Let’s talk about design for AI (which more and more, I’m agreeing means GenAI to those outside the data space). The hype around GenAI and LLMs—particularly as it relates to dropping these in as features into a software application or product—seems to me, at this time, to largely be driven by FOMO rather than real value. In this “part 1” episode, I look at the importance of solid user experience design and outcome-oriented thinking when deploying LLMs into enterprise products. Challenges with immature AI UIs, the role of context, the constant game of understanding what accuracy means (and how much this matters), and the potential impact on human workers are also examined. Through a hypothetical scenario, I illustrate the complexities of using LLMs in practical applications, stressing the need for careful consideration of benchmarks and the acceptance of GenAI's risks. 

I also want to note that LLMs are a very immature space in terms of UI/UX design—even if the foundation models continue to mature at a rapid pace. As such, this episode is more about the questions and mindset I would be considering when integrating LLMs into enterprise software more than a suggestion of “best practices.” 

Highlights/ Skip to:

(1:15) Currently, many LLM feature  initiatives seem to mostly driven by FOMO  (2:45) UX Considerations for LLM-enhanced enterprise applications  (5:14) Challenges with LLM UIs / user interfaces (7:24) Measuring improvement in UX outcomes with LLMs (10:36) Accuracy in LLMs and its relevance in enterprise software  (11:28) Illustrating key consideration for implementing an LLM-based feature (19:00) Leadership and context in AI deployment (19:27) Determining UX benchmarks for using LLMs (20:14) The dynamic nature of LLM hallucinations and how we design for the unknown (21:16) Closing thoughts on Part 1 of designing for AI and LLMs

Quotes from Today’s Episode

“While many product teams continue to race to deploy some sort of GenAI and especially LLMs into their products—particularly this is in the tech sector for commercial software companies—the general sense I’m getting is that this is still more about FOMO than anything else.” - Brian T. O’Neill (2:07) “No matter what the technology is, a good user experience design foundation starts with not doing any harm, and hopefully going beyond usable to be delightful. And adding LLM capabilities into a solution is really no different. So, we still need to have outcome-oriented thinking on both our product and design teams when deploying LLM capabilities into a solution. This is a cornerstone of good product work.” - Brian T. O’Neill (3:03)

“So, challenges with LLM UIs and UXs, right, user interfaces and experiences, the most obvious challenge to me right now with large language model interfaces is that while we’ve given users tremendous flexibility in the form of a Google search-like interface, we’ve also in many cases, limited the UX of these interactions to a text conversation with a machine. We’re back to the CLI in some ways.” - Brian T. O’Neill (5:14) “Before and after we insert an LLM into a user’s workflow, we need to know what an improvement in their life or work actually means.”- Brian T. O’Neill (7:24) "If it would take the machine a few seconds to process a result versus what might take a day for a worker, what’s the role and purpose of that worker going forward? I think these are all considerations that need to be made, particularly if you’re concerned about adoption, which a lot of data product leaders are." - Brian T. O’Neill (10:17)

“So, there’s no right or wrong answer here. These are all range questions, and they’re leadership questions, and context really matters. They are important to ask, particularly when we have this risk of reacting to incorrect information that looks plausible and believable because of how these LLMs tend to respond to us with a positive sheen much of the time.” - Brian T. O’Neill (19:00)

Links

View Part 1 of my article on UI/UX design considerations for LLMs in enterprise applications:  https://designingforanalytics.com/resources/ui-ux-design-for-enterprise-llms-use-cases-and-considerations-for-data-and-product-leaders-in-2024-part-1/

Before the onset of COVID-19, the healthcare community was already moving to meet the challenges of a growing global population. By collecting record amounts of clinical data electronically and making significant progress on neural network-based AI approaches, the industry now has the potential to build powerful predictive analytics systems. The focus will accelerate the shift from a one-size-fits-all approach to individualized medicine. But several questions remain. What are the plausible outcomes for the world of predictive analytics in both the short and long term? What does the care pathway look like if everything is predicted? And with patient populations and healthcare needs increasing exponentially, how can the industry deliver care in a sustainable and cost-effective way? This comprehensive report, written by Jaquie Finn and Dr. Gavin Troughton with Cambridge Consultants, explores the possibilities. You’ll learn: How predictive analytics plays a part across all stages of the care pathway The foundational enablers for predictive analytics How healthcare economics figure into the equation Predictive analytics and today’s healthcare system The future of predictive analytics in healthcare

