{"id":55989,"date":"2026-06-08T14:47:32","date_gmt":"2026-06-08T18:47:32","guid":{"rendered":"https:\/\/overcentral.com\/en\/?p=55989"},"modified":"2026-06-08T14:47:32","modified_gmt":"2026-06-08T18:47:32","slug":"apple-ai-shortcuts-natural-language","status":"publish","type":"post","link":"https:\/\/overcentral.com\/en\/apple-ai-shortcuts-natural-language\/","title":{"rendered":"Apple Brings AI to Shortcuts for Natural Language Workflows"},"content":{"rendered":"

Apple<\/a> is fundamentally reworking how users interact with its automation platform by integrating Apple<\/a> Intelligence directly into the Shortcuts app, transforming a tool once reserved for power users into something anyone can command with plain English. The update, announced at the company\u2019s Worldwide Developers Conference on Monday and slated to ship with iOS 27 later this fall<\/a>, allows users to describe what they want to accomplish in natural language rather than manually assembling a chain of actions, variables, and conditions. For a company that has long positioned Shortcuts as a bridge between simple app usage and genuine personal automation, this shift represents one of the most significant usability overhauls the application has received since its introduction.<\/p>\n

The Shortcuts App Before the AI Overhaul<\/h2>\n

Shortcuts began its life as a third-party application called Workflow, which Apple acquired in 2017 and later rebranded and deeply integrated into iOS, iPadOS, and macOS. The core premise remained consistent through every iteration: users could string together actions from different apps and system functions to create automated sequences that saved time and reduced repetitive manual steps. A user could, for example, build a shortcut that grabbed a screenshot, cropped it, converted it to a PDF, and emailed it \u2014 all with a single tap or voice command to Siri.<\/p>\n

The problem, however, was that building these automations required a level of technical literacy that many users simply did not possess. The visual-scripting interface, while elegant in its drag-and-drop design, still demanded an understanding of variables, logic gates, action ordering, and app-specific parameters. Finding the right action, configuring it correctly, and debugging a broken workflow were tasks that frequently overwhelmed casual users. Apple itself acknowledged this friction during the keynote. Celcia Dantas, Senior Manager of Home Software Product Marketing, stated plainly that \u201cwhile super powerful, the process of creating these shortcuts can feel, well, complicated.\u201d<\/p>\n

How Apple Intelligence Unlocks Natural Language Workflows<\/h2>\n

The new natural language feature is powered entirely by Apple Intelligence, the company\u2019s on-device AI framework that processes language, images, and context without sending user data to external servers. When a user types or speaks a request into the updated Shortcuts app, Apple Intelligence parses the semantic meaning of the description, identifies the relevant apps, system functions, and data inputs required, and then assembles the necessary steps automatically.<\/p>\n

This means a user no longer needs to know, for instance, which specific action in the Maps app fetches an estimated travel time or how to structure a conditional statement that checks whether a location has changed. The AI handles those decisions behind the scenes. The user simply writes something like \u201cNotify my partner when I leave work with my ETA,\u201d and Shortcuts determines the rest: triggering the automation when the user\u2019s location matches a stored work address, querying Apple Maps for the drive time, composing a message with the relevant information, and sending it through Messages.<\/p>\n

The system is not limited to simple one-line commands. Apple indicated that users can layer complexity into their requests, describing multi-step workflows that involve several apps and conditions, and Shortcuts will attempt to interpret and build the entire sequence. The AI also understands context and intent, meaning it can infer what the user means even when the phrasing is imprecise or incomplete.<\/p>\n

What Makes This Different From Traditional Shortcuts Creation<\/h3>\n

Traditional Shortcuts creation required explicit instruction. Every action, every input, every output had to be manually placed and configured. The new approach shifts the paradigm from explicit programming to intent-based generation. The user provides the desired outcome, and the system reverse-engineers the process. This is a meaningful conceptual change because it removes the primary barrier to entry for non-technical users: the need to understand the internal logic of how automations are structured.<\/p>\n

Apple Intelligence does not entirely eliminate the visual-scripting interface for those who prefer it. Users who want fine-grained control over their workflows can still open the action editor and inspect, adjust, or override what the AI has generated. This dual-mode approach \u2014 AI-assisted creation alongside traditional manual editing \u2014 ensures that the update serves both newcomers and experienced users without forcing either group into a workflow they find limiting.<\/p>\n

