Emerging AI-Enhanced Features

As I observe the ongoing developments in Generative AI, several distinct trends are becoming apparent. These patterns can be analyzed through traditional design elements found in AI features or by examining user experience patterns. The trends I have identified emphasize how Generative AI is being utilized to enhance the effectiveness of products in fulfilling their intended purposes. These trends manifest either through a ripple effect—where one company's innovation inspires a wave of imitators, each adapting the concept for their specific user base—or through a convergence around genuinely transformative capabilities of Generative AI. These exemplary applications are now standard on the roadmaps of many companies, showcasing the tangible value this technology offers.

Consider the rise of mobile technology: early mobile apps often merely replicated existing desktop services, sometimes awkwardly adapted for smaller screens. Many were simplistic mobile versions of company websites, extending existing functionalities without unlocking new potential inherent in mobile platforms.

However, as companies became more adventurous in experimenting with handheld device capabilities, alongside the growth of high-speed internet, genuinely innovative solutions began to surface. Uber, for example, utilized the ubiquitous access of mobile technology to revolutionize ride-hailing, transforming not just the service but creating substantial value by fully leveraging the platform’s capabilities. Just as mobile technology evolved by embracing its unique attributes, the LLM revolution is now poised to transform our interactions with digital environments in equally significant ways.

Content Rewriting: A Natural Starting Point

One of the most significant applications of LLM technology is in content rewriting, which effectively utilizes these systems’ robust text generation and refinement capabilities. This application is a natural fit, enhancing user content while interacting with a service. Early implementations included Notion's feature that allowed users to turn brief inputs into organized templates, alongside Grammarly's expansion of its editing tools for comprehensive content rewrites. These integrations represent a logical evolution for products aimed at streamlining and improving written content.

As more companies adopt this application, innovative methods for assisting users in rewriting content are emerging. In sales, both HubSpot and Salesforce have developed systems that create personalized outreach to potential clients by utilizing stored information. This capability generates a demand for high-quality data, reinforcing the need for investment in their respective CRMs. UpWork, known for matching freelancers with job opportunities, has a feature that enhances job postings to attract more qualified candidates. On the consumer side, Bumble recently introduced a tool to generate icebreaker messages.

In all these examples, features go beyond simple content rewriting to enhance the effectiveness of the message. Some improve content quality based on best practices, while others identify unique elements to include in the message based on additional variables collected within the system. This can even result in more concise messages, boosting overall efficiency.

Summarizing Content Overload

While content rewriting remains a common application of Generative AI, the extraction of actionable insights from large volumes of information is rapidly becoming a crucial use case. Summarization is the most straightforward method, processing extensive data to identify patterns and key points. For instance, LinkedIn’s article summarization feature condenses lengthy posts and articles into manageable snippets, making complex information easily digestible.

Similar to summarization but with an evaluative component, features like Microsoft Teams' CoPilot’s call transcript summaries distill lengthy discussions into vital bullet points, highlighting significant moments or insights. HR technologies like Rippling AI and Greenhouse have introduced features that do the same for interviews, summarizing candidate transcripts and interviewer notes to outline strengths and weaknesses. ServiceNow is applying this in the B2B arena, summarizing support tickets and identifying patterns, with plans to expand into report creation and advanced search features.

With summarized data, the next logical step, particularly in SaaS products, is report generation. By identifying key variables or common patterns, products can automatically generate deliverables for users to build upon. ServiceNow offers this for customer service, while HubSpot does the same for sales data. Tableau takes it further by automatically producing summaries based on trending data. Priceline has also launched Generative AI features, including automated itinerary creation, which outlines a plan for users during their trips. Though still in the early stages, there is significant potential for further development and value creation.

The capacity to 'understand' nuanced language through summarization naturally extends to advanced search functionalities. ServiceNow allows customer service agents to search tickets for recommended solutions and clarify jargon used by various agents. Similarly, ZenDesk retrieves relevant information related to service tickets, while Concur facilitates faster expense report submissions by pulling up pertinent documents.

Dropbox has introduced new features allowing users to ask questions about documents stored within their systems, quickly extracting insights without needing to open multiple files. Glean employs an interactive chatbot for continued questioning. The most innovative applications involve integrating quantitative data, as seen with Mixpanel’s Spark AI, which enables users to query their data and uncover insights from metrics.

