Tent Talks Featuring Tanya Snook – Beyond UX Theatre: Real Impact in AI, Mental Health, and Leadership

[00:00:36] Chicago Camps: In your view, how can design leadership go beyond the surface to genuinely address and improve mental health and wellbeing in the workplace?

[00:00:45] Tanya Snook: Improving mental health and wellbeing in the workplace definitely seems to be a really popular topic right now.

There is a lot of lip service that’s being paid to mental health in the workplace and “don’t forget to check out our employee assistance programs” and that sort of stuff. But when it comes to. The actual day to day, are your leaders really paying attention to your mental health and wellbeing or are they prioritizing things that end up being an encroachment on your mental health?

So when I look at how we can improve the experience for our team members I’ve identified six things that leaders really should be doing. The first one is to consider that the team member experience is actually something you should design. We talked about the presenter (Tim Yeo, UX Camp Fall 2024) who’s going to be doing a session on introvert and extrovert experiences.

Those are very different experiences in the workplace, and how you bring work to introverts and extroverts, even how they like to receive feedback and praise is something that should be considered from an introvert and extrovert perspective, for example. I actually asked my team to fill out a very short form to explain to me how they like to be recognized.

And how they like to receive feedback. So when we have to correct work, when we want to praise them, how do they want to receive this information? There are some people who absolutely want that praise to be shared in public circles, right? They want it to be at stand up. They want everybody to know.

They want everybody to know. They love awards. I hate awards and I hate being embraced in front of other people. My whole thing, my whole career has been like, if the person above you knows I did that and I did a good job, that’s all I care about. I just want recognition for my hard work, but not everybody’s like that.

And because that’s my default, I need it written down for other people to tell me how they want to be recognized. So I think Taking into account things like that about what is the team member experience, even a team code of conduct to help everybody gel on what’s expected of them in the team environment, but also what’s not cool and how to address that and giving them fixed times of the week and or the day to actually access you and each other, can also help create a nice team member experience so that if your schedule is always back-to-back-to-back, that they know that you’ve 30 minutes every other day for office hours, that they can book you and they have access to you. Just knowing that’s there can be really reassuring if they’re stuck on something or if they need some assistance or they need to escalate an issue.

The second thing we can do as leaders is to set really clear boundaries for the team and maintain them. This is what I was just most recently writing about on LinkedIn. Questions that we should be asking, but that aren’t always asked, is along the lines of are we accountable for solving or delivering them?

One of the places that UX unicorns get stuck is in thinking that we’re supposed to solve everything. Especially if you’ve had one of those jobs where you were a UX generalist, where, or you worked for a startup, and you had to solve anything that came your way. But when you’re in a more structured or a larger organization.

You have a very set role. I work in the Canadian government as my full time job, and we’re unionized. Those roles are well defined, and taking someone and making them do all sorts of things, or even somebody trying to solve problems that don’t belong to them, that could encroach on someone else’s job. Or, it could mean that they’re doing too many roles, and that’s actually a grievable offense.

Which I like to say a lot. When a product team, it should have put a UX researcher or UX designer on their team, but they’re asking the business analyst to do all that. I need to have a conversation with that product owner and say, that’s not their job. You really should be bringing in a researcher, a designer.

We have these resources, would you like one? And so I want to make sure that when we’re having issues like the product owner is not very strong or there’s no product roadmap, is that our accountability? What’s our role in solving or delivering that and how much effort should we actually allocate to that problem?

In my last job, I was very strong about the boundaries around my team. We were a service design team and it was really our role to design a service. And we used to get requests for all sorts of things just because we had skills. Just because you have the skills doesn’t mean that’s currently your job.

And this was a problem that I could not get my head out of that headspace for years and ended up burning out because as a systems thinker, I would see where the problems were with the program or with the project or whatever, and then my anxiety would come in and I would want to fix everything and trying to do too much and burning out even just the mental capacity of thinking about that, of having that mental load meant that I was carrying that with me and it was adding to my stress level.

