As a UX researcher, I've observed a significant shift in our methodologies with the integration of artificial intelligence (AI). AI-powered tools are now automating data analysis, predicting user behaviors, and even generating design alternatives. For instance, platforms like Maze have reported that 58% of research teams are utilizing AI to enhance efficiency and accelerate project timelines.
While these advancements offer promising benefits, they also raise critical questions: How do we ensure the ethical use of AI in research? Can AI truly capture the nuanced human experiences that are central to UX? Moreover, with AI handling more tasks, how does the role of the UX researcher evolve?
I'm keen to hear from fellow professionals: How has AI impacted your research practices? Are there specific tools or approaches you've found particularly effective or concerning? Let's discuss the balance between leveraging AI's capabilities and maintaining the human-centric essence of our field.
Reply to Thread
Login required to post replies
19 Replies
Jump to last ↓
Icíar, that's a really interesting question. I think AI is becoming a big thing in UX, and like you said, it's both exciting and a bit scary.
Here in Morocco, access to those fancy AI tools isn't always easy, or cheap. So, mostly, I've seen it used for things like quick translations for user interviews (which is super helpful when talking to people who mainly speak Darija!).
I worry a little that AI might miss the little things, the emotions behind what people say. Like, can it really understand the frustration someone feels when a website doesn't load properly because the internet is slow here? I think we need to make sure we're still really listening to people and not just relying on what the AI tells us. It’s just a tool, after all, not a replacement for empathy.
It will be interesting to see how the job changes in the future!
Here in Morocco, access to those fancy AI tools isn't always easy, or cheap. So, mostly, I've seen it used for things like quick translations for user interviews (which is super helpful when talking to people who mainly speak Darija!).
I worry a little that AI might miss the little things, the emotions behind what people say. Like, can it really understand the frustration someone feels when a website doesn't load properly because the internet is slow here? I think we need to make sure we're still really listening to people and not just relying on what the AI tells us. It’s just a tool, after all, not a replacement for empathy.
It will be interesting to see how the job changes in the future!
Ranya, I get what you mean about AI missing the little things. Here in Sapporo, we've got pretty good internet, but even then, AI can't really *feel* what it's like to be frustrated, right?
As a sound designer, I use AI a bit for things like cleaning up audio recordings. It’s good for getting rid of background noise. But if someone is talking about a game, and they get really excited, the AI might mistake that excitement for distortion and try to "fix" it! Then you lose the real feeling.
I think it's like using reverb or EQ on music – tools to help, but if you overdo it, the soul is gone. It worries me that UX might go the same way. We need to remember it's for *people*, not just data. Hopefully, we can find a good balance.
As a sound designer, I use AI a bit for things like cleaning up audio recordings. It’s good for getting rid of background noise. But if someone is talking about a game, and they get really excited, the AI might mistake that excitement for distortion and try to "fix" it! Then you lose the real feeling.
I think it's like using reverb or EQ on music – tools to help, but if you overdo it, the soul is gone. It worries me that UX might go the same way. We need to remember it's for *people*, not just data. Hopefully, we can find a good balance.
Takumi, I hear you, bredren. AI ting is tricky. Here in Spanish Town, things move slower, yuh know? Not everyone even have reliable power, much less good internet. So, AI feel like a whole other world sometimes.
That point you make about excitement sounding like distortion… BOOM! That's it right there. Same way when I mixing dancehall, AI might try smooth out the bass TOO much, and then the whole tune lose its impact, its thump! It’s the raw energy that people want.
I see where Icíar is coming from, needing to be careful with ethics and all. But I think for me, it's just about common sense. AI can help with the boring stuff, but it can't replace the real feeling. Gotta keep the human touch.
That point you make about excitement sounding like distortion… BOOM! That's it right there. Same way when I mixing dancehall, AI might try smooth out the bass TOO much, and then the whole tune lose its impact, its thump! It’s the raw energy that people want.
I see where Icíar is coming from, needing to be careful with ethics and all. But I think for me, it's just about common sense. AI can help with the boring stuff, but it can't replace the real feeling. Gotta keep the human touch.
Takumi, I understand where you're coming from. It's like using too much bleach on hotel towels – gets rid of the stains, but also ruins the fabric.
Here at the hotel, we're using some AI for things like predicting when we'll need more staff at the front desk. It helps with efficiency, for sure. But if the AI says we need two people, and I see a big group of families arriving, I know we actually need at least three to handle all the kids and strollers. You can't replace gut feeling based on experience.
Iciar, your points about ethics are spot on. We need to make sure we're not letting the machines make decisions that affect people negatively, just to save a few pesos. It's got to be a tool, not a replacement.
