helpful AI Overview and the option to ask follow-up questions with AI Mode. The company says Chrome will also soon be able to use its Gemini Nano model to detect and protect against scams, such as fake virus alerts and fraudulent giveaways. These scams often impersonate trusted brands and use generative AI to create convincing phishing attempts. Google also announced that it’s using AI to help users fix compromised passwords with a single click on supported sites, like Coursera, Spotify, Duolingo, H&M, and more.
Google now lets you share your custom Gemini AI assistants known as Gems with friends, family, or co-workers
Google is making it possible to now share your Gemini Gems — custom AI assistants and experts designed for specific tasks — the company announced on Thursday. The feature launched last year, initially as part of the Gemini Advanced paid subscription, allowing users to write instructions to create an AI chatbot for different scenarios. For instance, Google launched with premade Gems like a learning coach, a brainstorming assistant, a career guide, a writing editor, and a coding partner. Now Google says you’ll be able to share your Gems with friends, family, or co-workers as easily as you can share a file from Google Drive. This would make Gems more accessible to more people, as not everyone uses the advanced customization feature. It could also help prevent people from building the same Gems as others. For instance, if multiple co-workers were using a similar type of custom Gemini assistant, they could just share the same resource instead of each making their own version that could have slight inconsistencies between them. Google suggests Gem sharing could also be useful for people working on family vacation plans and guides, meal planners, or collaborative writing projects. Also similar to Google Drive, you can control who can view and use your Gems and who’s allowed to edit them.
Google TV gains presence sensing through TCL QM9K starting at $2,999 with automatic wake functionality and conversational Gemini AI assistant
Google TV is adding support for TVs with presence sensing hardware, but the first TV set with support is not cheap, as the TCL QM9K is now available starting at $3,000. The TCL QM9K series is now on sale at Best Buy in the US. The new flagship TV from TCL is a 4K QD-Mini LED TV, and Google TV is taking advantage of its new hardware. Firstly, this is the first place that Gemini is debuting, at least according to TCL. More importantly, though, the TCL QM9K is the first Google TV set with presence sensing hardware built in. Using mmWave radios, the TV can detect when someone is in the room and turn on the TV to show an ambient screensaver or select information. TCL previously explained that this can be adjusted to your preferences, including the distance from the TV in which the presence sensor is used. The TCL QM9K series starts at $3,000 at Best Buy. That’s for the 65-inch model, with larger sizes of course being more expensive. Pricing breaks down as follows: 65-inch: $2,999; 75-inch: $3,499; 85-inch: $3,999; and 98-inch: $5,999. Sales will likely follow in time, though. TCL’s QM8K series started at $2,499 when it launched earlier, but you can now get a 65-inch model for $1,500.
Stripe and Visa team to power Fold Bitcoin credit card- a bitcoin-only rewards product designed to turn everyday spending into a direct path to bitcoin ownership
Bitcoin financial services company Fold has selected Stripe, the programmable financial services company, to power the upcoming launch of the Fold Bitcoin Credit Card™, a bitcoin-only rewards product designed to turn everyday spending into a direct path to bitcoin ownership. The card enables users to accumulate bitcoin with every purchase, offering a simple and consistent way to build long-term wealth. Issued on the Visa network and powered by Stripe Issuing, the Fold Bitcoin Rewards Credit Card delivers up to 3.5% back on every purchase, with no categories and no deposit requirements. Cardholders earn an unlimited 2% back instantly, plus up to 1.5% back when they pay off purchases using their Fold Checking Account with qualified activity. In addition, cardholders can earn up to 10% back with top brands in the Fold rewards network, including Amazon, Target, Home Depot, Lowe’s, Uber/Uber Eats, Starbucks, DoorDash, Best Buy, and hundreds more. Fold’s reward system is designed to be simple and transparent, offering bitcoin-only rewards without the complexity of tokens, staking tiers, or exchange lock-ins. Fold’s integration with Stripe Issuing marks a key milestone in Fold’s product development and reflects growing demand for digital asset integration in consumer financial tools. With Stripe’s infrastructure in place, Fold is positioned to bring the Fold Bitcoin Credit Card™ to market with the reach and reliability users expect.
