P16: Survey Data and Session Summary
Survey Responses
| Question | Response |
|---|---|
| Age | 35–44 |
| Education | Master's degree |
| Current role / position level | Individual contributor |
| Job title | Senior Product Designer |
| Years of professional experience | 8–15 years |
| Organization description | We support commercial customers with SaaS solutions. |
| Industry | Other or not sure |
| Individual AI tools used | Text generation (e.g., creating documents, emails, summaries), Media creation (images, audio, video), Search and information retrieval, Data analysis and synthesis, Code generation and completion |
| Organizational AI tools deployed | Customer-facing chatbots or virtual assistants, Internal search and knowledge summarization, Customer recommendation systems, Predictive analytics for business forecasting, Content moderation or filtering systems |
| AI adoption involvement | No direct involvement in adoption or deployment (mostly a user of a deployed AI system) |
| Biggest work win with AI | Using its tools to quickly obtain insights from meetings or research. Using Figma Make for quick iterations. |
| Biggest disappointment with AI | Some of the search capabilities are lackluster. It can skew data so you have to make sure to review it; some people don't and it shows. |
| Organization's biggest AI success | Ability to quickly use it to generate content or synthesize ideas and research. |
| Organization's biggest AI challenge | Unsure |
Background
P16 is a Senior Product Designer at a B2B SaaS company supporting commercial customers. She holds a master's degree, has 8–15 years of professional experience, and works as an individual contributor. Her organization has positioned AI use as expected to the point that non-use is "a problem," but she does not know how compliance is measured. She is one of the more enthusiastic personal-life adopters in the study, using AI for meal planning, financial forecasting, retirement modeling, room decoration, hidden-subscription audits, and even the design of a backyard play structure based on her property's plot plan.
Her professional adoption was catalyzed by an in-house LLM tool that another colleague built to help employees draft year-end and mid-year development objectives. The first time she used it, the cognitive load of figuring out what to write dropped sharply and her view of AI shifted. The biggest day-to-day change since then has been Figma Make for early-stage design work and various synthesis tools (Figma's research synthesis, Microsoft Teams transcripts) for collapsing what used to be days of manual sticky-note sorting and recording review into a quick summary.
The session's central focus is what happens at the boundary between the design team and adjacent teams who have access to the same generative tools but lack design training. P16 is balancing two pressures: hold the line on design guidelines and craft, while not being seen as the bottleneck slowing down a five-minute "design." She also names her own cognitive style ("process brain first versus a visual person") and uses Figma Make as a bridge across that gap, while already worrying that the trade-off is atrophy of her own design edge.
Key Findings
The Five-Minute Problem
P16 describes a recurring pattern: product owners on adjacent teams use Figma Make to produce a finished-looking design in five minutes, then bring it to designers and ask why their work takes so much longer. The friction is not about whether AI tools should be used (P16 uses them herself) but about what an AI-generated artifact actually represents. P16 has had to articulate the difference between an artifact-as-conversation-starter and an artifact-as-build-spec, and to do so without alienating her co-workers' product owners.
"The product owners have been using either Figma Make... and they will come up with a design and essentially it's like, 'well, why does it take you so much longer to do what I could do here in five minutes?'"
Her response is a balancing act: gently saying "stay in your lane" while also using the artifact as a communication tool, asking, "tell me more about why you think this is the best solution and maybe there's gaps in our communication." The vulnerability she names is that AI-generated designs are not in compliance with the organization's design guidelines, so accepting the artifact as ready-to-build means accepting unaligned work into the production pipeline.
She illustrates the failure mode with a concrete example: a high-leadership project where someone put together an idea in Figma Make, the development team and leadership read it as ready-to-build, and the result was effort spent treating a conversation prop as a specification.
"There's this big project that is being done at the very high leadership level, that someone put together an idea in Figma Make and the development team and leadership thought, like, 'okay this is ready to go and ready to be built.' And it was really just a tool to explain what could be done."