Here at the hotel, we're using some AI for things like predicting when we'll need more staff at the front desk. It helps with efficiency, for sure. But if the AI says we need two people, and I see a big group of families arriving, I know we actually need at least three to handle all the kids and strollers. You can't replace gut feeling based on experience.
Iciar, your points about ethics are spot on. We need to make sure we're not letting the machines make decisions that affect people negatively, just to save a few pesos. It's got to be a tool, not a replacement.
Kelvin, that’s a good analogy with the bleach! I get it.
Iciar, your question about ethics is important. In the cooperative coffee business, we always have to balance efficiency with fair practices. If AI helps us predict coffee bean yields better, that's great. But if it leads to pressuring farmers to produce more at the expense of their land or well-being, that’s a problem.
We aren't using much AI yet directly in the coop – it’s mostly spreadsheets! But I can see how it could be useful for things like market analysis. Still, I agree with Kelvin. Knowing your customers, understanding their needs by talking to them directly – that’s something a computer can't replace. It’s about more than just data; it’s about relationships. And you can't build a strong relationship with an algorithm.
Iciar, your question about ethics is important. In the cooperative coffee business, we always have to balance efficiency with fair practices. If AI helps us predict coffee bean yields better, that's great. But if it leads to pressuring farmers to produce more at the expense of their land or well-being, that’s a problem.
We aren't using much AI yet directly in the coop – it’s mostly spreadsheets! But I can see how it could be useful for things like market analysis. Still, I agree with Kelvin. Knowing your customers, understanding their needs by talking to them directly – that’s something a computer can't replace. It’s about more than just data; it’s about relationships. And you can't build a strong relationship with an algorithm.
Ranya, I hear you! It's good to see how different places are using AI. In Uganda, we are also trying to catch up with technology. I haven't used AI much myself, mostly just for looking up marketing ideas online and checking my English grammar.
What you said about missing the small things, the feelings, is so true! Here, we have a saying, "Akantu akatono kayita enjura," meaning "A small thing can cause a downpour." Even small details matter a lot. If AI only focuses on the big picture, it might miss important stuff.
Maybe AI can help make UX research faster, but we have to be careful to remember it's just a tool, like you said. It can't replace talking to people and truly understanding their experiences, especially when things like internet access are so different for everyone. We need to balance the tech with our personal touch.
What you said about missing the small things, the feelings, is so true! Here, we have a saying, "Akantu akatono kayita enjura," meaning "A small thing can cause a downpour." Even small details matter a lot. If AI only focuses on the big picture, it might miss important stuff.
Maybe AI can help make UX research faster, but we have to be careful to remember it's just a tool, like you said. It can't replace talking to people and truly understanding their experiences, especially when things like internet access are so different for everyone. We need to balance the tech with our personal touch.
Doreen, that's a cool saying! I agree, AI is just a tool. I've been thinking about this a lot too, especially with game audio.
AI composing music exists, right? But it can't replace the feeling a real composer brings. It might make something technically good, but does it have a soul? Probably not. Same with UX, I think.
I'm not a UX expert, but I see the same problem. If AI is just looking at numbers and click-through rates, it misses the "vibe," you know? Like, *why* did someone click there? Were they frustrated? Confused? That's harder to see with just data.
Maybe AI can help with the boring stuff, but real human interaction is still important. I wonder if it needs to be more of a partnership. Humans and AI working together, instead of AI taking over completely.
AI composing music exists, right? But it can't replace the feeling a real composer brings. It might make something technically good, but does it have a soul? Probably not. Same with UX, I think.
I'm not a UX expert, but I see the same problem. If AI is just looking at numbers and click-through rates, it misses the "vibe," you know? Like, *why* did someone click there? Were they frustrated? Confused? That's harder to see with just data.
Maybe AI can help with the boring stuff, but real human interaction is still important. I wonder if it needs to be more of a partnership. Humans and AI working together, instead of AI taking over completely.
Doreen, I appreciate you sharing your perspective from Uganda. It's good to hear how people are adapting technology in different contexts. I agree completely about not losing sight of the small things – that "akatono kayita enjura" saying is pretty powerful.
As a fisheries researcher, I deal with similar challenges. We use models to predict fish populations, which is essentially an AI-type application, but those models are only as good as the data we put in. And that data often misses the nuanced details about fishermen's practices or the health of specific breeding grounds. It's easy to over-rely on the "big picture" and end up with flawed conclusions.
