Proshanjit Mitra, Lead UX Research and Strategy at Lucid Motors, makes the case for UX professionals to broaden their skills into strategy, design, and adjacent domains while using AI tools to handle more work without sacrificing quality. The talk covers the current UX job market, the expanding skill expectations in job postings, and practical AI use cases drawn from Mitra's own experience. It closes with best practices for responsible AI adoption and advice on staying competitive as the field evolves.
Key Takeaways
- Adopt a generalist mindset: deep research expertise paired with skills in strategy, basic design, and cross-functional communication makes you harder to replace and more influential.
- UX job postings dropped sharply from 2022 to 2023 but remain 53% above 2018 levels, and product design and UX research roles were recovering as of mid-2024.
- AI tools accelerate repetitive, volume-heavy tasks such as secondary research, thematic analysis, persona drafting, and survey design. They are starting points, not finished outputs.
- Always review AI-generated content for hallucinations and inaccuracies. Asking AI tools to cite reputable sources during secondary research is one practical check.
- Follow your company's IT and infosec policies before inputting any proprietary or user data into external AI platforms. The level of openness varies significantly by organization.
- Human judgment, empathy, and strategic thinking remain the core of UX work. AI handles the heavy lifting so practitioners can focus on interpretation, storytelling, and decision-making.
Questions & Answers
- Are people using AI to get to requirements? It seems to fall short when finding specifics beyond larger themes.
- Mitra noted that the process of getting to requirements is still driven by surfacing insights and building recommendations from them. AI can generate useful summaries of large text bodies and helps reach insights faster, but thematic extraction is still immature and requires significant human review and intervention. The technology is improving but not yet reliable enough to replace that step.
- What level of security do you look for in platforms like Dovetail where research data is uploaded?
- Mitra defers to the company's IT and infosec teams on these decisions. If those teams are satisfied with a vendor's data handling practices, that clears the tool for use. Mitra does not personally evaluate the specific security standards of each platform.
- Are there other AI-enabled tools you use regularly, whether in-house or closed versus public?
- Mitra described two categories: established research platforms such as Dovetail, Miro, and Lucid Spark that are adding AI features to their existing ecosystems, and newer AI-focused companies including OpenAI's enterprise offering. Both should go through a company's standard IT and infosec review. OpenAI and similar companies are building enterprise-grade infrastructure, but compliance needs to be verified for each organization.
- What challenges have you encountered with AI use cases and how did you overcome them?
- Image generation has been the most common challenge. AI frequently misreads details and produces inconsistent results, though the tools are improving rapidly. Mitra now iterates on specific areas of an image rather than regenerating entirely. On thematic analysis, the technology is still not reliable enough, so Mitra's team continues to rely on manual processes for qualitative analysis while using AI-generated summaries for reports, and is waiting for further progress from platforms like Dovetail.
- As AI progresses, how can researchers make themselves more competitive?
- Mitra's advice is to keep high-level skills strong, specifically the combination of creativity, scientific rigor, and empathy that defines UX work. That combination is unlikely to be automated. Use AI to do more with less and focus human energy on interpretation, storytelling, and strategic impact. Internally, researchers can also build influence by helping their organizations develop a clear, consistent position on AI adoption.
Session Notes
Context: The UX Job Market
Mitra opened with a data-grounded look at how demand for UX professionals has shifted over the past several years.
- Digital transformation during the pandemic drove significant UX hiring in 2021 and 2022, with demand nearly doubling from pre-pandemic levels.
- A market correction followed. UX research job openings dropped 73% from 2022 to 2023 as companies stabilized their workforces.
- Despite the correction, postings were still 53% above 2018 levels, signaling long-term growth alongside short-term volatility.
- As of mid-2024, product design roles had rebounded, product management opportunities had surged, and UX research positions were recovering, though more slowly.
What Job Postings Now Expect
Looking across current UX and insights job openings, Mitra identified a consistent expansion of requirements beyond core research methods.
- Standard expectations remain: methodology fluency, process rigor, strong insight communication, and clear recommendations.
- Newer expectations include: crafting video summaries, acting as a strategic partner to product and design, influencing stakeholders across functions, and contributing to strategic planning.
- Some roles also list experience with generative AI, Unreal Engine, Unity, or extended reality (XR) as nice-to-have skills.