The Grayscale Prototype Analogy
P16 articulates a useful framing for why vibe-coded artifacts cause downstream problems. The same reason designers show grayscale wireframes at the beginning of ideation, not finished clickable prototypes, applies to AI-generated work: high-fidelity output bypasses the conversation about whether the underlying problem has been understood correctly.
"The same way that we would want to only show grayscale designs at the beginning of an ideation session, you don't want to show a full clickable prototype either."
The analogy applies because design teams already have an established practice for managing fidelity. Wireframe-first conventions exist because high fidelity at the wrong stage forecloses iteration; participants stop critiquing the underlying idea and start critiquing the colors. P16 is naming the same dynamic at the boundary with non-designers: a Figma Make output, presented as a finished design, ends the conversation about whether the design should exist at all.
Process Brain Versus Visual Person
P16 names a specific cognitive style and locates her AI use within it. She describes herself as "a process brain first versus a visual person," meaning she can describe what she wants but struggles to produce the visual artifact directly. Figma Make produces the visual she can then reason about and refine.
"A lot of the time I'm the type of person that, like, I can describe what I want, but I'm having a difficult time with, like, I'm a process brain first versus a visual person."
"Once you have the visual then I'm like, oh okay, I can run with this."
This is a more specific framing than the general "AI as productivity tool" stance. P16 is not claiming AI makes her faster across the board; she is claiming AI bridges a particular gap in her cognitive style. The pattern matches how P6 and P9 describe AI as compensating for self-identified cognitive weaknesses, with P16 articulating the gap with unusual clarity (a named cognitive style paired with a specific tool that produces the artifact she can then reason about).
The Skill-Erosion Trade-off
P16 is already worried about her own design edge. Because Figma Make and similar tools now handle the early-stage work (wireframes, beginning screens), she is not designing in Figma or Sketch as frequently as before, and she frames this as a probable loss of skill.
"Maybe losing my edge with actual design. I mean, it was never my forte to begin with, but because we're using it to build at least beginning screens or wireframes, I'm not improving my craft as I used to. I'm not designing in Figma, previously Sketch at our company, as frequently as I used to. So, I feel like I might be losing some of my skills."
The connection to apprenticeship erosion is direct. P16 names two cohorts at risk of having no equivalent of the experience she had during her first four years: people who started their career during covid (lost in-office mentorship, lost networking, lost real-people-to-talk-to access) and the next generation who has never worked without AI.
"I spent the first four years of my IT and design career in an office where we had access to leadership, we had access to networking, we had real people to talk to."
"I have major concerns, especially having young kids... you need to be able to think on your own. You need to be able to think on the fly and come up with your own ideas."
P16's worry, taken across her own situation and these two younger cohorts, is that the chain of skill development is breaking at multiple points at once: experienced practitioners losing depth because AI handles the practice, mid-career practitioners who never got mentorship because of remote work, and a generation that never built the foundation at all.
Supervisor Feedback and the Authenticity Question
P16 raises a relational concern not heavily covered in earlier sessions: the use of AI to generate performance feedback. She finds AI useful for writing her own development objectives, and the in-house tool that supports this was the moment that flipped her relationship with AI at work. But she has noticed that supervisors use the same kinds of tools to write feedback, and she questions what that does to the developmental conversation.
"Using it for my personal objectives at work is fantastic and everyone does it. Which I will say is kind of a drawback, though, because supervisors are also doing it to give feedback. And I'm like, is that feedback something that you truly think, or are you getting it from the tool?"
This articulation prompted an expansion of the AI Slop Detection theme to include the relational consequences of recognizing AI authorship, not only the reputational consequences of recognizing low-effort AI output. The harm is not "this person looks lazy." The harm is that the developmental conversation between a supervisor and a report has been hollowed out, because the words evaluating the report's work were not written by the person delivering them.
