#LLMs are full of #UX challenges. These magical things....also frustrate us...because they're so "smart"....and so stupid at the same time. Should we be yelling at the product teams? A notable user experience element I've been thinking about when working with these is whether the LLM understands when I am frustrated - does it realize I'm twisting, turning, and morphing my prompts in a desperate attempt to squeeze out something better? Something more useful? Does it know that I'm struggling? Does it know that I don't know if the problem is that there is no good answer, insight, or better version to be had -- or that it's "user error" because I'm too stupid to speak "prompt" fluently, and don't feel like downloading a bazillion LinkedIN PDF hack sheets on how to game the system? Does it know I'm not a gamer, and having to learn cheat-codes is a tax and not a benefit? This feels like an area that could be improved - understanding intent, rephrasing of questions/prompts, and seeing patterns that suggest "this user is trying to get something out of the system that doesn't exist or cannot be improved upon." It seems LLM #UX could be improved here -- with AI -- because the training data seems to be right there in the prompts. How do you most often feel when you end up down one of these holes? Do you feel like "that's the cost of using the tech"? ~~~~~ ⬆️ Join 2k readers: tap "Visit my website” to get my latest thinking in your inbox 🚫 Don’t like LinkedIn’s algo deciding what you read? Tap the bell 🔔 under my profile banner to get all my posts ♻️ Please repost if this was useful
Brian T. O'Neill’s Post
More Relevant Posts
-
#LLMs are full of #UX challenges. These magical things....also frustrate us...because they're so "smart"....and so stupid at the same time. Should we be yelling at the product teams? A notable user experience element I've been thinking about when working with these is whether the LLM understands when I am frustrated - does it realize I'm twisting, turning, and morphing my prompts in a desperate attempt to squeeze out something better? Something more useful? Does it know that I'm struggling? Does it know that I don't know if the problem is that there is no good answer, insight, or better version to be had -- or that it's "user error" because I'm too stupid to speak "prompt" fluently, and don't feel like downloading a bazillion LinkedIN PDF hack sheets on how to game the system? Does it know I'm not a gamer, and having to learn cheat-codes is a tax and not a benefit? This feels like an area that could be improved - understanding intent, rephrasing of questions/prompts, and seeing patterns that suggest "this user is trying to get something out of the system that doesn't exist or cannot be improved upon." It seems LLM #UX could be improved here -- with AI -- because the training data seems to be right there in the prompts. How do you most often feel when you end up down one of these holes? Do you feel like "that's the cost of using the tech"? ~~~~~ ⬆️ Join 2k readers: tap "Visit my website” to get my latest thinking in your inbox 🚫 Don’t like LinkedIn’s algo deciding what you read? Tap the bell 🔔 under my profile banner to get all my posts ♻️ Please repost if this was useful
To view or add a comment, sign in
-
#design #ux #ai #tool I just tested the new Bard image generator, and it's really bad! Google really don't understand what is quality. The generated pictures are often small, pixelated, with a very dark vibe. Back to ChatGPT :-)
To view or add a comment, sign in
-
Google has just unleashed a game-changer onto the scene: ScreenAI. 🌟 Imagine an AI that not only understands user interfaces but also infographics! 😲 Want to know more about how this groundbreaking technology is set to revolutionize the future of UX? Dive into my latest article for all the juicy details! https://lnkd.in/gq_Hkxf8 🚀 #Google #ScreenAI #UX #Innovation #AIRevolution #ArtificialIntelligence #MachineLearning #ChatGPT #SoftwareDesign #Architecture #Mentorship 🌐🔍
To view or add a comment, sign in
-
I’m dyslexic. Not everyone respects or appreciates a dyslexic mind because they typically think different. Despite the advantage they give businesses by simply thinking different, our society still maintains a negative perception of this group. In the beginning of my UX journey, some would say, I was lucky to be surrounded by critical thinkers and true UX experts. I’d share who they are but obviously it would be my bias. One size don’t fit all. So I guess the big question for me was how do I know if this or that senior UXer is legit or nothing more that a shadow consumed by self interest? How do I know what is right or wrong with my understanding of UX? What voices do you follow or book do I buy (and may never actually read) to deepen your understanding? Who do I trust to help with my personal development? How do I know what I know is true? Yeah I know. Lots of questions that lead to uncertainty and in some instances self doubt because of shallow understanding. Glad I’m past that stage. It was a bomber. While I can’t answer those things for you, I’ll share my secret. I sought to understand the history of Design and its separation from art to embrace science and other disciplines for the good of humans first, then business. This history is the early beginnings of how we ended up with UX Design, a story that begins way before Don Norman. It was then, I began to understood what I was observing. Only then the wierdness of the UX community made since and I truly began to understand what UX Design actually is. “You can’t design a solution for a problem you don’t really understand.” And guess what? There’s still things I’m figuring out as I go but I have a solid foundation I’m building upon. UX is fluid, taking the shape of its container but its nature and substance remains the same. Water in a glass cup covers less area because it’s contained. If you poor that same glass of water into a plate, its reach surpasses the diameter of the glass cup previously in. This is what the industry needs from me as a designer, volume (depth in understanding) and reach (the ability ro collaborate, persuade, convince and show). This my elementary point of view of UX. If you want to grow at a healthy pace, I’d check out this guy and his UX booklist/review. It’s amazing where UX history will take you. I hope this helps someone as a UX designer or researcher.