It's not just enough to just communicate your insights. You would want to drive product decision making and product roadmap through your contributions to strategy and design.
The T-Shaped UX Professional
Mitra described a T-shaped model as the target profile: deep expertise in research as the vertical bar, with horizontal breadth across adjacent disciplines.
- Core depth: All major research methods, a strong research process, and the ability to generate and communicate insights.
- UX strategy: Persona development, customer journey mapping, product visioning, design thinking workshops, and feature prioritization frameworks.
- Basic design skills: Concept sketching, and the ability to build clickable prototypes when designers lack bandwidth. Mitra described doing this in a previous role to unblock a research study.
- Cross-functional awareness: Understanding how product management and engineering teams work improves how researchers approach and communicate with those partners.
- Broader domain knowledge: Familiarity with adjacent industries and domains enables faster context-switching and more effective conversations with participants or stakeholders from different fields.
Why the Generalist Approach Strengthens Your Position
- Cross-functional ability enhances job security and makes you a more valuable team member.
- Understanding product strategy lets researchers advocate for user needs in terms that align with business goals, giving research more influence over outcomes.
- Taking on design or strategy contributions enables faster decision-making and adds value beyond insights alone.
- Broader domain knowledge means insights are more actionable and do not stay siloed in reports.
Challenges of Going Broader
Mitra acknowledged the real costs of expanding scope.
- Spreading too thin risks losing depth in core research skills.
- Building new competencies takes time and adds mental load on top of day-to-day responsibilities.
- Mitigation strategies include mentorship, formal training and development opportunities, and building a peer network of UX professionals.
AI as a Complement, Not a Replacement
On the question of AI displacing UX roles, Mitra drew on research suggesting that roles combining creativity, interpersonal skill, empathy, and strategic thinking are among the least likely to be fully automated. The more useful framing is AI as a support layer.
- AI handles volume-heavy and repetitive tasks, freeing practitioners to focus on interpretation, storytelling, and high-stakes decisions.
- Tools such as ChatGPT and Microsoft Copilot can accelerate brainstorming, content creation, and data synthesis.
- The combination of human judgment and AI throughput is where the productivity gains are.
Practical AI Use Cases in UX
Secondary Research
AI tools can benchmark competitor features, identify market trends, and pull together organized insights across multiple sources in a fraction of the time previously required. Mitra noted that exploratory secondary research that once took hours now takes roughly an hour with well-crafted prompts.
Persona Development
At Lucid, persona development that previously took weeks now takes a couple of days. AI provides a research-grounded starting point rather than a finished artifact. Mitra emphasized not copying AI-generated personas directly, but using them to accelerate the process.
Thematic Analysis of Interviews
Mitra's team uses Dovetail to transcribe interview recordings, extract themes, generate summaries, and link findings back to source video clips. This has meaningfully reduced the time needed to move from raw interviews to a written report.
Classifying Customer Feedback
Lucid receives large volumes of customer comments. An internal AI tool classifies that feedback by theme, surfacing the most critical issues without requiring manual review of tens of thousands of responses.
Survey and Interview Design
AI provides a structured starting point for discussion guides and survey instruments. It does not replace researcher judgment on question design but reduces the time to a usable first draft.
Prototyping and Concept Sketching
Figma's AI features reduce the manual work of connecting interaction flows. Tools such as Claude can generate front-end code snippets from wireframes or design files. Image generation tools such as DALL-E and Midjourney can produce concept sketches for evaluation studies when designers are unavailable, with appropriate caveats to participants that images are AI-generated and representational only.
Ideation
AI can generate feature ideas or UX elements quickly during ideation sessions, helping teams move past creative blocks. The outputs are a prompt for further thinking rather than final recommendations.
Best Practices for Using AI Tools
What to do
- Define clear objectives. Specific, detailed prompts produce better outputs than vague ones.
- Iterate on responses. Treat the interaction as a conversation, refining outputs across multiple exchanges. AI tools retain prior context within a session.
- Use AI as an enhancement only. These tools accelerate repetitive tasks and support ideation. They should not replace critical thinking or user empathy.
- Maintain quality control. Always review AI output for accuracy and alignment with project goals. Ask AI tools to cite sources during secondary research to reduce hallucination risk.