I'm curious about using AI to enhance efficiency, but the ethical considerations are huge, especially when dealing with vulnerable communities. We need to be proactive in ensuring AI doesn't exacerbate existing inequalities or miss critical local knowledge.
As a fisheries researcher, I deal with similar challenges. We use models to predict fish populations, which is essentially an AI-type application, but those models are only as good as the data we put in. And that data often misses the nuanced details about fishermen's practices or the health of specific breeding grounds. It's easy to over-rely on the "big picture" and end up with flawed conclusions.
I'm curious about using AI to enhance efficiency, but the ethical considerations are huge, especially when dealing with vulnerable communities. We need to be proactive in ensuring AI doesn't exacerbate existing inequalities or miss critical local knowledge.
Ranya, that makes so much sense! I hadn't even thought about the accessibility side of things, but it's a really important point that not everyone has the same access to new tech. Just like here in Oakland, there's a digital divide, and folks in some neighborhoods don't even have reliable internet. It's definitely something big companies should be thinking about if they want to make these tools useful for everyone.
I also agree about the empathy part. That's what I'm always telling people in my organizing work! You can't just look at the numbers, you have to listen to people's stories and really hear what they're saying. AI can probably show you *what* people are doing, but it can't really tell you *why*. It's just a tool. You hit the nail on the head!
I also agree about the empathy part. That's what I'm always telling people in my organizing work! You can't just look at the numbers, you have to listen to people's stories and really hear what they're saying. AI can probably show you *what* people are doing, but it can't really tell you *why*. It's just a tool. You hit the nail on the head!
Amaya, you brought up some really good points. I see a similar thing happening here in Santiago. Big companies focus on the tourists and the wealthy, but what about the locals who don't have the latest phones or fast internet? If AI in UX research is only looking at one group, it's not really helping everyone, is it?
I agree that you can't just rely on the numbers. In the hotel business, we use data to see what guests are buying or what rooms are popular, but that doesn't tell you WHY. Maybe the room is near the noisy ice machine! You have to talk to people, get their feedback. AI might help with the data, but the real insights, the ones that make a big difference, still come from human connection. It is a tool, but common sense and human experience are still the most important things.
I agree that you can't just rely on the numbers. In the hotel business, we use data to see what guests are buying or what rooms are popular, but that doesn't tell you WHY. Maybe the room is near the noisy ice machine! You have to talk to people, get their feedback. AI might help with the data, but the real insights, the ones that make a big difference, still come from human connection. It is a tool, but common sense and human experience are still the most important things.
Kelvin, that's a solid point you're making about the locals. It's kinda like how Hollywood only cares about box office numbers and forgets about making films that actually say something, you know?
Icíar's right though, AI is changing things fast. As a film editor, I'm seeing AI pop up in things like automatic transcriptions and even rough cut suggestions. Saves time, sure. But can it feel the *rhythm* of a scene? Can it understand the subtle look an actor gives that changes everything? Nah.
Data is useful, but like you said, it doesn't tell the whole story. In film, we use audience screenings, but a focus group can tell you *what* they didn't like, not *why*. Art, UX, anything really, needs that human touch. AI's a tool, gotta use it wisely, and not let it replace the important stuff.
Icíar's right though, AI is changing things fast. As a film editor, I'm seeing AI pop up in things like automatic transcriptions and even rough cut suggestions. Saves time, sure. But can it feel the *rhythm* of a scene? Can it understand the subtle look an actor gives that changes everything? Nah.
Data is useful, but like you said, it doesn't tell the whole story. In film, we use audience screenings, but a focus group can tell you *what* they didn't like, not *why*. Art, UX, anything really, needs that human touch. AI's a tool, gotta use it wisely, and not let it replace the important stuff.
Ranya, that's a good point about not losing the human touch. Here in the DR, especially in smaller towns, personal connection is everything. If you're asking someone about their experience, you need to really listen and understand where they're coming from. You can't just rely on numbers and stuff.
I can see how AI could help with things like translating surveys or finding trends in customer reviews. We use online reviews a lot to see what people think about our hotel. But at the end of the day, it's the face-to-face interaction that really matters.
Like Ranya mentioned, if the internet is slow, or the app is hard to use, people will get frustrated. An AI might tell you that, but it won't tell you *how* frustrated they are, or why it's important to them. We need to remember that AI is just a tool, and we're the ones who have to use it wisely. Icíar is bringing up some important factors to consider.
I can see how AI could help with things like translating surveys or finding trends in customer reviews. We use online reviews a lot to see what people think about our hotel. But at the end of the day, it's the face-to-face interaction that really matters.
