There are a total of 24 patterns for you to explore, and you can find all of them in the “Trending AI-UX Pattern” ebook by AIverse.
Pattern 1 — Linear back & forth
This concept is quite straightforward. ChatGPT’s chat functionality was the first access point to its powerful language model for everyday users like me. It’s akin to the command-line interface of the 1960s, but far easier to use. The chat user experience is what made it intuitive and allowed it to spread rapidly.
Honestly, incorporating a chat feature could be the solution for your product. Imagine if the chat could assist users in accessing data through a generative user interface, similar to Meter’s command. Additionally, consider the possibility of your product functioning as an API, integrated into a larger operating system-based chatbot.
Pattern 2— Non-linear conversation
Subform’s exploration of non-linear conversation reflects how we think. Humans don’t think in a straight line; we connect ideas and concepts. Why shouldn’t our thinking tools operate in a similar way?
Pattern 3— Context bundling
Figma’s dual-tone feature allows users to adjust the tone of text by simply dragging the cursor across a 2D matrix. This means there’s no need to search for the exact words; you can visually modify the tone using the grid. It’s easy and intuitive. While it functions similarly to prompting, it is integrated into a matrix format. Instead of being a standalone button like Grammarly, the goal is to incorporate pre-created prompts directly into the existing user interface.
Pattern 4— Living documents
Elicit’s bulk extraction feature exemplifies a hybrid approach that combines AI capabilities with a familiar Excel-like user experience. It includes subtle animations to load answers in bulk, ensuring that the loading process does not distract users. Additionally, the feature sets appropriate expectations by clearly marking any answers with “low confidence” if there is uncertainty associated with them.
Pattern 5— Work with me
This is my favorite example. Granola’s summarization feature effectively captures the “human in the loop” trend. They understand that not everyone, especially the founders of unicorn startups, wants the full transcription of a meeting, only to later spend time blocking out their calendar to review and extract key points. Instead, they create a summary based on your rough notes. The micro-interaction aspect is also worth exploring!
Pattern 6— Highlighting text content
Lex’s “@lex” comment feature is a great example of enhancing a user’s daily workflow. Microsoft Word and its commenting function have been fundamental to our activities on the internet. Allowing users to curate their output or even edit their own text by highlighting and commenting mimics the traditional process we use with paper and pen. As someone who writes regularly, I appreciate this approach because it provides me with autonomy while also offering support. There’s no need to switch apps or interrupt my flow.
Pattern 7—Do you exist?
Ford’s lane assist feature is an excellent example of a technology that operates seamlessly in the background. It takes control when needed and quickly hands it back to the driver without any hassle. This concept of AI-driven user experience (AI-UX) extends beyond digital products; it’s present in many aspects of our surroundings.
What began as a simple “sparkle” button has evolved to make everyday objects smarter. This aligns with the Agentive UX macro trend, where an AI assistant takes control from the user (“hand-off”) and returns it effortlessly (“take-back”). For feedback, the system not only provides visual indications but also uses sound alerts when the driver releases the steering wheel, along with haptic feedback when the system takes over steering.
This is a perfect illustration of a helpful agent—present when needed and unobtrusive when not.