What to avoid
- Do not input proprietary or user data into public AI tools without confirming your company's policy. Some organizations have approved private or enterprise instances; others have not yet cleared any external tools.
- Do not over-rely on AI for critical design decisions. AI output can contain inaccuracies and unintended biases. The UX professional remains accountable for final decisions.
- Be aware of AI bias. Like any research method, AI models have inherent limitations and known biases. Understanding those limitations is part of responsible use.
The Role of UX in an AI-Driven Future
- Generalist UX professionals are well positioned to leverage AI as a collaborative partner rather than a tool to fear.
- UX practitioners have an opportunity to shape how AI is applied in product development, ensuring it serves user needs and ethical standards.
- Mitra's team at Lucid built an internal narrative around AI adoption, helping align the broader organization. Individual researchers can drive similar influence within their own companies, even where AI adoption is still cautious.
Transcript
Read the full transcript
all right thank you so much um hi everyone um um thank you so much for the acceler accelerent research team to give me an opportunity uh to speak to this um group on this webinar uh I'm going to talk about the ux generalist Advantage powered by AI um my name is pro shanit Mitra um but my colleagues and friends call me Pro that's easy to remember I'm I'm I work as a lead ux research and strategy at Lucid Motors um I work here to improve the user experience of the electric vehicle ecosystem and I live in the San Francisco Bay Area um a little bit about uh my team uh here at Lucid we are uh I'm part of a team that won the if design Gold Award for ux um uh for the said in vehicle experience um our Lucid customers will soon going to be able to experience this award winning user experience um in the upcoming Lucid gravity SUV which uh gets out next month um I have been working here for nearly 3 and a half years now and it feels like such a long time because this is a startup environment and so many things have changed uh before this I was working as a ux researcher at Toyota Research Institute which was also in the San Francisco Bay Area um where I was working on autonomous vehicles um yeah um so let me just dive into what the agenda is going to be like for this talk um and what I'm going to try to cover um here so uh I'll first start to talk about the value of Versatility in ux stores why am I advocating for um ux professionals to be U more of a generalist um and then I'm going to talk about how we can Embrace AI to expand our impact as ux professionals uh or inside professionals that for that matter um practic uh then we'll look at some practical applications of gen J gen AI um generative AI in user experience and uh I'll also talk about some best practices um and toward the end we'll also uh take some questions from the audience um so if you have any questions throughout the presentation uh I I think you uh you can um place the questions um at in the chat uh and we will address them after the presentation oh I'm seeing some questions already do they give you Lucid car to drive yes they they do give me Lucid car to drive from time to time as part of my job as a ux researcher all right um so let's dive into it uh I wanted to start uh on a somber note but this presentation is not going to be somber it is a hopeful presentation so uh let's talk about some uh insights around the ux job openings um so I during the pandemic especially in 2021 and 2022 um businesses adapted to pandemic challenges and there was a significant shift towards digital transformation what this resulted was um that there was a heightened need for ux professionals so a lot of hiring was done um but and by 2022 we sort of saw a peak in the demand for ux designers and researchers and it had nearly doubled uh since the prepandemic times uh reflecting the critical role of ux in digital product development however uh in 2023 um you know as we may have all sort of experien that the job market experienced a correction and after the rapid hiring spree of the previous years companies uh began to stabilize their workow force and the adjustment led to a decrease in US job posting for exam for instance the ux research job openings dropped from dropped by 73% from 2022 to 2023 however when you compare this to um 2018 um six years ago uh there was still a 53% increase in US research job postings uh indicating there's still long-term growth despite some short short-term fluctuation as of mid 2024 um the most recent reports suggest that the US job market is showing signs of resilience and growth product design roles have uh rebounded product management opportunities have surged and ux research positions are also coming back slowly um um so overall I feel that the Outlook remains positive with significant growth opportunities in the ux field um but we we I we will see how you know the demand for the ux professional is changing um so there is increasing demand for versatile skill sets uh I'm just looking at a few here but um this I've seen seen a trend across a lot of different um job openings across different companies for ux research roles or similar similar uh Insight professional roles uh so here's one from Reddit um so aside from from all all job all researcher job openings ask for the regular um you know whether you can whether you know different methodologies for doing qualitative or quantitative research they are they they want to look at what your processes uh are um they want to see what are what your strengths are in terms of you know different methodologies um they want to see how well you can communicate your insights and what kind of recommendations you make these are all common across most job openings but more and more we're seeing things which are um beyond the basic skill set for a ux researchers we're seeing um you know things like craft video summaries of interviews and