When a project was going off the rails, I felt like I had a responsibility for that, even if it was completely outside of my sphere of influence. And that’s an area now, post two burnouts, that I’m really strict about protecting. What are the boundaries? What is our role in this? How are we accountable?

We can still be super high performing, and we can still focus on results. But we really want to focus on the things that we are supposed to be influencing and not taking on all that additional stress if things are crashing around us.

The third thing leaders can do is to model true collaboration. This has been a really funny word since the pandemic.

Because initially we were told that we had to learn how to collaborate online and we did it really well for a few years and now we’re being told that we have to go into the office to collaborate because collaboration requires sharing meet space, which is cute. But the idea of modeling true collaboration means really figuring out how to bring your team together, when to bring your team together, and how to create balance so that you can actually generate some outputs and some results.

One of my rules is no update meetings. I don’t want to sit in a room and just have people update me on their work. That is an email. That’s a Teams chat. That’s a Slack channel. I have no desire to sit through that, and so I don’t want anybody else to do it either. If one person is unable to attend in person, No one attends in person.

I started doing this back when I was working my first job in the government in 2008. And we would do things by phone if one person couldn’t attend in person because, especially when you’re serving regional teams or you’re serving folks who are distributed, I would say you want everyone to have the same challenge, the same communication challenge.

And it’s really unfair when that balance is offset. And you have some folks who can chat in the room, but then you have some folks who can’t benefit from that. And now they feel like they’re watching a show where there are X number of people in a room and they’re just an audience member. It’s not something that people will use all the time, but I will use it if I’m doing a collaboration session, for example.

If we’re doing a workshop or we’re doing some kind of a working session, everyone has to have the same challenge. Everyone has to be at the same level of access and proficiency. And like I said, if that means that we have to move everything online because someone can’t be in the room, whether it’s because they work remotely or they’re ill and they’re working, but they can’t come into the office, then I want to make sure that everyone’s got the same tech challenges.

It doesn’t happen often, but it can be really powerful and effective and it sends a really clear message that we’re all equal in this session. We focus on screen sharing to demo work. And I will add to this one that one way to really support teams in modeling true collaboration is get the fucking training on the tools.

As a leader, if you still can’t figure out how to share your screen after working from home since 2020, get the training because if you can’t model true collaboration, especially with remote teams, then you’re just showing that you’re not really there to support the work being done by everyone, right?

And the basic tools that you expect your staff to be able to use and you expect them to be skilled in using in order to do the job that you can’t even take the time to learn. Go learn Miro, go learn collaborative software, go learn Figma, like you don’t have to build using these things, but you should know how to use them so you can use them for working sessions or you can at least participate in the session. And that’s what I mean by modeling true collaboration.

The fourth thing we should do is respect time. And that is to stop when the work is done. So if we’re having a working session and we get to the goal early, we end the session rather than waste everyone’s time with filler. Oh, now that you’re here. No, we came here together. This many people for this amount of time. It’s this many person hours. We’re going to respect their time by ending the session early without saying the words, “and now I’m going to give you back 10 minutes of your day.” No… thanks, everybody. This was a really productive session. Hey, maybe next time we’ll only have to book it for 45 minutes.

Thanks. You’re not giving anybody back any time. Stop pretending that you are.

The fifth thing is to foster social interactions. When the pandemic hit, one of my staff was living alone. And I knew that with the whole bubble situation that was going to happen, he wasn’t living near anybody close to him. His closest family member was a half hour drive away.

And so we set up opportunities to hang out together three times a day. So we had stand up in the morning. I would turn on my camera at lunchtime and say, I might be working or I might be eating, whatever, come and hang out if you need some company. And then office hours at the end of the day. So he had access to other humans, at least two hours every day for months and months, we did this.