Emerging Themes
| Theme | Description | Key Quote |
|---|---|---|
| Vibe Code Governance | The challenge of evaluating AI-generated artifacts produced by non-designers | "Why does it take you so much longer to do what I could do here in five minutes?" |
| Trust Calibration | Deliberate practices for evaluating AI trustworthiness | "I never trust it at face value. I will still go review it." |
| Skill Erosion | Perceived atrophy of professional craft attributed to AI handling the work | "I'm not improving my craft as I used to." |
| Apprenticeship Erosion | Concern that AI prevents junior practitioners from developing foundational skills | "I spent the first four years of my IT and design career in an office where we had access to leadership." |
| Organizational AI Adoption Challenges | Mandate without measurement | "If we're not using AI, that's a problem... I don't think they [measure] it." |
| AI as Cognitive Prosthetic | Using AI to compensate for a self-identified cognitive style | "I'm a process brain first versus a visual person." |
| Pervasive Integration | AI use spanning work, finance, food, home, family | "I can do meal planning for my family without having to do all this extra work." |
| AI Slop Detection | Recognizing AI authorship and its reputational and relational consequences | "Is that feedback something that you truly think, or are you getting it from the tool?" |
| Disclosure Norms | Layered formal training and informal team practice | "I know that there are norms... I haven't really read them." |
| Knowledge Displacement | Concern that AI dependency erodes critical thinking | "They need to be able to learn what are ways to still work without AI." |
| Augmentation Not Replacement | Locating human contribution in originating ideas | "It still needs people to feed it." |
| AI as Sounding Board | Using AI as a thinking partner for unfamiliar questions | "Bouncing ideas off of someone when I don't necessarily know the answer." |
| AI as Learning Partner | Wishing for AI tutoring that does not yet exist | "If Figma had an AI tool that I could interact with in real time." |
P16 contributes a second session to Vibe Code Governance (the first was P14). Where P14 framed the issue from a head-of-design vantage at a small startup, P16 articulates the practitioner-level friction at a larger B2B SaaS company. Her grayscale-prototype analogy is a portable framing device for explaining why fidelity matters at the wrong stage. The "stay in your lane / tell me more" balancing act she describes is a coping mechanism for staying in productive relationships with adjacent product owners while still defending design guidelines.
"It can be great as a conversation tool or great from a design tool of iterating quickly, but I still think we need to be cautious because the same way that we would want to only show grayscale designs at the beginning of an ideation session, you don't want to show a full clickable prototype either."
P16's trust calibration evidence draws an explicit parallel to general internet research. She does not trust AI output at face value, and she frames the verification practice as continuous with how she would evaluate anything found through a Google search. The undergrad stats class is named as the lens that gave her this framing.
"I think that goes with anything that we research online. You know, just because you Google something doesn't mean it's fact. It's all in reference to the context that you give it. So analyzing metrics, analyzing statistics, I took a stats psychology class back in my undergrad and it's given me a whole different perspective on statistics. Like, you can look at statistics five different ways and it's going to give you five different answers."
P16's skill erosion contribution is specific to design at the wireframe-and-early-screens stage. The concern is not that she will lose advanced skills she already has but that she will stop deepening her craft because the practice opportunities have been moved upstream of her work.
"Because we're using it to build at least beginning screens or wireframes, I'm not improving my craft as I used to. I'm not designing in Figma, previously Sketch at our company, as frequently as I used to. So, I feel like I might be losing some of my skills."
P16's apprenticeship erosion contribution adds a covid-era cohort that prior sessions had not articulated. The framing is concrete: she had four years in an office with access to leadership and networking before remote work changed entry-level conditions, and she names that access as the thing newer cohorts have not had.
"The same thing goes for the people who started their career, um, during covid. They never, like, I spent the first four years of my IT and design career in an office where we had access to leadership, we had access to networking, we had real people to talk to."
P16's organizational AI adoption challenges evidence is a clean instance of mandate-without-measurement. Leadership has positioned non-use as a problem, but accountability is invisible to her at her level. She infers compliance from peer behavior and from the development teams' tooling rather than from any explicit metric.