experiences of users be a strategic partner to product and design influence stakeholders across different functions um lead idea and strategy discussions through research um stay up to date on industry Trend contribute to strategic planning um communicate results and illustrate suggestions in creative ways fuel ideation and and for some specific roles we're also seeing you know um requirements are rather nice to have around certain Technologies um they want you to know Unreal Engine or Unity um you know or have S experience with Gen AI or or immersive technology such as XR so what we see from here is that the role of the ux professional is expanding or or evolving rather um is is going Beyond just gathering and analyzing user insights uh to contributing across multiple areas such as strategy and design it's not just enough to just communicate your insights you would want to drive product decision making product roadmap through you know by by your contributions to strategy and design that's how researchers may be able to make more impact uh in in the coming years so what we're looking at is a is a t-shaped ux profession um a t-sh ux professional has deep expertise in research and this is what a t-sh professional looks like you're expected to have expertise in your research uh methodology um you know you you need to know all the common ux research methods but you're and you're also expected to have a strong process around research but your advantage really begins I feel uh when you are also able to contribute to um ux strategy perhaps or business strategy uh through Persona development customer Journey mapping product visioning design by conducting or helping conduct design thinking workshops um feature prioritization Frameworks as a strategist you are um the user um um means of communication um in a business world you are speaking the language of the business by understanding the business needs but you're also speaking the language of the user um by by inserting the you insights you have created or found through your research um you also might want to learn or might have to do um some certain ux design skills it's it might be good to know uh how to you know do some concept design work or sketching uh or maybe uh you know your designer is unavailable to do build a quick and dirty prototype and they all they've built is certain W wireframes um and you have some time to fill that Gap uh with and I've I've experienced this in uh in my uh previous roles as well where there was a need for have building cck clickable prototypes but there were there was no time or bandwidth for anyone to do that and in order to run that study that was the biggest blocker um so I started doing those things myself but of course you know these things take time um aside from knowing you know just just about all the ux functions uh you may also have an advantage if you um are able to understand how other functions that work with ux um you know how how how they work having some knowledge about the product management process about you know Engineers who frequently work with you may give you an uh better Insight or may may teach you to better approach them um and communicate your insights better to them um and U aside from knowing what you know best which is the the your area of expertise or specialization um you may gain Advantage if you know beyond your area of specialization across different Industries maybe or across different domains um this may not only help you um um you know quickly switch over um and adapt to a different situation but also maybe if you're speaking with someone from a different industry uh you may be able to communicate better with them or understand them better as a US researcher um so how do you strengthen your position by doing more um as I mentioned uh being the ability to work across functions enhances job security and makes you a valuable asset in any team uh when uh when ux professionals understand product strategy they can better advocate for user needs in ways that align with business goals uh giving research outcomes more influence um and with regard to proactive problem solving when you can take on ux design or contribute to strategic decisions you enable faster decision making create more cohesive experiences and add tangible value Beyond insights Alone um I want to talk a little bit about broadening how your domain knowledge um I feel um we can always start with understanding adjacent domains knowledge in areas like product management or uh data analytics make your insights more actionable and relevant um for cross functional collaboration with broader domain knowledge you can communicate effectively across different teams ensuring your insights don't just sit in reports but inform real decisions and you can have an enhanced empathy for users uh by understanding their Industries the industries that they come from the the backgrounds that they come from or the context Beyond surface level interactions it allows you to uh allows for richer insights and more effective Solutions uh but of course um I'm I I I said so many things that a ux researcher must do uh um but it it has its own set of challenges um in you know expanding your ux role being a generalist doesn't come without its challenges right it can it can be easy to spread yourself too thin losing depth in your core skills uh expanding skill sets requires time and mental load which can be hard to balance with day-to-day responsibilities however these challenges are manageable especially with help of tools and strategies to support your expanded role uh you uh if if there are younger professionals here uh who are listening to this talk um this means that you could seek out maybe mentorship um take advantage of training and development opportunity and building a supportive network of of ux professionals who can guide you um there's also a lot of talk or rather concerns about how AI will impact um job roles in ux or rather job security in ux um so there's a lot of concerns that you know people think