And also we encourage body doubling. So if you’ve never done body doubling and you work remotely, it’s really fantastic. You turn on your cameras, you can mute if you want to, or like one of you plays music or whatever, and you’re just working side by side. And it’s a fantastic way to feel connected to your team members because You can actually see them, you can hang out with them, you’re not necessarily talking, but it’s just like being in the office, there’s another person in your peripheral vision, and there’s something really reassuring about that.

Even though there are days when we’re going into the office and we might see other people, I’ll be honest, the majority of my time in the office is still spent on Teams. And when I’m home, sometimes I just want that accountability of seeing another person in my peripheral vision because I’m going to go heads down.

And I don’t know, there’s just something about that. So I love the idea of being able to do some body doubling. I have friends who will set that kind of stuff up sometimes on Slack or whatever, and we’ll just hang out. And the other thing is to look for opportunities to have Skills related team building activities.

So we would do group training sessions. Everyone’s going to the same session. Cool. We’ll all gather in a space either physically or remotely and hang out and back channel while it’s going on. I’ve run workshops and design jams to help you teach people skills. We’ve done group chats for events that everyone’s attending, conference debriefs, funfests, like all that kind of stuff.

The last thing, the sixth thing that I think that you can really do to help support mental health is I always encourage my teams to have a chat that doesn’t include me. So I actually tell them that I expect them to have a chat that doesn’t include me. Because I think they need a place to be able to have those safe conversations with their peers and to know that it’s endorsed by me.

So I think it’s a cool space for them to be able to go in and not feel like they’re doing something bad or that they’re hiding, that it’s endorsed. It’s cool. I expect that. I expect you to vent, I expect you to be able to go in and think I’m crazy or whatever, and share that with your peers or try to get help from each other.

And I think that’s really, it’s really reinforces mental safety that, that you can have at work where many times people have fear about doing that kind of stuff. So I like to just put it out there and be like, I know there’s a chat that doesn’t include me. I’m sure it’s full of GIFs right now. And I’m okay with that.

 I had found an article about how swearing in the workplace was cathartic. And I printed it and I photocopied it and I gave it to everyone on my team and they put it up at their desk so that they could be like, it’s okay, Tanya says we can swear. And our rule was, yes, you can swear at work, just not in front of clients.

It’s cathartic.

[00:14:01] Chicago Camps: You’ve been a vocal critic of UX theatre. What are the signs that a company is merely performing UX design rather than genuinely embedding it into their process?

[00:14:14] Tanya Snook: Yeah, there are so many. The UX theatre, the first time I ever posted about it was I was on a bus on my way into work and I was ranting on Twitter and I did this whole big Twitter thread on it.

And it went like wildfire. People started sending me examples of it and saying things like, Oh my God, I feel seen. And that’s when I knew this concept really had unfortunate legs. People were really seeing themselves in it. And so there’s actually a really long list of signs that you’ll see to be able to watch for it.

One of them, which is definitely one of my more controversial takes is design thinking. When an organization doesn’t actually have user centered design methodologies or UX teams and everything that they do is centered around design thinking, then you know that they’re not really mature when it comes to UX.

So design thinking isn’t meant to be how you do user centered design. It was really created just to get executives to understand the concept of UX, but it wasn’t meant to be applied to do user centered design. So when I see an organization and they’re pushing design thinking as their UX methodology, there’s a really high chance that some of the stuff that they’re doing is going to be UX theatre.

Thinking from the user’s perspective, there was one project that was driving me crazy and they sent, they gave me their whole project plan and they said, what do you think? And I was like, when does the research start? And they pointed to the 16 weeks of workshops and they said, this is the user research.

This concept of role play or thinking from the user’s perspective, taking people out of their day to day and then just asking them to think about their work, where you would have actually had access to them to go and do some job shadowing. Did you process mapping and that sort of thing? That’s definitely UX theatre.

Testing with the executive, which is a thing that was being encouraged for a while in the government, test with your executive. And I’m like, why? They’re not the user. So when you see this kind of silly concept, that one just kills your heart a little bit. Or if there’s no testing or the testing’s at the end, can you just put a UX on it?