"Our organization is all about it. They're like, 'Use AI for anything and everything. Find ways to create efficiencies.' And even to the point of, like, if we're not using AI, that's a problem... I don't think they do [measure compliance]. I don't know how to answer that question."
P16's AI as cognitive prosthetic articulation is the cleanest in the dataset so far. She names a cognitive style, identifies the gap (description-to-visual), and points at the specific tool that bridges it. This is a more targeted use of AI than the general "AI as productivity tool" frame, and aligns with how P6 and P9 use AI for self-identified cognitive corrections.
"A lot of the time I'm the type of person that, like, I can describe what I want, but I'm having a difficult time with, like, I'm a process brain first versus a visual person. Once you have the visual then I'm like, oh okay, I can run with this."
P16's pervasive integration spans more domains than most participants in the study: meal planning, recipe generation, retirement planning, room decoration, hidden subscription detection, charge analysis, and a backyard play structure designed against the property's plot plan. The breadth is striking because she is otherwise reflective about the costs of AI dependency at work.
"Where do we have gaps in our finances that we could cut that we're not seeing? Like, here's all of our charges, where are hidden subscriptions that we don't see, or where are areas that we could cut back?"
P16's AI slop detection evidence prompted an expansion of the theme. She offers two distinct detection contexts: visual-style mismatch in design output (the existing reputational pattern) and supervisor authorship in performance feedback (a new relational pattern). The supervisor-feedback case is the one that prompts the question in her voice: "is that feedback something that you truly think, or are you getting it from the tool?"
"Using it for my personal objectives at work is fantastic and everyone does it. Which I will say is kind of a drawback, though, because supervisors are also doing it to give feedback. And I'm like, is that feedback something that you truly think, or are you getting it from the tool?"
P16's disclosure norms evidence is layered: she knows her organization has integrity training classes around AI disclosure, but she has not read the official documents. The norms that actually govern her behavior come from her team's informal practice ("you always disclose if this is something that you made in Figma Make"), reinforced by visible patterns ("this was created with AI" taglines, "AI-generated image" labels) appearing in the broader environment.
"More people are disclosing, like, 'this was created with AI' as a tagline, or 'AI-generated image,' or 'made with the assistance of this tool.' We have always been, the people on my team have always been encouraged of, you always disclose if this is something that you made in Figma Make."
P16's knowledge displacement evidence connects the personal to the structural. As a parent of young kids, she worries about a cohort that has never thought without AI assistance, and her concern spans innovation, security awareness, job-market navigation (resume writing, interview integrity, posting authenticity), and the basic ability to come up with original ideas.
"They need to be able to learn what are ways to still work without AI, but how do we use it to make the world better?"
P16's augmentation-not-replacement stance locates human contribution in the originating-ideas role. AI can expand on existing ideas and remove cognitive load on familiar tasks, but it cannot generate genuinely new things without people to feed it.
"I think it can expand on ideas that are already generated, but it still needs people to feed it. We still need people to come up with new things, right, to feed the tools."
P16's AI as sounding board contribution is brief but explicit. The use case is unfamiliar questions: when she does not know the answer, she uses AI to bounce ideas rather than to receive a definitive response.
"I use it a lot for just bouncing ideas off of someone when I don't necessarily know the answer."
P16's AI as learning partner contribution is forward-looking: she names the unmet need rather than describing an existing practice. She wants a Figma-integrated tutor that can give real-time feedback on her visual design as she works, and she frames the alternative as paying thousands of dollars for one-on-one training.
"It would be really cool to have a tool created for learning and training that can adapt based on, like, for example, I really want to learn more about visual design. If Figma had an AI tool that I could interact with in real time and I can learn it, and it can give me real feedback of, 'okay, the way that you did this isn't the most efficient way of doing it. Let me show you how to do it in real time,' rather than having a one-on-one session with a person."
Interview Transcript
00:00:00
Paul: I'd like you to tell me the story of your first "oh wow" moment with AI. So what was going on that made you try it and what happened that made the light bulb turn on for you?
P16: In personal or work life?
Paul: Either one is fine.