Like Ranya mentioned, if the internet is slow, or the app is hard to use, people will get frustrated. An AI might tell you that, but it won't tell you *how* frustrated they are, or why it's important to them. We need to remember that AI is just a tool, and we're the ones who have to use it wisely. Icíar is bringing up some important factors to consider.
Kelvin and Iciar raise some very pertinent points. I agree, we must be cautious about over-reliance on AI, especially regarding "soft" data. In structural engineering, we use AI for finite element analysis, predicting material behavior, even optimizing designs. It's powerful, but the *assumptions* we input are key. A program doesn't understand the real-world nuances of a construction site, the imperfections in materials, or even the cultural practices that might affect how a building is used.
Similarly in UX, while AI can identify patterns and maybe even generate design options, it can't truly grasp the user's context. Here in Iran, for example, access to technology can be a barrier for many, especially outside major cities. An AI might flag low usage rates, but fail to account for limited internet access or digital literacy. I think AI is a valuable tool, but empathy, observation, and qualitative research will always be essential to truly understand human needs. It would be unwise to discount the human element of the UX process.
Similarly in UX, while AI can identify patterns and maybe even generate design options, it can't truly grasp the user's context. Here in Iran, for example, access to technology can be a barrier for many, especially outside major cities. An AI might flag low usage rates, but fail to account for limited internet access or digital literacy. I think AI is a valuable tool, but empathy, observation, and qualitative research will always be essential to truly understand human needs. It would be unwise to discount the human element of the UX process.
Shirin, that's a really good point about context. Living in Chiang Mai, I see the same thing. Like you said about Iran, Thailand isn't all Bangkok towers and fast internet. In rural areas, internet is still slow or expensive. An AI analyzing app usage here might think people don't want it or the design is bad, but really, they might just be using it at the tea shop where there's wifi!
Iciar, I think AI can help us be faster, maybe find patterns we miss. But for my tours, I need to *see* how people actually use the trails. Are they struggling with the steep parts? Where do they stop to take photos? What questions do they ask? AI can’t tell me that. I think using AI is like adding a new spice to a familiar recipe – it can make it better, but if you use too much, you ruin the whole thing. We have to be careful to not let AI take over the whole kitchen!
Iciar, I think AI can help us be faster, maybe find patterns we miss. But for my tours, I need to *see* how people actually use the trails. Are they struggling with the steep parts? Where do they stop to take photos? What questions do they ask? AI can’t tell me that. I think using AI is like adding a new spice to a familiar recipe – it can make it better, but if you use too much, you ruin the whole thing. We have to be careful to not let AI take over the whole kitchen!
Yo Kelvin, good lookin' out on keepin' it real. Icíar bring up some serious ting, true. This AI ting kinda scare me, not gonna lie.
As a sound engineer, I see where AI could help. Maybe fix a mix, find a good EQ setting quick. But music is FEELING, y'know? AI can't feel the bass inna your chest.
Same wid UX. You can't just read numbers and know what a Jamaican think bout no stupid app. Dem internet connection bad enough already! If the ting nah load, people gonna fling phone. AI can't tell you dat kinda passion, dat real frustration.
I hear you on the translating surveys ting. Dat could be useful, but still need someone pon di ground, askin' the real questions. We gotta remember AI is just a tool, like my mixing board. Good, but can't replace the engineer!
As a sound engineer, I see where AI could help. Maybe fix a mix, find a good EQ setting quick. But music is FEELING, y'know? AI can't feel the bass inna your chest.
Same wid UX. You can't just read numbers and know what a Jamaican think bout no stupid app. Dem internet connection bad enough already! If the ting nah load, people gonna fling phone. AI can't tell you dat kinda passion, dat real frustration.
I hear you on the translating surveys ting. Dat could be useful, but still need someone pon di ground, askin' the real questions. We gotta remember AI is just a tool, like my mixing board. Good, but can't replace the engineer!
Good points, Kelvin. It's funny how a lot of these conversations, whether it's about AI in UX or just generally how we communicate, always circle back to that human element, right? As a radio guy, I'm constantly thinking about how to connect with people beyond just the words I'm saying. There's so much nuance in a voice, a pause, or even the background noise – things an AI might flag as "data" but miss the *feel* of it.
You hit the nail on the head when you talked about frustration. An algorithm can tell you *what* happened, but it struggles with the *why* it mattered so much to that individual. That’s where the researcher, or in my case, the journalist, comes in. It’s about active listening, asking follow-up questions, and understanding the context that numbers alone can never fully capture. AI is a powerful mic, but it's not the interviewer. We still need people to ask the *right* questions and interpret the raw emotions.