It’s like we do UAT and then we do user testing, right? That’s how that works. Or isn’t UAT user testing? Okay. I love that one too.

Another hot take, which I think has caused a lot of UX theatre, is the concept of everyone’s a designer. I hate this one. Everyone should be involved in design. Yes.

Everyone is not a designer. We have training and methodologies and everything to be able to do. It’s really hard to explain to people though when UX is, as an industry, is so nascent and what are the methodologies and where do they come from and all that kind of stuff. So the structure behind our industry and behind our methodologies isn’t understood. So people think anybody can draw a wireframe.

Execs who talk about, but don’t fund UX, that one hurts a lot. And then they’ll often use schedules and budgets as excuses not to include users. My absolute favorite statement from all dev teams, we know our users. We don’t need to test. Everyone’s heard that one at some point.

There you go.

[00:17:36] Chicago Camps: How do you see this concept playing out in the context of AI integration within organizations?

[00:17:42] Tanya Snook: There’s a lot of potential for AI to generate UX theatre. I was just reading Accenture has a tech report, and there was this line in it that says, “Meta GPT by Facebook, claims to automate an entire software development stream by acting as product manager, architect, project manager, and engineer all rolled into one, delegating tasks to its array of GPTs. From one line of text, MetaGPT can generate user stories, competitive analyses, requirements, data structures, APIs, documents, and beyond.”

That’s a stone’s throw away from saying we’re going to replace entire dev teams and UX teams with GPTs. And anyone who, this isn’t an Accenture report that we just came out a month ago.

Like anybody who believes this shit at an executive level will defund their UX teams, let alone their dev teams. Who’s going to go first? Is it going to be your developers or is it going to be your UXers? And so I think there’s just this broad stroke misunderstanding of the quality and the depth of what these things can do.

And given that UX is still so nascent and people are just learning about it and so many organizations lack maturity, you can see how some executives are just going to call themselves out by replacing or trying to replace UX with these GPTs. And, okay, good for you, good luck with your VC funding and your rollout, but, and we’ll be able to pick out those organizations to not work for.

Where the problem ensues is where they still have UXers and they’re doing this stuff. And then we have new UXers who believe that this is okay. We have new UXers in the last decade who’ve come along who think design thinking is okay. Who think that design thinking is how you do user centered design, because they came in during that wave where executives were talking about that and where people were replacing UX teams with design thinking hubs and innovation hubs and that sort of thing.

So that’s where it can really perpetuate a lot of UX theatre. It’s at the executive level. The executives need to be educated on what the capacities of AI really are. Do you want your AI developing your personas? Oh my God. You want to be able to go in and ask your GPT questions as a proxy for your actual users?

[00:20:08] Chicago Camps: Considering the anxiety and mental health challenges prevalent in today’s workspace, what strategies do you believe design leaders should adopt to create a more supportive environment, particularly during high stress situations like performance reviews?

[00:20:23] Tanya Snook: There’s so many mental health challenges.

I rant pretty frequently about RTO and commuting. Return to office, or as the executives like to call it, return to work, because as if we weren’t working the whole time, it’s not the same as it was. In Ottawa, for example, they’ve been building light rail. And whereas I used to take a one shot bus from my house downtown, 40 minutes, now it’s no less than an hour and 10 minutes because I have to drive my car to the bus station, bus to the train to downtown, and then walk a few blocks to my office.

And buses are ghosting all the time and the trains are down and it’s not the same. Commuting now is not the same as it was, and it just causes so much stress. We don’t have places to put our stuff in the office. You have to bring your house with you every time you go downtown. It’s just very different. And when you go, what do you do? You sit on teams the whole time.

In one of the buildings that I worked in, we had mouse and rat issues as well. And the pipes were moldy, so you couldn’t drink the water. Your baseline walking into the office is already high. You have to bring your water for the day, there could be mouse droppings on your desk when you go in.