P16: Okay. So I would say I started using it at work because we have an in-house [tool] for [company]. It's an LLM tool that uses, we can just decide which engine to use, and we've been encouraged to use it by leadership for various things. The first time that I used it was really for my development objectives for year-end and midyear. Someone at the company had created a tool that uses the basis of the objectives that we're supposed to use, and it asks you all of these questions and then it spits out a summary at the end. So it makes it really easy to utilize. So that was really the first time I had used it at work, and that was mind-blowing because it took a lot of the brain power out of trying to figure out what to write.
00:01:22
P16: I still gathered a lot of the information to feed into the tool. But I didn't also have to think of prompts on my own. In my personal life, I use it for many different things.
For example, we're trying to figure out a play set structure in the backyard, and being able to put in our plot plan and have it give the overall recommendation. But I think the first time I actually used it for my personal life was trying to figure out a recipe for dinner. I'm like, I have this at my house. I need it to be ready in 30 minutes. Like, what do I do? And it's been great. So, ever since then, my husband and I have used it for forecasting finances and stuff and retirement, and we can put in a bunch of different data points.I use it a lot for just bouncing ideas off of someone when I don't necessarily know the answer, or like getting, doing quick research.
00:02:39
P16: My husband is a very numbers person. So being able to use it for that has been, I think, the most, like, my big "oh wow" moment was like, oh my gosh, I can do meal planning for my family without having to do all this extra work. And then I can feed it different variables. And then using it for. My husband does it for more of a technical deep dive, because I'm not the person that you talk to about all of these servers and data and stuff. But being able to essentially be a financial planner for us without paying for one. Because we tend to research a lot of things on our own and it just really speeds up that process.
Paul: I'd like you to think about one thing that you do regularly that AI has changed the most. So, walk me through what you used to do versus how you do it now.
00:03:52
P16: The other area that I've used it for most recently is with Figma. I'm not a big fan of Miro's AI capabilities. I don't think they're that good, but I'm using it for synthesizing research. So having spent almost an entire day synthesizing sticky notes or notes or interviews, if I'm able to do interviews, we have a research team, but being able to take a bunch of information that I would have to manually sift through to identify trends, being able to put a bunch of different formats of data, of information, and have it summarize it is really the biggest difference. So, especially with Figma, unfortunately we have a token limit now, but being able to just bounce ideas through the system and come up with a solution that I'm like, okay, how could I imagine this complex solution? Because a lot of the time I'm the type of person that, like, I can describe what I want, but I'm having a difficult time with, like, I'm a process brain first versus a visual person.
00:05:34
P16: Once you have the visual then I'm like, oh okay, I can run with this. So I think most recently Figma Make has been the biggest change. But I think just the synthesizing of information, that you can just put it in a tool and have it spit out a summary, that has been the biggest change for me as a designer.
P16: And then same thing for personal, like most recently I've used it for, here's a picture of my room, I need you to decorate it for me. Or, here is our finances. Where do we have gaps in our finances that we could cut that we're not seeing? Like, here's all of our charges, where are hidden subscriptions that we don't see, or where are areas that we could cut back? That's the biggest change that we've been able to make, is doing our own forecasting without actually manually writing the tool ourselves.
00:06:56
Paul: I'm using Claude Cowork for this project and the velocity increases I've gotten are insane. I'm interested to hear how your organization is handling AI adoption.
P16: Our organization is all about it. They're like, "Use AI for anything and everything. Find ways to create efficiencies." And even to the point of, like, if we're not using AI, that's a problem.
Paul: Tell me more about that. How is compliance measured? How does your organization know if you're using it or not?
P16: I don't think they do. I mean, in my organization, at my level, they know if we share it. I don't know how to answer that question. I mean, I think in the development teams they're using it a lot more because there's other AI integrations in the development tools that they use.