there's a fear of AI replacing ux roles especially with increasing automation This concerns that you know there could be a reduced need for human cented roles but uh research suggests uh that um you know there are some studies that have been done um that suggest that combining creativity and interpersonal skills um such as the way we do in ux design and research are among the least likely to be fully automated uh rather what we can think of AI is uh more of a complement rather than um you know a replacement um for roles that you demand user empathy storytelling and strategic thinking AI serves as a supporting tool rather than a replacement uh what what it allows as ux professionals to do is um it allows us to focus on the higher level TRS and the heavy lifting work such as interpreting uh user needs um or um and and in designing meaningful experiences this is the stuff that we can focus on whereas a repetitive processes um or data um heavy work can be delegated to AI so there is a balance between uh the uh between Ai and the ux professional that can go hand inand to improve uh our work environment and improve our productivity that's what I'm going to talk about in the rest of the talk don't fear AI EMB it um AI can help speed up critical workflows such as brainstorming content creation and research it can help augment human capabilities and creativity not replace them um so uh I want to talk about how AI tools such as chat GPT or co-pilot help you do more with less um chat GPT can help brainstorm uh user scenarios suggest research questions and even uh assist with synthesizing data to streamline your process um how can you become an expert in uh a new field need to understand quick need to quickly understand a new field um these tools can provide instant insights helping you upskill without extensive reading or training um especially this this will come in handy when you're you trying to talk with someone who is from a different industry or trying to understand more about their background this this skill comes in very handy it can also help you in your prototyping and ideation um you know by creating or generating sample scripts user stories and even some design ideas um there's tools like Cloud uh. I'm not sure if I'm pronouncing that correct uh it can also assist with quot Snippets for interactive prototypes speeding up ideation um and you know you just have a design and uh ask the AI to build a code snippet that can create the front end prototype for you um and it also helps you with documentation and communication can help you draft concise summaries create alignment documents or even role play stakeholder questions um using chat GPD to ensure seamless Communications across teams uh there are some practical use cases for AI that directly impact our ux work um such as um synthesizing qualitative data which can be a lot of text um and may take ours or even days sometimes to extract themes and sentiments from you know 20 interviews that you've done um uh it can help you as I mentioned you know build code Snippets for your protypes can help you generate personas with minimal inputs from your side um so let's look at some of these use cases that I have personally been uh using um and I I I can recommend based on my experience um so the first one that I'm going to talk about is conducting secondary research so AI assists in secondary research in a few ways uh you may be able to B Benchmark your competitor features uh you can identify market trends you can uh organize your research data more efficiently or you you have so you know different secondary research sources you can create a a an organized Insight um across those secondary B socialist so here has an example I have from chat GPD what are the key ux features of top e-commerce platforms um this is just an example um I I and for uh confidentiality S I can't I can't um give examples from my own work but I have given similar examples from generic um Industries like for example um here it it it has given you know um all the different features uh that are key for having good user experience to enhance customer satisfaction and drive sales and now these prompts that we that go in gen geni uh portal such as chat gbt um they can be as creative or as specific as you want um so as that you're not spending a lot of time just looking for things that you need to look for like for exam for instance I used to spend hours looking for the right paper or right article to support um you know to do just exploratory research work now it takes me a few minutes um maybe an hour tops uh to just get to all the data that all the um secondary research data that I want to find just by putting in the right words on this portal um so I feel that this is really powerful for conducting secary research uh the next one I want to share is person Persona development um and uh this is also really sped up our process of um you know building personas uh I'm in the automotive industry and we coming up with different um you know car platforms um in the course in due course of time uh it is essential for us to understand what's the market for different products um which means personal development is key strategy piece um in not just um the ux team within Lucid but also uh you know our business in general um so uh we heavily rely on building personas that are um backed up by research as a research team developing personas um that that process used to take days um weeks even to come up with a a Persona that is grounded in real work real research work um now it doesn't take weeks or months it rather takes a couple days uh to get the right research in um and start building a pro Persona based on the accumulated research just by you know a few prompts on chat GP uh what I'm I'm not suggesting that you just copy paste this uh Persona from here and um get started and make your final Persona from chat GPT but this is a great starting point and really helps us uh progress much farther um than you know looking for all the research yourself uh and and