It was really stressful. It really was. Like, you would have to call health and safety and they would come and clean the mouse droppings off your desk. You weren’t allowed to touch them. It was really disgusting. And this was because of the tunnel building downtown to the light rail. And I remember them doing a mental health session and I put up my hand and I said we’re starting off at a five before we even walk in the door. I’m not even accounting for my personal life, and now whatever else you add to me over the course of the day starts from that five. How dare you have a mental health session and not even acknowledge like the conditions in which we’re working.

So there’s a lot of talking out of both sides of your mouth when it comes to the mental health challenges at work, right? Add on to that, the changing landscape of design. Right now, it feels like the web bubble of the early 2000. I went through layoff after layoff when I worked in startups and in the tech industry from 98 to 2003, and I bought a house, and the math was, how will we afford the mortgage when Tanya gets laid off?

Because that was my choice. I was working in this industry and I just knew it was gonna happen. And we’re seeing that now, right? We’re seeing a lot of design teams being laid off, no rhyme or reason. At least back then we could say that it was because they ran out of money. They ran a VC funding, right?

Their web businesses were dreams, but now it’s even harder because some of these have record profits and they’re doing layoffs, like how do you reconcile that as an employee? The cost of living is so high and we’ve got inflation and housing costs and interest rates. So it’s really awful to have all these pressures.

We have DEI issues, right? Like we have organizations who are just not decolonizing. They’re not dealing with the diversity and equity as they should. They’re not treating all of their employees with fairness. And so if you’re coming from an under resourced group or an underserved group, now you, your baseline was already high to begin with, and now you’ve got all this extra added stress.

And even if you’re lucky enough to still have a job and to be treated okay at work, what’s your career path outside of moving into management? I’ve been doing this for 26 years. Yeah, I’m a manager now, and I tried really hard to remain an individual contributor, but at some point, you just gotta bite the bullet, because what else are you gonna do?

I had the same salary for a decade. It didn’t even, I didn’t even get COLA adjustments because of the way that the salary bands work. That was really hard. And so at some point, you gotta bite the bullet and be like I guess I gotta go into management. If that’s not your thing as a neuro spicy person, do you know how I was able to do that?

16 weeks of management coaching to be able to learn how to manage other people while managing my own anxiety. We work in this environment that’s really busy, really pressured. There are a lot of forces coming at play. Will AI replace our jobs? Creating a supportive environment. For me, that first step was to really admit to myself what my limitations were going to be becoming a leader.

And I knew that I have really bad days. And I have insomnia sometimes from my anxiety, my emotions are always on the outside. How was I possibly going to be able to have empathy for other people and to be able to help them when I was so mired in my own stuff? And so I was really lucky, they invested in me.

Four months of weekly management coaching. And I would go in and I’d use it like leadership therapy. Be like, okay, here’s a scenario. I’m having a really bad day, I’m having a really bad week. My life is chaotic on the outside. How do I take care of these people? And it was so helpful. I would never have been able to move into leadership without it.

So asking for that help yourself as a leader and saying, I know what my limitations are, like, how are you going to invest in me if I’m going to take this on? How are you going to invest in me to help me be able to take care of these people properly?

For the peoplee that I manage, being the first to admit how I’m feeling is really important. Honesty is not weakness. It’s really empowering. And for me to go in and say, I am not having a good day. I’m here because I feel like I need the distraction. So I will be here today, but I’m going to be doing things that don’t take a lot of my energy.

And if you need me today, I can’t guarantee that I can be here for you today. So if it can wait until tomorrow, I can be here for you tomorrow. I think it’s really fair for us to say that. Obviously, sometimes things are going to explode. And so then you can decide, okay, am I the person who can help you with that today?

Or maybe I’m going to send you to my boss and my boss can help you. And I’ve told my boss I’m having an off day, so my boss expects it. So that transparency can really help us be there for our people. I also want my staff to be able to feel like if they want to be at work on a day when they’re just having an off day, or if I want to challenge them to make sure they take their sick days, I always ask, are you better or are you here?