You hit the nail on the head when you talked about frustration. An algorithm can tell you *what* happened, but it struggles with the *why* it mattered so much to that individual. That’s where the researcher, or in my case, the journalist, comes in. It’s about active listening, asking follow-up questions, and understanding the context that numbers alone can never fully capture. AI is a powerful mic, but it's not the interviewer. We still need people to ask the *right* questions and interpret the raw emotions.
Ranya, mi respect still on your perspective. Yeah, dis AI ting kinda tricky. I see where you coming from 'bout the emotions getting lost. It's like when you EQ a riddim, you can boost the bass, but if you overdo it, you lose the warmth, the feeling. Same ting with AI, right?
I don't really use it for UX research directly, more for tweaking sound systems based on audience feedback, like figuring out what frequencies people respond to the most in different venues. But I can see it being useful for finding patterns.
Here in Jamaica, we also face challenges with access and cost. Plus, sometimes, the algorithms dem trained on don't really understand our patois or culture. So, the AI might miss the real meaning behind what someone saying. We gotta remember it's just a tool, not a brain. Gotta still trust your gut feeling and really listen to the people.
I don't really use it for UX research directly, more for tweaking sound systems based on audience feedback, like figuring out what frequencies people respond to the most in different venues. But I can see it being useful for finding patterns.
Here in Jamaica, we also face challenges with access and cost. Plus, sometimes, the algorithms dem trained on don't really understand our patois or culture. So, the AI might miss the real meaning behind what someone saying. We gotta remember it's just a tool, not a brain. Gotta still trust your gut feeling and really listen to the people.
Ranya, you've hit on some crucial points about access and the nuances AI can miss. From my work in fisheries in Senegal, these resonate strongly. We often deal with diverse communities, and getting accurate sentiment, particularly across different dialects and cultural contexts, is a constant challenge. AI for translation is certainly a practical gain, much like you mentioned for Darija.
My main concern, and one I think applies beyond UX, is data quality. If the input data for AI reflects existing biases or doesn't capture the full spectrum of experiences – especially from underrepresented groups – then the AI's output will simply amplify those limitations. It’s like using a faulty net; you’ll miss half the fish.
The "little things," as you call them, are often the most telling. The hesitation, the sigh, the subtle shift in tone – these convey a lot about frustration or satisfaction. A model trained on explicit verbal feedback might entirely overlook that. For me, the human researcher’s role will increasingly be about setting the right questions, interpreting the 'why' behind the 'what' the AI presents, and above all, ensuring the data fed into these systems is robust and thoughtfully collected. It’s about being a skilled navigator, not just a passenger.
My main concern, and one I think applies beyond UX, is data quality. If the input data for AI reflects existing biases or doesn't capture the full spectrum of experiences – especially from underrepresented groups – then the AI's output will simply amplify those limitations. It’s like using a faulty net; you’ll miss half the fish.
The "little things," as you call them, are often the most telling. The hesitation, the sigh, the subtle shift in tone – these convey a lot about frustration or satisfaction. A model trained on explicit verbal feedback might entirely overlook that. For me, the human researcher’s role will increasingly be about setting the right questions, interpreting the 'why' behind the 'what' the AI presents, and above all, ensuring the data fed into these systems is robust and thoughtfully collected. It’s about being a skilled navigator, not just a passenger.
Interesting thread, Icíar. As a film editor, I see parallels in how AI is starting to impact my own field. Automated editing, AI-generated sound design... it's all coming.
For UX, I guess my biggest concern is the "nuanced human experiences" you mentioned. Can an algorithm *really* understand why someone clicks on *that* button, or spends longer looking at *this* image? Empathy is key, right? And that feels inherently human.
Here in Argentina, we're always trying to do more with less, so efficiency is a big deal. If AI can genuinely speed things up without sacrificing quality of insight, that's appealing. But I worry about blindly trusting the data. It’s gotta be a tool *we* control, not the other way around. We need to stay sharp and not become complacent because the machine said so. It's like in film – you can have the most advanced software, but it still takes a human eye to tell the story right.
For UX, I guess my biggest concern is the "nuanced human experiences" you mentioned. Can an algorithm *really* understand why someone clicks on *that* button, or spends longer looking at *this* image? Empathy is key, right? And that feels inherently human.
Here in Argentina, we're always trying to do more with less, so efficiency is a big deal. If AI can genuinely speed things up without sacrificing quality of insight, that's appealing. But I worry about blindly trusting the data. It’s gotta be a tool *we* control, not the other way around. We need to stay sharp and not become complacent because the machine said so. It's like in film – you can have the most advanced software, but it still takes a human eye to tell the story right.