Presents a useful guide for applications of SEM whilst systematically demonstrating various SEM models using Mplus Focusing on the conceptual and practical aspects of Structural Equation Modeling (SEM), this book demonstrates basic concepts and examples of various SEM models, along with updates on many advanced methods, including confirmatory factor analysis (CFA) with categorical items, bifactor model, Bayesian CFA model, item response theory (IRT) model, graded response model (GRM), multiple imputation (MI) of missing values, plausible values of latent variables, moderated mediation model, Bayesian SEM, latent growth modeling (LGM) with individually varying times of observations, dynamic structural equation modeling (DSEM), residual dynamic structural equation modeling (RDSEM), testing measurement invariance of instrument with categorical variables, longitudinal latent class analysis (LLCA), latent transition analysis (LTA), growth mixture modeling (GMM) with covariates and distal outcome, manual implementation of the BCH method and the three-step method for mixture modeling, Monte Carlo simulation power analysis for various SEM models, and estimate sample size for latent class analysis (LCA) model. The statistical modeling program Mplus Version 8.2 is featured with all models updated. It provides researchers with a flexible tool that allows them to analyze data with an easy-to-use interface and graphical displays of data and analysis results. Intended as both a teaching resource and a reference guide, and written in non-mathematical terms, Structural Equation Modeling: Applications Using Mplus, 2nd edition provides step-by-step instructions of model specification, estimation, evaluation, and modification. Chapters cover: Confirmatory Factor Analysis (CFA); Structural Equation Models (SEM); SEM for Longitudinal Data; Multi-Group Models; Mixture Models; and Power Analysis and Sample Size Estimate for SEM. Presents a useful reference guide for applications of SEM while systematically demonstrating various advanced SEM models Discusses and demonstrates various SEM models using both cross-sectional and longitudinal data with both continuous and categorical outcomes Provides step-by-step instructions of model specification and estimation, as well as detailed interpretation of Mplus results using real data sets Introduces different methods for sample size estimate and statistical power analysis for SEM Structural Equation Modeling is an excellent book for researchers and graduate students of SEM who want to understand the theory and learn how to build their own SEM models using M plus.

Have you ever attended a conference? Did you know that analysts over-index towards introversion? Have you ever struggled to figure out how to start a conversation over a cold pastry and a cup of tepid coffee at a conference breakfast? IS there actually a point in developing and executing a strategy when it comes to attending a conference? Is it annoying to listen to people who speak pretty regularly at conferences pontificate about speaking at conferences? Some of these questions are answered on this episode! For complete show notes, including links to items mentioned in this episode and a transcript of the show, visit the show page. We made this up, but it seems plausible.

Algorithms are pervasive in our society and make thousands of automated decisions on our behalf every day. The possibility of digital discrimination is a very real threat, and it is very plausible for discrimination to occur accidentally (i.e. outside the intent of the system designers and programmers). Christian Sandvig joins us in this episode to talk about his work and the concept of auditing algorithms. Christian Sandvig (@niftyc) has a PhD in communications from Stanford and is currently an Associate Professor of Communication Studies and Information at the University of Michigan. His research studies the predictable and unpredictable effects that algorithms have on culture. His work exploring the topic of auditing algorithms has framed the conversation of how and why we might want to have oversight on the way algorithms effect our lives. His writing appears in numerous publications including The Social Media Collective, The Huffington Post, and Wired. One of his papers we discussed in depth on this episode was Auditing Algorithms: Research Methods for Detecting Discrimination on Internet Platforms, which is well worth a read.

Beginning Database Design, Second Edition provides short, easy-to-read explanations of how to get database design right the first time. This book offers numerous examples to help you avoid the many pitfalls that entrap new and not-so-new database designers. Through the help of use cases and class diagrams modeled in the UML, you'll learn to discover and represent the details and scope of any design problem you choose to attack. Database design is not an exact science. Many are surprised to find that problems with their databases are caused by poor design rather than by difficulties in using the database management software. Beginning Database Design, Second Edition helps you ask and answer important questions about your data so you can understand the problem you are trying to solve and create a pragmatic design capturing the essentials while leaving the door open for refinements and extension at a later stage. Solid database design principles and examples help demonstrate the consequences of simplifications and pragmatic decisions. The rationale is to try to keep a design simple, but allow room for development as situations change or resources permit. Provides solid design principles by which to avoid pitfalls and support changing needs Includes numerous examples of good and bad design decisions and their consequences Shows a modern method for documenting design using the Unified Modeling Language What you'll learn Avoid the most common pitfalls in database design. Create clear use cases from project requirements. Design a data model to support the use cases. Apply generalization and specialization appropriately. Secure future flexibility through a normalized design. Ensure integrity through relationships, keys, and constraints. Successfully implement your data model as a relational schema. Who this book is for Beginning Database Design, Second Edition is aimed at desktop power users, developers, database administrators, and others who are charged with caring for data and storing it in ways that preserve its meaning and integrity. Desktop users will appreciate the coverage of Excel as a plausible "database" for research systems and lab environments. Developers and database designers will find insight from the clear discussions of design approaches and their pitfalls and benefits. All readers will benefit from learning a modern notation for documenting designs that is based upon the widely used and accepted Universal Modeling Language.