Real-World Application: The Leaving Work Automation<\/h2>\n

During the demonstration at WWDC, Dantas provided a concrete example that illustrates how the natural language feature functions in practice. A user could design a shortcut that automatically notifies their partner when they depart from work and provides an expected time of arrival at home. In the current version of Shortcuts, building this automation would require the user to manually select a location trigger, configure the Maps ETA action, set up a Messages action with the correct text template, and ensure the whole sequence runs without errors.<\/p>\n

With the new AI-powered<\/a> approach, the user simply types a prompt describing the desired behavior: something along the lines of \u201cWhen I leave work, send my partner my ETA.\u201d Shortcuts, powered by Apple Intelligence, interprets the request, identifies the user\u2019s stored work address (likely from their Contacts or Maps history), sets the location trigger, accesses the Maps ETA function, constructs a message that includes the travel time, and designates the recipient from the user\u2019s Messages contacts. The entire workflow is generated in seconds.<\/p>\n

Equally important is the ability to modify an existing shortcut using natural language. In the same example, if the user later decides they want to add an action that starts playing a favorite podcast when the shortcut runs, they can simply describe that change. \u201cAdd an action to play my favorite podcast when this shortcut runs\u201d is enough for the AI to locate the relevant Media action and insert it into the existing automation at the appropriate point. This edit-by-description capability removes the friction of having to navigate the action library and manually reconfigure the workflow.<\/p>\n

The Broader Implications for Automation and Accessibility<\/h2>\n

Bringing natural language input to Shortcuts is not merely a convenience feature; it represents a strategic expansion of who can benefit from automation. The Shortcuts ecosystem has always been one of the most powerful tools in Apple\u2019s software stack, capable of orchestrating complex interactions between apps, system settings, and hardware sensors. But its power was underutilized because the learning curve deterred all but the most determined users. By lowering that curve dramatically, Apple is effectively opening the door to a much larger audience.<\/p>\n

This shift has implications beyond individual productivity. In education, students and teachers who lack programming experience can now build automations for their daily workflows. In small business settings, employees can create task-specific shortcuts without waiting for a technically inclined colleague to set them up. In accessibility contexts, users who rely on voice input or who find visual-scripting interfaces challenging can describe what they need in natural language and receive a working automation in return.<\/p>\n

The update also positions Shortcuts more competitively against other automation platforms. Microsoft Power Automate and similar enterprise tools have begun incorporating AI-assisted workflow creation, and IFTTT (If This Then That) has long offered a simplified interface for connecting services. Apple\u2019s approach, however, benefits from deep system-level integration. Shortcuts can control hardware features, network settings, system toggles, and every app that exposes actions through the Shortcuts framework \u2014 a level of access that third-party automation tools cannot match on iOS.<\/p>\n

How Apple Intelligence Interprets Natural Language Descriptions<\/h2>\n

Understanding how Apple Intelligence translates a user\u2019s description into a functional shortcut requires looking at how the system processes language. The on-device model performs several tasks in sequence. First, it identifies the overall intent of the user\u2019s request: is the user trying to send a notification, change a setting, fetch information, or combine multiple actions? Second, it extracts key entities from the text \u2014 people, places, times, apps, data types \u2014 and maps them to the corresponding parameters in the Shortcuts ecosystem. Third, it determines the most logical order of operations and any conditional logic that might be needed.<\/p>\n

For example, if a user types \u201cEvery morning at 7 AM, check the weather and if it\u2019s going to rain, send me a reminder to grab an umbrella,\u201d Apple Intelligence must recognize a recurring schedule, a weather-check action, a conditional branch based on precipitation data, and a notification action. The AI handles all of this implicitly, generating a shortcut that includes a time-based trigger, a Weather app action, an if-condition that evaluates the forecast, and a Reminders or Messages action to deliver the alert.<\/p>\n

Apple has not disclosed the specific architecture of the language model used for this feature, but it is consistent with the company\u2019s broader approach to on-device AI: models are trained to handle a wide range of user intents while maintaining privacy by processing everything locally. The system does not require an internet connection to interpret a request and build a shortcut, which also means it works offline and avoids the latency associated with cloud-based inference.<\/p>\n