then build on that research uh create those insights and build on that um the next um use case I'm uh I want to share is the Thematic analysis of interviews AI can summarize user interviews by identifying common themes and sentiments Saving Time on manual analysis this has really been helpful for us as a team we have been using a platform called dovetail um and this is not a plug really it's um just a share out of this tool that is available on the market that what it does is um it's able to you know all the video all the interview videos that you upload for example on this platform it's able to create uh transcript uh transcript from those videos and then um it it's a features and um create summaries as well as you know create extract HS from um these uh transcripts and once uh you create those report using those themes on the platform you can tie back to the videos um all all of this happens pretty SE seamlessly and it has really improved the way we write our reports um and it has really that process of getting to the inside uh really quickly um another use case that I feel is really helpful is summarizing and classifying customer feedback um so tools like chat GPT can classify large volumes of user feedback by theme allowing quick identification of user sentiment and priorities uh I just put here an example a generic example of um a generic um um data set but uh we do this uh internally as well um with our customer feedback that we get from our customers which which has thousands and thousands of lines of thousands and thousands of comments um and this uh chat GPT or similar products um we have our own internal product here at Lucid uh which can look at this data read it and classify it uh making it much easier to digest and highlight you know what are the most important issues for our customers without having to go through 10 tens of thousands of comments okay um the next next one is designing surveys and interviews um it it kind of seems like I'm bordering on um you know all the tasks that a user researcher does but again I have a caveat here uh AI can only help brainstorm and structure questions it can help you get to a starting point um but I I'm not suggesting that you know this is AI can do the entire job of a ux researcher but you can really speeden your process of you know getting to this point uh if you if you have a head start and that's that's what it what it will really that's where it saves your time to be honest provides a head start you need uh as I previously mentioned there are tools uh such as figma which with its new AI features it's uh letting you professional add interactivity to its prototype without extensive coding knowledge or spending a lot of time connecting the strings if anyone here has tried to prototype using figma or similar tools you may have realized that it takes such a long time to just you know draw those strings even if it um a simple looking job to do um it it can get pretty complex uh and getting a head start um using AI based uh prototyping can really help and um then there are tools such as cloud. that I previously mentioned that can help uh you that can look at a prototype or that can look at a design rather or C some wireframes and and provide you with um codee Snippets that you can use to build the front end based on just visuals um AI can also help you brainstorm new ideas can generate feature ideas or ux elements for quick ideation sessions uh helping overcome creative blocks so again this is all about getting a quick start um you know generate here's an example generate 10 10 feature ideas for a meditation app and it'll give you a few um ideas and not all of them are maybe Super Creative but sometimes when you have a block um these ideas can help you you can you can use these ideas to build on them to get to some place which might be really useful or creative you can also you know tailor these prompts to be more specific and you know um specific around certain requirements and it will still give you um you know um some ideas that will be around that and uh another important one just related to that is you know if you have you have these different ideas different concepts that you're coming up with um or recommending uh the next uh obvious process is to sketch those out and sometimes our designers may not have the bandwidth to sketch out A New Concept idea that you may be testing and I this myself in my current role uh where we uh when I'm about to do a concept evaluation I do have all the concepts but I don't have any concept sketches that's where I've used the help of image generation tools such as D which is also part of open AI or mid Journey um which can help you you know uh do some concept sketching it's not accurate it's not uh perfect at the moment uh but with a few few iterations you can get to a stage where it it it is visually representative of what you want to show your participant uh and of course with some caveat you can um you can you can caveat your surveys or questionnaires uh that you know all the imagery that you're seeing are AI generated and only representative in nature they are not part of final design helps you uh you know break the ice quickly and get to um the process of doing research rather than waiting on concept sketches from someone else so yeah those are some of the use cases that I wanted to share um and uh I will talk about some best practices for using AI tools effectively um some dos uh are um that you you know as these tools can be incredibly powerful asset for ux professionals um U here are some best practices um let's start with defining clear objectives be specific in what you ask uh the more detail your prompt the better the output from uh tools like chat GPT um iterate on your responses treat your interaction with chat GPT like uh a conversation refine the responses item atively uh until you get the quality and relevance you need um and chat GPT and similar tools they are they understand context so you know something that you've said previously uh it it knows that