00:08:26
P16: But from design it's like they're all about using AI, but it's almost scary because I have some teams that work with business that are like, "well, here, I just made this in Figma Make. Can you just make this?" And there has to be some boundaries with certain product teams because that's not necessarily something that we would want to make. But yeah, I mean, from a measuring perspective, I'm sure there are ways that they're measuring it. I don't know what they are, unfortunately.
Paul: Let's return to what you just said about there being people using Figma Make and saying, "okay, this is what I'm trying to do or what we should do," and you said, "well, that's not how we can do it or should do it." Tell me more about that.
P16: Yeah. So, luckily it hasn't been with my direct team. It's been with a co-worker. Their product owners have been using either Figma Make or, I don't know if they've been able to use Miro.
00:09:42
P16: But those are the two programs that we use most, and they will come up with a design and essentially it's like, "well, why does it take you so much longer to do what I could do here in five minutes?" So I think it's a balance of using it as a communication, like gently saying "stay in your lane" but also using it as a communication tool of, "okay, well, tell me more about why you think this is the best solution and maybe there's gaps in our communication." Because we also have to. I think there are integrations that they can use, like our own design system. Because just because Figma Make comes up with a design doesn't mean that it is in compliance with our design guidelines. So keeping that in perspective: there's this big project that is being done at the very high leadership level, that someone put together an idea in Figma Make and the development team and leadership thought, like, "okay this is ready to go and ready to be built." And it was really just a tool to explain what could be done.
00:11:09
P16: So that's kind of the other aspect of design, is using it and sharing it with business and they're like, "okay, well, let's go build it." And it's like, well, this is just a tool, this is a starting point, let's build on it, like, understand where are the data points and what. This is not actually built in a true design fashion. We haven't figured out dependencies or anything like that. So I see it on both ends. It can be great as a conversation tool or great from a design tool of iterating quickly, but I still think we need to be cautious because the same way that we would want to only show grayscale designs at the beginning of an ideation session, you don't want to show a full clickable prototype either.
Paul: That's interesting. Yeah. You are not the first person who's mentioned exactly that situation, where people who are not in a design role will vibe-code something and present it as the solution, when maybe it should be taken more as a concept communication.
00:12:32
P16: Yeah. I tried to give more prompts to something that I was pretty solid on, and then I ended up breaking my entire design. And I'm like, I have to revert back. Trying to learn it on the fly as well is also great.
Paul: You said "learn it on the fly." Learn what?
P16: Oh, learn Figma Make, like learn how to best use it, and just what are prompts that it understands a little bit easier.
Paul: What do you think's been your biggest win with AI at work so far?
P16: Moving faster on the things that I mentioned before, like synthesizing information that may have taken me a couple of days of manual work, that I can do quickly. I can come up with a quick. When it comes to Miro, I'm still not a big fan of their AI capabilities. I'll use it to compare like, okay, this is what I came up with.
00:13:41
P16: Like, is their summary similar to what I came up with? Or like grouping stickies together. I still like to do that manually. But when it comes to. Oh, the other thing is transcripts. In Microsoft Teams, we use the AI notes and transcripts that way. Not the best, but also better than nothing. Being able to just take a lot of information and pull it together into a summary is the biggest win for me, because I used to spend days on synthesizing research data, or recording, you know, I used to sift through recordings and take notes and all that. But I think that's the biggest win overall. But like using it for my personal objectives at work is fantastic and everyone does it. Which I will say is kind of a drawback, though, because supervisors are also doing it to give feedback. And I'm like, is that feedback something that you truly think, or are you getting it from the tool?
00:15:13
Paul: What's been your biggest disappointment or surprise failure when you tried to use AI for something that you previously hadn't?
P16: I don't know if I've had design failures per se yet, but I will say getting down a rabbit hole with it and then having it not take my prompts. But again, that's not something that I necessarily had before. I wouldn't say. Well, I guess maybe losing my edge with actual design. I mean, it was never my forte to begin with, but because we're using it to build at least beginning screens or wireframes, I'm not improving my craft as I used to. I'm not using Figma design. I'm not designing in Figma, previously Sketch at our company, as frequently as I used to. So, I feel like I might be losing some of my skills.