Mastercard debuts program to help its customers reach new markets, offering customers a tailored and scalable go to market approach
Mastercard introduced its Global Reach Partner Program, a new global expansion framework that makes it easier for customers to expand into new markets around the world. Mastercard Global Reach offers customers a tailored and scalable go to market approach that streamlines processes, accelerate launches, and delivers against their goals to serve cardholders across multiple markets. Program partners will also be able to leverage Mastercard’s consulting, advisory, and implementation expertise to support with their go-to-market definition, target operating model design, and operational roll out. Mastercard’s Global Reach Partner Program features include: Speed to Market: Accelerating international market entry; Simplified processes: Removing barriers and providing greater flexibility; Customized approach: Tailored to each customer’s unique needs; Efficiencies: Streamlining and helping to manage operational cost; Dedicated global enablement team: Providing consistency and in-market expertise across multiple markets.
Apple’s Foundation Models Framework gives developers API access to on-device 3-billion-parameter LLM powering 20+ apps including SmartGym, Stoic, and OmniFocus with zero-cost inference, no token limits and offline functionality
Proving Apple Intelligence’s worth, third-party developers are now using it to make apps more personal to users, and users more productive. By using the Foundation Models Framework, developers get an API that means they can pass prompts to Apple Intelligence. It’s done privately, and with specific limitations — but also specific freedoms: No limit on user requests; No tokens or API keys for the user to install; Access to the same Apple Intelligence on device. That last part is significant, because it is both a limitation and a guarantee of privacy. Developers can’t use the full Apple Intelligence LLM in the cloud, nor can they use extensions such as directly making requests of ChatGPT. Nonetheless, developers have been implement Apple Intelligence across a wide range of apps. Apple has now picked out more than 20 to champion, ranging from To Do apps to mental health ones. CellWalk takes users through a 3D journey around molecules. Now it can automatically tailor its explanations to the user’s level of knowledge. More heavyweight To Do apps such as OmniFocus are adopting Apple Intelligence too. It can generate whole projects and suggested tasks to help a user get started. Starting with iOS 26, iPadOS 26, and macOS 26, Apple provides app developers with access to a new Foundation Models framework that allows their apps to tap into the on-device LLM at the core of Apple Intelligence. Apple today highlighted some of the iPhone, iPad, and Mac apps utilizing the Foundation Models framework to power new features and capabilities. For example, in the fitness app SmartGym, users can now describe a workout and turn it into a structured routine with sets, reps, rest times, and equipment adaptation. SmartGym also generates insightful summaries of workout data, and more. In the journaling app Stoic, users can now receive contextual journaling prompts that are generated from their recent entries, and there are other new features. SwingVision, an app that helps users with their tennis or pickleball skills, uses the framework to generate advice for players to improve their game. Foundation Models powers new Listen Mode and Scan Mode options in the to-do app Stuff. The task management app OmniFocus 4 can now generate projects and next steps on a user’s behalf, such as helping them know what to pack for an upcoming trip.