you already said it and then you can you know build on that uh use AI as an enhancement only AI tools are here to augment your expert expertise not replace it uh they're excellent for speeding up repetitive tasks and supporting iation but shouldn't replace critical thinking or user empathy and uh lastly maintain quality control super important uh remember that AI output is not perfect um always review and ensure that whatever AI generated content you want to use it aligns with your project goals or is is grounded in reality sometimes AI has this habit of hallucinations uh and one way I've realized uh one way to avoid that is to ask ask uh these AI generated tools um for reputable citations while doing your secondary research so it can you know point to the source that it is um you know citing a particular Insight from and uh lastly some don'ts um there are also several key things to avoid when using AI in ux uh avoid using proprietary data with external tools uh to protect use user privacy and prevent data breaches um and I want to talk a little bit more about that um different companies may have different approaches uh with with AI um some companies may be strongly for using AI related tools uh they have embraced that AI is the future While others are still skeptical or on the friends about this um wherever you're working um make sure that you are aligning with the company goals in terms of uh you know what is the company policy around using these AI tools um and which means that in certain cases you may not be able to use uh use or input proprietary data into these tools in other cases you may be able to do it where the company has um you know vetted a particular AI tool and you know they are uh they have a private instance of your um and the data is secured if if the company already knows that and if you already know that maybe you can use those tools more um and use these um less private or rather public tools such as open AI chat gbt uh for more generic um searches for example for ux researchers if you're uh trying to learn about a new domain um these open AI related open AI tools are very effective doesn't you don't have to really tell about give any confidential information from your com company on that portal but rather learn from The Tool uh or or it help it's helping you build um quickly build a a template for a research um study that's that's something that you can also perhaps do even even if your companies has not embraced these tools uh yet um and lastly don't over rely on AI for critical design decisions um avoid relying too much on AI for critical design decision as this it can perhaps lead to inaccuracies and unint unintended consequences that's why we have us the ux professional uh that's uh in the middle making those critical decisions uh at the end because it's uh it's on you um and you you can't just say say that the AI said so so whatever you're doing make make sure that you're using it as a support tool only and and only to you know maybe speeding up your process but not necessarily to make decisions for you um lastly we are in a world um we're headed toward a future where um ux is going to be in an AI driven World um and our seal is going to change with it ux generalists will be well positioned to leverage AI as a collaborative partner uh ux professionals have the opportunity to influence how AI is used in design uh ensuring it serves user needs and ethics uh ethical considerations around AI also are also becoming more crucial as we incorporate these tools into our work uh we must remain aware of our biases uh of biases or limitations in AI outputs we've seen certain examples and news articles um across the internet where uh a lot of of these AI models U have certain bies so it's good to be aware of these bies while using them um you know so that you can like any other research method this is also this has it inherent B and it's good to know what they are all right uh that brings us to the end of the presentation uh I want to end on um just three things again uh in conclusion we've covered some best practices for AI tools effectively um explored the future of ux in an AI driven world and emphasize the importance of staying adaptable and curious uh here are the key takeaways Embrace a generalist approach expanding your role Beyond traditional ux research empowers you to contribute more deeply across the product's life cycle use AI tools to amplify your impact uh AI tools like chat GPT allow you to handle more with sa without sacrificing quality bring you to focus on higher quality work uh be adaptive and stay curious the ux field is evolving and so should we uh cultivate a mindset of learning and adaptability to stay valuable as the field advances um do more with less right um thank you so much for your attention and I will open up the floor for any questions um I can go over some questions we got throughout the presentation so the first one's from Janna she was asking if uh anyone is using AI to get to requirement she says she's found it hard to do well when finding larger themes or Doo yeah do well when finding larger themes but if falls short when getting to specifics and requirements I I'm not entirely sure if I understand um the meaning of requirements here um I I I have personally seen uh used AI for I I I'll try to answer how I have used AI here um our requirements for the product okay yeah I I I I feel uh requirements for the product is the process of getting to the requirements is still similar um I I feel that you are always driven by getting to the in uh Insight first and recommendations based on those insights uh and the process of getting to the Insight is still pretty similar to what we typically do um uh AI can help you create those summaries pretty well sometimes it will extract certain themes but it's not always perfect um when you're you know um asking it to extract the the themes from a bunch of uh text um I feel that that technology is still very nent um and it it requires a lot of human in introspect uh