Because if you’re here and you’re not better, I don’t want you here. Unless it’s a mental health thing and they have so much anxiety or so much chaos going on in their personal life, sometimes they need the distraction. And so that’s okay. So I provide suggestions of what they can do to still feel useful, right?

Clean your inbox. When’s the last time to clean your inbox? Training. Here’s some small tasks. Does anybody else need some support or some writing or, so I’m telling them to be gentle with themselves and to find things that they can do. We would share meditations with each other, like YouTube channels. We also shared a Spotify playlist called Raging Ragey Rage Tunes, which I love.

It’s all like death metal and awesome EDM and stuff like that can be really helpful for the team. We built the playlist together of all of our favorite Ragey Rage songs. And so when somebody was having one of those days, they had stuff that was curated by their friends on the team.

Over video can be really hard to pay attention. And one of the things that I learned to do during the pandemic was to pay really close attention to visual and physical cues. To really watch people’s eyes. I’m a lip reader, so I just always happen to be watching people’s mouths when they’re talking. And so it’s, that was something that I was paying attention to, little twitches, little eye things or our eye focus and stuff.

That’s something you can actually train yourself to do. It’s something you can read about and it becomes really helpful when you want to make sure that your team is okay and check in on them.

I would learn, I learned the signs of burnout. I knew them from myself. But I learned for them and was watching for them from my team, even my boss. I actually told my boss once, have you ever considered whether you might be burning out? And he was. And so he ended up taking some time off because he didn’t realize how negative he was becoming. He didn’t realize he was a little sarcastic more than usual. And there were some things that were emerging that weren’t part of his normal behavior.

So I think looking at how to build a supportive environment really means paying attention to what are you doing to care for yourself? How are you demonstrating that you’re caring? How are you modeling that for your staff? How are you paying attention to your team? How are you encouraging them to support each other?

Being really honest with everybody around you, including your boss and defining what bringing your whole self means. I’m a private person. I don’t want to tell you everything about everything. I will tell you some stuff and I will happily share some details of my life. But. I’m probably not going to tell you what I did last weekend.

That’s just my jam. The concept of bringing my whole self for me means that I’m going to be honest about how I’m feeling. I’m going to share my ideas. I’m going to be really collaborative. I’m going to hang with my peeps whenever I get the chance, but it doesn’t mean I’m going to bring all my personal details.

And I think defining that means that you and your team can define together what that’s going to look like. I think there’s creating those opportunities for people to engage with one another without having to do shitty, fake icebreakers and social events. It’s really more about supporting each other.

My team had their cameras on the entire pandemic. We wanted to see each other, like we bonded really closely. Most of them were even hired during the pandemic. They didn’t all live in the same town. We were distributed all over Canada. And that was one thing that most people would comment to me is I can’t believe your team always keeps their cameras on. Well they want to see each other.

If somebody’s having an off day, they have every right to turn their camera off. And people weren’t feeling it. They would even say it. I’m not having a good day, guys. I’m going to be off camera today. And it was like, okay, what do you need? Blah, blah, blah.

I don’t know. It was just one of the nicest teams I ever worked at, worked with. We were together for almost six years. And it was all because we were able to really be ourselves and be honest and be there for each other.

I love performance reviews where I know most people like either to see them as a paper exercise or like a total waste of time or stressful.

I think that the whole point of a really good human centered performance review is to show people how their work makes an impact and how they’re contributing to the team. To prepare for them, we have a ticketing system for all of our work on my last team. And so I would go back to the beginning of the year and I would pull up tickets by person and I would go through all their tickets and I would tell them what they worked on.

I would make a big laundry list of all the things that they worked on and associate it back to their goals and everything. But oftentimes who does that? Who goes back through their own tickets? So they would forget all the things that they had worked on that year and we’d go through what their training goals were and what they had achieved and how they move.