00:16:43
Paul: Have you noticed any specific changes already, or is it just more of an anticipation or a concern that hasn't manifested?
P16: Yeah, I think that's my biggest concern is, when is this going to make us irrelevant? I don't think so, because there's still something to say about best practices, and understanding the problem fully, and being able to ask questions on the fly. Could someone use a tool to do that? Maybe. But it's like the memes that you see out there, like the rules. It's like, you still have to understand all the complex problems that a business customer needs has, right? Or understand the technical architecture. I could see part of someone's job putting in all of the design criteria, but then I think it's going to fall flat because people aren't going to actually do that. They're just going to say, "Well, I'm just going to design it the way that I want to."
00:18:19
P16: As a developer, I still have developers coming to me, not with AI ideas but coming to me and saying, like, "well, this is going to be better for the customer." I'm like, well, what research do you have that supports that? Because I didn't notice the pattern that you were trying to show me until you called it out. I just thought you had. It was an inline table snack bar with a global snack bar on top of it. And I'm like, well, to me that just looked like things all over the page and I didn't understand what was going on. But you still have to have some level of understanding that what the AI tool is presenting to you is not the best fit. Like, there needs to be some level of that in the role for it to be effective. Now a company could say, "well, we don't care about that," but I'm not going to become a developer because I have something that codes, I'm not going to know if that code is good or not because I don't have the knowledge.
00:19:36
Paul: Has there been a time where you trusted AI and you shouldn't have? What happened? And how do you decide whether to trust the output AI gives you?
P16: I don't know if there's actually been a time that I trusted it and I shouldn't have. If I've been unsure, I'm always one that reviews it. So I never trust it at face value. I will still go review it and determine, like, is this actually what I told it? Is this not? So I can't really say that's something I've encountered. It could have been something I encountered if I didn't review it. It messed up my interactions in a prototype that I was trying to create because the prompt didn't account for it. It messed up the interaction. So, I had to go back and revert and rewrite the prompt. But I think to make sure that we don't take things at face value, you still have to have that analytical thinking, right?
00:20:53
P16: So you review it, you determine, does this make sense? And if it doesn't, research it on your own and come up with the answer. I think that goes with anything that we research online. You know, just because you Google something doesn't mean it's fact. It's all in reference to the context that you give it. So analyzing metrics, analyzing statistics. I took a stats psychology class back in my undergrad and it's given me a whole different perspective on statistics. Like, you can look at statistics five different ways and it's going to give you five different answers. It's just depending on, how do you want that to come off? How do you want that to be portrayed by your reader?
Paul: Are there norms forming at your workplace around when you disclose that AI helped you with something? Like, when do you mention it, when do you not?
P16: I know that there are norms. To be honest, I haven't really read them.
00:21:54
P16: But I'm always. I feel like more people are disclosing, like, "this was created with AI" as a tagline, or "AI-generated image," or "made with the assistance of this tool." We have always been. The people on my team have always been encouraged of, you always disclose if this is something that you made in Figma Make.
I know that there are norms and there's integrity training classes around it and all that.You can tell, I mean, you can tell if it's Figma Make or not, versus our own design system.
Paul: Have you encountered people at work or in your personal life who resist using AI? What do you think's behind the resistance?
P16: Oh yeah. My, I mean, my husband was one of them and now he talks to Gemini more than he talks to me. I always laugh about that.
00:23:04
P16: I was, like, what's Gem up to today? What's on the topic list? But again, it's on very technical topics that I'm not going to care if you talk to me about this or not. But yeah, I think it's a generational thing.
Paul: Tell me more about that.
P16: Well, like, my parents are older, so they fall into one of two categories. Either they believe everything that's on the internet, or they believe, like, China is going to steal all of our information, and they're very generational when it comes to that and very judgy. Like, TikTok was a big thing to my parents. "Oh no, you can't use TikTok." I was like, you don't even know what TikTok is, but okay. But then I would say there's also individuals that don't want to share data with them. And believing that. It's not wrong. The more information that you feed into an agentic AI is going to give it more information about you.