introspection uh and intervention um to get to the Insight uh but it does help you get to the inside Faster by creating summaries of uh what you long bodies of text um the next question we had coming was what level of security do you look for platform do you look for with platforms like dovetail where your research is uploaded for its analysis and Magic um so um we uh so I for for me as a ux professional I um let the IT team and the infosec team in my company um handle these questions uh so whatever their policy is in terms of um you know how do we uh how how do these third party vendors are managing our data uh our data meaning you know company's data um if the infos team and the IT team if they are satisfied with their answers um that's good enough for us to be using these products um so that's that's I defer to those teams essentially uh and I I particularly may not be able to answer what exact levels of security um Day May okay um another one going off of dovetail are there any other AI enabled tools you're using regular regularly where they are inhouse or closed unlike public ones um I I think more and more of these platforms are um I there are there are two types of uh vendors here one is uh traditional research tools such as dovetail and uh perhaps you know uh tools such as Meo they are or Lucid spark for example they are embracing AI features into their ecosystem and enhancing their features um so these these uh you're probably using all of these tools already um and then there are there is another a different set of um AI specific companies which are now coming up including open AI uh open AI has their own enterpr Enterprise um version which I am I'm I'm not entirely sure what their security looks like uh but I I feel that you know if it should go through the regular process of your company's it and infosec protocols to make sure that they are all compliant with them so I believe that a lot of these uh newer companies the AI focused companies they are also you know building up those infrastructures to serve uh larger clients uh Enterprise clients with more um secure platforms okay um that's all I see online but I actually had a question um so you mentioned different use cases for AI when you're using Ai and those have you come across challenges with that and how did you overcome those challenges um I did come across several challenges across different use cases I would say um one of the most common challenges uh is around image Generation Um AI doesn't really fully understand uh what you want out of it often or U you know Mis messes up the details quite a lot uh but you know the these tools are also progressing they're getting better by the day and I've seen that uh these tools have evolved in a very short time right like uh last year this time it was it was pretty messy but now there are features within the tools where you can iterate uh pretty quickly and specifically to a particular ular area for example if there's an image and only one part of the image is not looking good or it doesn't make sense or it's gibberish you can highlight that area and you know iterate on that area ask it to focus or defocus on it so that's how I have met uh I have tackled the image Generation Challenges I've also had challenges as I previously mentioned around some of the Thematic analysis pieces and for those uh I have deferred to as of now um manual processes and uh relied more on AI summaries for my reports so for qualitative analysis I um use AI based summaries a lot and don't use schematic analysis as much so that's where uh I hope that the progress is made by tools such as dovetail which we are currently using and um yeah hopefully uh it'll be it'll be better in the coming weeks and months okay and as AI progresses how can like individuals make themselves more competitive with that changing environment as researchers as a whole and then as employees to companies yeah I mean um as I mentioned throughout this presentation uh AI is uh not here to replace you uh um this there is always going to be a need for that creative professional who um for us uniquely we stand on the precipice of um professionals who are both creative and scientific in some way um as well as empathetic so these combination these these this combination is really critical for us to continue emphasizing on and that human eye will always be needed to um uh you know handle these tools better uh and give the right direction of um you know finding the finding the appropriate data um finding the appropriate insight and what you do with the Insight is still your uh it's a it's a human thing it's going to be dependent on heavily on the ux professional going forward so um focusing on my my advice to someone who is thinking in that direction would be focus on having your high level skills strong um and you know make L make larger impact um through you know using these tools by doing more with less um and lastly I want to talk about just um how you can I I just gave a few use cases with regard to uh how you can impact your ux work uh with uh with with these AI tools but as a ux professional you can also make impact within the company if your company is still on the fence um of using AI maybe skeptic uh you can perhaps try and understand how uh your company can benefit um across different domains uh in the comp within the company uh how they can benefit through AI based on their use cases it's something we did our research team did at Lucid um and we really were able to bring together the idea of uh Ai and have a consistent narrative within the company and this is how you can make a larger impact um within the company as well uh for driving this AI Revolution um yeah that's what I would say okay thank you um I'm not seeing any more questions come in but I do want to thank you for your time AI is a huge conversation now and I only anticipate it becoming a bigger conversation so thank you for your insights and making time for this all right thank you so much it was great speaking to this audience thank you