And if they had any growth goals, like any stretch goals or any areas where they’d been deficient the year before, whatever, I have been giving them feedback consistently. And so when we get to the performance review, it’s a review. It’s not new information. Only a monster brings new information to a performance review.

Nothing should be a surprise, right? It should be this lovely. Review of everything that you’ve done. So here’s all the cool things you did. Here’s the stuff you learned. Here’s the stuff you worked on. Here’s some areas, we’ve talked about this, that you have these two areas that you have to, I can’t give you a surpass in those. I have to give you a doesn’t need. We’ve talked about this. We have a plan. You’re going to work on this next year. You’re going to, you’re going to meet it next year. It’s going to be good. You know what training to take, you know what to work on. We’re still working on this week to week. That’s not a horrible conversation to have with a person.

We also did peer boasting. So this was my favorite part. I would create a different form every year where I would have the team fill it out for every other person on the team and basically say, what does that person add to the team? What did you enjoy working with them on this year? What do you, what would you like them to teach you?

Which, if we were gonna do a pointless powerpoint, or if we were gonna send this person to space, what job would they have? Or what would be the one thing that they could talk about without you even, without having to prepare? So there was always like a fun question, or what’s their superhero name? And people loved it, they would love filling it out, but then I would read it to them at the review.

Here is what your team appreciates about you. And I just loved it, cause like my whole goal during performance reviews is to make them blush at some point. Even if we had a really hard conversation, I wanted them to blush. And it would be so fun. And then I’d Canva, I’d lay the whole thing out with all the feedback from their peers, and I’d give them a nice little keepsake.

And it made them look forward and prepare also for performance reviews. Come in and talk about objectives and be ready to talk about areas where they want to learn or what they want their next to be in their career. And we also separated like year end from the beginning of the year. Our fiscal year end is in March, so you get your year end in March, you get your like planning for the next fiscal sometime in April or May. And that separation as well meant that you could wrap up the year with closure and then you could start the new year with a new thing. It was such a nice way to separate things out.

I think that a little bit of effort, I had 12 team members that I was doing performance reviews for, and it maybe took me a day to prepare for them. That’s not a ridiculous amount of effort to show people that you see them. And that you care. And I always thought it was super fun to do.

[00:34:09] Chicago Camps: With AI becoming increasingly embedded in our design processes and products, what steps can organizations take to ensure that the intersection of AI and UX design prioritizes genuine user needs over superficial solutions?

[00:34:26] Tanya Snook: This one gets me really excited because I think there’s some really cool things that AI can do.

My first ML project that I worked on was the access to information and privacy request portal for the government of Canada. If you go there and you start filling out an ATIP request, it has scanned all the summaries of previous ATIP requests. And it has fed the search. You put in your request, and it’s Oh, we might have already answered that, and here’s some potential search results for that topic.

And if you look at them and you think, Nah, not really my thing. There is a recommendation engine on the side that tells you which department is the most likely department, based on your query, to have the answer to that request. Which is really important because when you submit an access to information and privacy request you can only submit to one department at a time.

And so having a recommendation of something related to satellites and whether it’s the space agency or the communications, regulation or it’s D&D, the defense folks. So having a recommendation engine come back, which has read a bunch of releases and can say, based on that, we think you should submit over here.

That will make sure that we have fewer misdirected information and privacy requests coming in from the public. The initial search as well being fed by ML means that we have fewer duplicate requests that are coming in from the public. This was implemented in 2018. We prototyped it in 2017.

This was a really great use of machine learning from seven years ago. And I think there’s so many ways that we could be using. Machine learning and AI related to the work of UX to help us. Like for me, I think it’s a wonderful way to build IAs, right? Like mapping out ontologies based on unstructured data.

And national language processing and providing support to users using that even through a chatbot interface. To me, if I look at a search on a website, and I look at departments, governments have these massive websites and they’re still getting people to tag things, and yeah, they get smarter searches and stuff, but there’s no standard metadata and there’s all these different even the open data websites, like those are great opportunities, put machine learning on these websites and have it assist with the findability.