00:24:30
P16: And if it has your login and your information behind it, it could be used to fuel whatever. It's basically like giving your information to Google. Like, my sister is very much, like, doesn't post pictures, doesn't. She uses Signal, she doesn't want to use texting on her phone, she doesn't want to use Shutterfly because it's owned by Google. I was like, "Well, you used to use WhatsApp and that's a Facebook product, so they already have your information. It doesn't go away." So, it is what it is. I think there are tools out there that are more safe than others, and I know someone who is very security conscious when it comes to technology, and they still use it but they use it for certain things. You just got to be careful of what information you feed into it. You know, don't feed in your social security number.
Paul: Let's talk about the next generation of people entering the work world who've never done the work without AI.
00:25:41
Paul: What concerns or excites you about that?
P16: I have major concerns, especially having young kids, that they, you need to understand, you need to be able to think on your own. You need to be able to think on the fly and come up with your own ideas. And there's a time and a place for technology. There's a time and a place for. Sorry, I got to go back to plug it. But there's a time and a place for any sort of technology tool. It's just how you use it that is what's going to make or break you, I guess. I don't know. I think the same thing goes for the people who started their career, um, during covid. They never. Like, I spent the first four years of my IT and design career in an office where we had access to leadership, we had access to networking, we had real people to talk to. And now, it's one of those things where, feel like people need to be more cautious too. Like, I know, like interviewing skills.
00:27:33
P16: Sorry, I think I'm, like, all over the board. So, I feel like there's an idea, there's an innovation piece to it. It can both fuel or hinder innovation. I think there's a security piece to it, where they still need to understand what you do and don't give to the internet. But then there's also, like, finding a job. Finding a job with AI tools right now is, like, insane. I have friends that are looking for work, and I have friends that, like, being on LinkedIn and figuring out, what's a real posting and what's not? How do you make your resume stand out with all of these tools? How do you know if somebody has two jobs? How do you know if someone, my husband does a lot of interviews, learning how to interview without somebody giving you the answer, you know? It's just learning, like, they don't know anything else and that's the problem. They need to be able to learn what are ways to still work without AI, but how do we use it to make the world better?
00:28:53
Paul: Let's zoom out. How does this increasing presence of AI in the world, both work and personal, make you feel?
P16: I used to be really resistant to it. I feel like, used in the right way, it's a benefit. I think it can expand on ideas that are already generated, but it still needs people to feed it. And I mean, I still don't know all of the ins and outs of it.
00:30:01
P16: Like, I'm not going to pretend to know, but it can expand on ideas that are already generated. But we still need people to come up with new things, right, to feed the tools. What really scares me though is the ability to create images with AI and not knowing if they're real.
Paul: What's a breakthrough that you're most hoping for that AI might enable?
P16: Finding gaps in scientific research, like health research. If it's being used for that, I think that would be great. But along with the image doctoring or creation, it really scares me with the voice recognition. I think being a mom, that scares me. If my daughter got, you know, scammed or like taken, or like [UNCLEAR: her boys got re]. You always hear about the people that get screwed of, like, making them seem like their kids are kidnapped.
Paul: That is terrifying.
00:31:30
Paul: So what's the biggest gap between what AI can do for you right now and what you actually want it to do?
P16: I don't know if it doesn't do this right now, but it would be really cool to have a tool created for learning and training that can adapt based on, like, for example, I really want to learn more about visual design. If Figma had an AI tool that I could interact with in real time and I can learn it, and it can give me real feedback of, "okay, the way that you did this isn't the most efficient way of doing it. [I'd want it to say] let me show you how to do it in real time," rather than having a one-on-one session with a person. I think that could help me in my career of learning design tools. I'm sure that there's other ways that it could help my life and my personal life, but I feel like in my day-to-day, figuring out how to be a more efficient designer would be, without having to pay thousands of dollars in an interactive one-on-one training, right?