Agent Mode in Excel and Word uses OpenAI reasoning models to orchestrate multi-step tasks, reaching 57.2% SpreadsheetBench accuracy compared to 20% for original Copilot, with PowerPoint integration coming soon
Microsoft Corp. envisions a future where everyone starts “vibe working” with the launch of its new Agent Mode in Office Apps and Agent Mode in Copilot Chat offering new ways to automate business and personal work. The company said the new agents will “do work on your behalf” by automatically spinning up “high-quality Office documents, spreadsheets and presentations” upon the user’s command. Users now have the ability to “steer” Microsoft’s Copilot tool from within those apps as it orchestrates complex tasks within them. Agent Mode can also create a loan calculator that will figure out monthly payments based on user inputs, such as the loan amount, interest rate and loan term. It will then generate a payment schedule in a neatly formatted table. Alternatively, it can generate a monthly financial analysis of a small business’s spending, complete with a breakdown by different product lines. As for Agent Mode in Word, this is designed to make more advanced features available to users. It’s true that most office employees are probably comfortable in Word and can already create a basic document – but they may not be familiar with many of its more advanced formatting tools for fleshing out documents and making them look more professional. The end result can take advantage of Word styles, formatting and other features to make sure the document not only reads well but looks good. Also announced was a new Office Agent in Copilot, which makes it possible for users to create Word and PowerPoint documents from directly within the Copilot Chat user interface. Excel integration will be added soon too. With this, all the user has to do is set Office Agent in Copilot a task, and it will follow up by clarifying the user’s intent, then engaging in web-based research to create whatever document is required. As it does this, it will reveal its “chain of thought” processing, so users can check how it’s going about each task. Once it’s finished, the Office Agent in Copilot will present a preview of its new document. At this juncture, the user can then go through it and request any amendments they feel are necessary.
AWS debuts Amazon Quick Suite for enterprises, an agentic AI teammate connecting 50+ apps to answer, visualize and automate work across Office 365, Slack and more via Spaces and connectors.
Amazon Web Services Inc. unveiled a new AI tool Amazon Quick Suite, which collaborates with users to get work done by connecting with enterprise information sources to provide answers, take action and automate work through an intuitive web-based interface and integrations with Microsoft Corp.’s Office 365, Slack and other digital work spaces. Quick Suite includes more than 50 built-in connectors to provide a complete view of company data and productivity tools, including Adobe Analytics, SharePoint, Snowflake and Google Drive. With integrations such as OpenAPI and Model Context Protocol customers can link in custom resources from even more services including Atlassian, Asana, Box and Canva among more than 1,000 apps. Users can organize these data sources into Spaces, a sandbox that integrates information and applications so the platform can index and understand company and personal context. Using this broad connectivity, users and teams can build custom AI agents capable of focusing on specific aspects of their work, conducting research and taking action. The platform offers powerful capabilities that leverage this unified view across apps and data. With Quick Suite, business users can request analytical insights using AI agents to pull in data from multiple business apps and build dashboards that combine business intelligence and research. Going beyond analytics and visualization, Quick Research provides an agent that can dive deep into data, web resources and company knowledge to deliver comprehensive reports on any topic. Routine and complex tasks can also be managed through Quick Flows and Quick Automate. Flows allow users to create digital workflows using simple prompts to handle repetitive work, reducing human error and freeing teams from busywork. Automate can coordinate complex business processes, even those spanning hundreds of steps and multiple departments. Marketing automation company Propulse Lab used Quick to streamline its customer service operations, reducing average ticket-handling time by 80%. With this performance scaled across its business, the company estimates it will save 24,000 hours annually.
Epic Games says Apple’s new install process cuts user drop-offs by 60%; includes only one screen showing users that they might miss some features, such as subscription management
Fortnite maker Epic Games said that Apple’s new installation workflow, implemented in iOS 18.6, has had a positive impact on its user growth, with a 60% decrease in drop-offs. The new process includes only one screen showing users that they might miss some features, such as subscription management offered by the App Store. The screen also informs users that the alternative app store’s developer will handle their data directly. Epic Games said that prior to the iOS 18.6 update, 65% of users attempting to install its game store abandoned the process. However, after the update, that figure dropped to 25%. The company noted that this drop rate is closer to the rate it observes with Windows and macOS users installing its game store. Despite this improvement, the company remains critical of Apple’s policies, including charging a core technology fee, imposing notarization and approval policies, and implementing rules that make it difficult for developers to distribute their apps through alternative app stores.