Veteran UCD practitioner | NNg Certified UX Manager | Host of The World of UX podcast | UserInterviews Active UX Leader to Follow | Adjunct Professor | TEDx & Conference Speaker | Author | Opinions are my own
If you’re learning something that includes things that will later deserve to be unlearned, renounced, or refuted…. it’s not a “good start.” Truth is, we all spent years having someone plot out our educational paths and do NOT possess expert skill at vetting out learning solutions at certain stages of our lives. Many people unsuspectingly opt into a learning experience thinking it’s a “good start” with absolutely no idea of the trail of destruction said “good start” will produce and/or how many years this path of destruction will be operable. #ux #realuxtalk Darren Hood, MSUXD
To view or add a comment, sign in
-
In the spring of 1995, I had an idea while struggling to sleep in a hot loft. What if there was a computer search program that could summarize vast amounts of information into a concise paraphrase tailored to individual preferences? I envisioned a tool based on the "Seven Learning Styles" concept, capable of delivering answers aligned with user modalities and communication styles. This ambitious project seemed like something only a tech giant like Adobe could tackle, requiring extensive collaboration for data provision. I thought, if such technology ever materializes, it would revolutionize how people access and utilize information. #generativeai #ux #humanfactors #machinelearning
To view or add a comment, sign in
-
#design #ux #automations #ai A few years ago, I worked on automations for almost a year. This is an important topic in the tech. Tools like Zapier or Make are often used but still complicated to use, often featuring an outdated, bloated user interface. I always evaluate new solutions like Claris Connect, SureTriggers or n8n. I found a new AI tool very promising. It's called Lindy and the marketing is kind of strong, clever. AI agents are clearly the next big thing! What is your favorite tool? https://www.lindy.ai
To view or add a comment, sign in
-
Tired of getting hit with unexpected charges just for trying out different user testing platforms? We know how frustrating that can be. 🤬 That’s why we’ve designed our free trial to be truly free: ❌ no credit card needed ❌ no automatic charges ❌ no surprises Just pure, insightful user testing at your fingertips. What You Get with Your Free Trial ⭐️ 2 Free Testers: Gain insights from our pool of over 120k+ qualified testers. ⭐️ Invite 5 users: Conduct tests with your own users. ⭐️ Access all Features: Enjoy Clips, AI Insights, Transcripts, Automated Reports and more… Start your 14-day free trial today, avoid the stress of unwanted charges, and start making informed design decisions with real user feedback. (Link in first comment) #UserTesting #FreeTrial #UX #Userbrain
To view or add a comment, sign in
-
The "Wizard of Oz" method is a technique that allows you to effectively test an app concept without having to develop complex algorithms 🧙♂️ In practice, it involves an operator, the "wizard," who controls and influences the system’s behavior to make it seem like it operates autonomously and intelligently. 🔸You can successfully use this method to test interactions between the system and users. The operator can simulate various scenarios and communication methods for the system and reactions from fictional users. They can also manage the software’s behavior during interactions with real users, which helps assess the effectiveness of the communication and tailor it to the target audience’s needs. 🔸A more tangible example might be testing a chatbot system. Before creating the bot, you can set up a page that simulates it. Users interacting with it will communicate with real people pretending to be the bot rather than with artificial intelligence. This allows you to check what users ask about, how they want to use the bot, and whether they are interested in interacting with it, all without unnecessary coding. Have you heard of this method before? #WizardOfOzMethod #UXTesting #AppDesign #StartupTips #ProductValidation #DesignThinking #UserTesting #UX #UserExperience
To view or add a comment, sign in
-
uxGPT tip: Creating a form with this example. Create a account form. Display form field type, recommended field length, error messages and tooltips for each field. It should be in list format. Try it, it's fun! #ux #chatgpt
To view or add a comment, sign in