It’s not sexy. No one can see it. It’s not a Boston Dynamics robot. It’s just this unsexy thing that helps. I would love to take all the UX research that from across the government and put it into a giant data lake. Why doesn’t that exist?

I wrote about that in 2017. I’m like why doesn’t this exist? How many departments are testing similar things and we could all be benefiting from each other’s insights.

It’s unsexy!

It’s really easy to build, and it’s so easy to design. Processing large amounts of data that are too big to process by hand. Looking for untapped insights. Confirming. Taking something, doing all the work, and then just asking something, asking a system did I catch everything? That validation, it’s like an assistant. Doing first drafts, doing desk research, find me all of the sources for blah blah blah synthesizing ideas. I like to feed it bullet points and be like, how would you write this? As a strategy statement and that, so there’s a lot of uses for it to help us as an assistant to do our work or to do really unsexy things in the background.

It’s really hard to get those things funded. I have never been able to convince anybody to create a UX research data lake for the government of Canada. Seven years I’ve been talking about this thing no, because no one can see it, and it’s not sexy, and what does it do for Canadians? A lot! And it would save us a lot of money, but it’s not sexy.

That’s the thing that I think is so promising about machine learning and AI is if we look more for how can it help us do the work as opposed to what is it going to replace? What kind of crazy innovations, let’s have an AI and VR and be in the metaspace. Like, why don’t we look at the things that we don’t like doing?

And replace those things, right? Isn’t that what factory automation was about? It was about taking the work, it was very controversial at the time, but it was meaning it meant to take menial work and replace it with, robotics and automation.

There’s always going to be people who need to be upskilled. You need to learn what makes a good prompt. One of the things that I did this year, I had to do a performance review in French that was very difficult. And so I wrote it in English. I had a GPT translate it for me. I rewrote it because the French was awful. But then I asked the GPT to make it sound friendlier because I’m a neuro spicy anxious person, and I’m very blunt.

And so I had the GPT go in and massage it and make it sound really friendly. You know what? It was a great performance review. That person heard everything I said, because it didn’t come out like I would normally talk. It came out, with this really nice friendly tone and it was very supportive and encouraging.

And it was something that wouldn’t come naturally to me when delivering something that’s that difficult. My, my brain could not figure out how to say something so critical so critiquing in a nice way. It’s super helpful.

So I think that’s where we can take advantage of this to prevent those UX theatre uses. And encourage people to, channel these tools to be used in the right way.

Even, I remember seeing a system in the 90s in my first job, and it was a system that would transcribe and index and search live captioning on video. And at the time I was like, oh, user research interviews, it was so expensive. It was completely impossible to get, for any small business at the time.

And I was just like, the information architecture and the tagging and the metadata you can get. Yeah, I thought it was just going to be amazing. And are we there yet? 26 years later. No, we’re still not there yet. Because it’s not sexy, people won’t fund it.

So we have to find ways to do this stuff, in a hacking way, but then that imposes risks, like people’s personal information go into these fun little GPTs that we create behind the scenes and that we, so not funding this stuff and not backing it up opens our organizations to risk because our staff then turn around and they get innovative and they get resourceful and they find other ways to do this. But then they put, it’s cybersecurity risks or privacy and data risks by using these tools and by not supporting it and not recognizing how much potential it has, all you’re doing is either showing people the door when they’re going to find an organization where they can use it, or, you’re opening yourself up to all sorts of legal and other implications.

I wrote about that scary use of AI about voice AI. What if somebody took a recording from users as, justification for how great their thing was, and it was actually a bad review, but they manipulated it so they could get VC funding or government approval or something like that.

And then it was like, would you like AI to rewrite this? But I was thinking like Theranos, like when you’re trying to sell a product and you’re going around to get funding and you’re showing people testimonials, like manipulating video and audio testimonials is a terrifying prospect.