The final aspect of deriving value from emerging content is recognizing novel insights. Mixpanel and Tableau, among others, are already doing this alongside a range of AI-specific products. Gong, a sales enablement tool, has introduced a feature that identifies successful trends during sales calls using Generative AI to detect patterns across similar yet different expressions, pinpointing effective communication strategies. On the consumer side, Fitbit leverages individual user actions and collective data to deliver personalized health insights. Combining tailored messaging with these insights, it seems likely that nudging users toward personalized actions will be the next step.

These design patterns illustrate the multifaceted capabilities of summarization as a core component of Generative AI. As experimentation continues, we can anticipate a proliferation of these applications, with successful models being refined and replicated across various industries. Just like content rewriting, summarization and its various forms represent just one of many emerging strengths of Generative AI, paving the way for future innovations. Another emerging strength is the ability to generate evidence-based predictions grounded in the collective knowledge embedded in the foundational LLMs due to their extensive training data.

Tapping into Collective Knowledge

Despite some controversies and unresolved issues, the undeniable fact remains that large language models (LLMs) are trained on an extensive body of human knowledge, encompassing a diverse array of data sources and content types. Although these systems may not truly “understand” content in the human sense, they excel at recognizing patterns in human behavior. This ability underpins their effectiveness in role-playing specific personas, achieved by crafting prompts that guide the system to adopt a desired stance.

Organizations are just beginning to unlock the potential of this collective knowledge in their product features. Rather than solely focusing on content creation or manipulation, new applications of these systems provide fresh perspectives and predict outcomes based on accumulated human experiences. The true value of these applications lies not just in enhancing efficiency but also in improving effectiveness, enabling users to make more informed choices. While still not widespread, I anticipate this approach becoming a cornerstone of company strategies, marking the next evolutionary phase in AI application. When an LLM generates a new insight, the immediate follow-up question becomes, “What actions should I take based on this information?” Employing Generative AI to guide decisions based on shared experiences offers a powerful tool for informed decision-making.

Scenario Planning

In both professional and personal contexts, we often face situations that seem unique to us. However, others have likely encountered similar challenges. Generative AI can tap into this vast repository of experiences to offer guidance and predict potential outcomes. For instance, BetterUp’s Difficult Conversation Scenario Planner uses this technology to help users strategize for difficult interactions. By simulating various conversational paths, the tool reveals the potential impacts of different approaches and helps users anticipate dialogue elements. In a B2B context, companies like AMEX are exploring Generative AI to predict customer behaviors, aiming to enhance financial planning processes.

Empathy Building

Beyond scenario planning, these systems can foster understanding between individuals. Our divisions often arise from a lack of comprehension, and Generative AI, trained on a wide range of perspectives, can serve as a bridge. An experimental feature I’ve observed enabled users to explore their potential reactions in various scenarios, fostering an understanding of why others might respond differently. While the full realization of this technology is still forthcoming, LinkedIn’s current feature suggesting questions following posts hints at future possibilities. Users could inquire, “Why would someone hold this perspective?” or “Why would this user feel comfortable sharing this?” Generative AI could then provide predictions that nurture empathy among participants.

As we continue to unlock the capabilities of summarization and the utilization of collective knowledge, these new Generative AI applications not only promise to replicate knowledge but also to contextualize and humanize it, potentially transforming our interactions and understanding of one another.

The Evolution Continues

I have explored nine design patterns for AI-enhanced features that have emerged across the product landscape—all in less than a year! Looking ahead, I am confident that some of these innovations will endure, while others may be surpassed by more refined applications. As AI-first products continue to emerge, it will be intriguing to observe the angles taken and how many of these initial features will persist. What is certain is that the evolution will continue as teams discover how to develop solutions that are well-suited to this innovative technology.

In this exploration, I have highlighted nine distinct design patterns for AI-enhanced features that have proliferated across the product landscape in under a year. As we look forward, some of these innovations will likely solidify their place while others may become obsolete, overtaken by more advanced applications. The rise of AI-first products holds the promise of introducing new perspectives and methodologies, potentially reshaping the relevance of these early features.

What remains indisputable is that the evolution of AI-enhanced features will persist as development teams refine their approaches to integrating this transformative technology. The key to future success will be the ability to adapt and innovate, ensuring that solutions are not only technologically advanced but also deeply aligned with human needs and contexts. As we navigate this exciting terrain, the principles of user-centered design will guide us toward creating functional tools and transformative experiences that redefine how we live and work. The next chapter of product design is being written today, with Generative AI undeniably playing a pivotal role in this unfolding narrative.