Editing and Iterating With Natural Language<\/h2>\n

One of the most practical aspects of the update is that the natural language interface is not limited to initial creation. Users can return to any shortcut they have already built \u2014 whether it was created with AI or manually assembled \u2014 and make changes by describing them. This iteration loop is critical for real-world use because automations rarely come out perfect on the first try. A user might find that their morning weather shortcut runs too early, or that the notification message could be more informative, or that they want to add an additional action.<\/p>\n

In each case, the user can simply describe the change. \u201cChange the trigger to 7:30 AM,\u201d \u201cAdd the high temperature to the message,\u201d or \u201cAlso set the thermostat to eco mode when I leave work\u201d are all valid requests that the AI can interpret and apply. This eliminates the need to drill into the action editor, locate the specific parameter, and manually adjust it \u2014 a process that, while straightforward for experienced users, can be intimidating for those who are less familiar with the interface.<\/p>\n

Availability and Rollout With iOS 27<\/h2>\n

The updated version of Shortcuts with AI-powered natural language creation will be part of iOS 27, which Apple announced will ship later this fall. The feature will be available on devices that support Apple Intelligence, which includes iPhones and iPads with the A17 Pro chip or later, as well as Macs with Apple Silicon. This hardware requirement is necessary because the on-device language models that power the natural language interpretation require the neural engine capabilities found in Apple\u2019s more recent silicon.<\/p>\n

Users who run the developer or public beta versions of iOS 27 during the summer months will likely gain early access to the feature, allowing them to test the natural language workflow creation and provide feedback before the public release. Apple has a history of refining features based on developer and beta user input, and Shortcuts \u2014 a tool that interacts with so many parts of the system \u2014 is likely to receive adjustments before the final rollout.<\/p>\n

The timing of this update is also notable. Apple has been steadily expanding the capabilities of the Shortcuts app with each major iOS release, adding more actions, improving the editor, and integrating deeper with third-party apps. The addition of AI-powered natural language creation represents the most significant user-facing change to the app in years and signals that Apple sees Shortcuts as a core part of its productivity strategy rather than a niche utility for enthusiasts.<\/p>\n

What This Means for the Future of Shortcuts<\/h2>\n

The introduction of natural language workflows does not mean the end of the visual-scripting editor, nor does it signal that Apple has abandoned power users. Rather, it suggests a maturation of the platform: Shortcuts is evolving from a tool that required its users to learn its language into a tool that learns the user\u2019s language. This is the same trajectory that many software products have followed as AI models have become capable of understanding human intent more accurately. The best tools are those that adapt to the user, not the other way around.<\/p>\n

Apple has also positioned this feature as part of a broader AI strategy that includes system-wide writing tools, image generation, and contextual intelligence across apps. Shortcuts, in this context, becomes another surface where Apple Intelligence can deliver tangible value \u2014 not by replacing user agency, but by reducing the friction between an idea and its execution. For millions of iPhone and iPad users who have never built a shortcut because the process seemed too technical, that reduction in friction could be transformative. The automation tools have always been there. Now, for the first time, the barrier to using them is a sentence, not a manual.<\/p>\n","protected":false},"excerpt":{"rendered":"

Apple is fundamentally reworking how users interact with its automation platform by integrating Apple Intelligence directly into the Shortcuts app, transforming a tool once reserved for power users into something anyone can command with plain English. The update, announced at the company\u2019s Worldwide Developers Conference on Monday and slated to ship with iOS 27 later […]<\/p>\n","protected":false},"author":7,"featured_media":94803,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"fifu_image_url":"https:\/\/cards.overcentral.com\/cards\/en\/55989.png","fifu_image_alt":"Apple Brings AI to Shortcuts for Natural Language Workflows","footnotes":""},"categories":[31],"tags":[],"class_list":["post-55989","post","type-post","status-publish","format-standard","has-post-thumbnail","category-technology"],"fifu_image_url":"https:\/\/cards.overcentral.com\/cards\/en\/55989.png","fifu_image_alt":"Apple Brings AI to Shortcuts for Natural Language Workflows","_links":{"self":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts\/55989","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/users\/7"}],"replies":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/comments?post=55989"}],"version-history":[{"count":0,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/posts\/55989\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/media\/94803"}],"wp:attachment":[{"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/media?parent=55989"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/categories?post=55989"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/overcentral.com\/en\/wp-json\/wp\/v2\/tags?post=55989"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}