My slides. So so let's kick it off. My name is Ibanichirich, and I lead, product strategy and delivery at Thoughtworks.Thoughtworks is a global consultancy that integrates design, engineering, and AI, and we enable our clients to thrive in times of great changes such as right now. And so everyone I know who's building digital products today feels overwhelmed by change. There are new tools to use. There are new ways to build, and there are just so many AI training courses that you don't know where to start, and it's definitely very difficult to keep up. So I'm curious if you can drop in the chat, what are your biggest challenges with using AI for building products? Let me know, and we'll continue to talk about some that I've seen in my experience.

So today, what we'll do is, and we're going to, take a look at, going through what you need to know about AI tools, whether you're product designer, whether you're a product manager, an engineer, and or really any other role in product development. We're going to cover some key skills that product teams need to have in order to build in this new AI enabled age. We'll cover what you need to know about making trade offs when using AI tool, and we'll cover how our ways of working are now going to shift. So to stay so that you can stay competitive as AI becomes more and more integrated into our product development and really our lives. So before we dive in, there are two different ways you can think of generative AI tools in product development. The first is as a capability. So an AI capability that you can build into your product for your users. And the second is as a tool.

That's a tool that you can use for product development, to make your to build your product faster, to improve your process. And in the latter, currently, GenAI has shown immense promise, improving productivity, improving performance as much as 66% in some cases, and this is just the beginning. But along with massive leaps in productivity, this AI revolution brings its own share of challenges as well. So today's focus is going to be on specifically AI tools for product developers. And to really understand the impact that AI tools have, let's start with the fundamentals of how we build product. So this part has not changed. There are roughly two key areas that product folks spend their time on. The first is discovering. So building the right thing.

Making sure that we understand our customer, understand their problem, we understand the market, and we define the best product to build. The second aspect is delivery. Actually executing on the product vision, that North Star that we created, and really dragging towards benefiting our customers and also benefiting our business. Both of these aspects are done continuously, and we have these six key areas that we focus on. Within these six key areas, and this is something that we call the product thinking playbook at Thoughtworks. So each section of the playbook contains various tactics and techniques that product teams use in order to build and, of course, to discover products that they're building. Now we're not going to talk about AI assisted coding today, but there have been huge advances in this area.

However, if you look at the rest of product development, there are AI tools or agents for almost every other product development task within the life cycle. So huge gains in AI assisted coding, but there are many, many tools for all of the other aspects of delivering great product. Just to illustrate a few, we have products such as synthetic users, which allow you to test your ideas with AI users. We have ProdPad that helps you define and refine your product vision, which is a foundation really for your product with Northstar. Product board, user doc, lovable is a great new product that, where you can create entire products just with a prompt. And then, of course, Amplitude as we work into, looking at analyzing how our products are doing. Some of these are started off as AI products.

Other ones started off without any generative AI and build AI features into their product to help you with various aspects of the product development life cycle. And if you can't find a tool for the specific task that you want to perform, you can always use general purpose LLMs such as ChatGPD, Gemini, Cloud, or many others that exist. So really across the entire product development life cycle, you have many options to use various AI tools that some have been around for quite a while, others are just getting started. And, again, the landscape is massive. So I'm curious in the chat and just throw it out there. Is there a go to generative AI tool that you use in your product development life cycle? And tell us what you use it for because there are many ways to use an LLM.

What if it's chat g p t, what's your favorite way to use it while you're building products? And, of course, we know that many of these tools are available. But just because we can use them doesn't necessarily mean that we should. And to understand why, we're gonna go back in time to draw some parallels between, Gen AI tooling and some earlier innovations. I see some great chatter, perplexity, cloud, mid journey. So some LLMs, some image generation. Let us know, and and keep sharing. So I thought email would be a really great, great innovation to compare to when we talk about, Gen AI and tooling and and benefit. So this is a direct quote that we thought when email first came out, it would reduce paperwork. It would make us so much more effective.

We'd have all the latest up to date information. And and the best one for me is decision making would be easier. That's all in theory. But we know in practice what really happened. Don't worry about digesting this entire diagram. I'll I'll walk you through it. And so as we started adopting email and we were able to write more quickly and easily. And then what do you think happened? All of these emails from all of these people who can now more easily write to you started to pile up. And today, any guesses on how much time we we spend on email and and messaging? Some sources say about 29%, of our workday. So huge, huge amount of time spent on something that isn't a core for the for the majority of us isn't a core of our, responsibilities.

And despite all of these benefits, decision making is still hard today. So what this diagram what I want you to take away from this is that while new tools can create efficiencies, they also dramatically impact the entire ecosystem around them, in and sometimes cause unintended consequences that down that that reduce some of the benefits that we see from them.

If you understand the big picture, when it comes to any new innovations, you'll adapt more quickly and you'll achieve better results in the long term. So, again, I'm curious. Is there a tool that you've used that failed to live up to its promise? So we'll give the example of email. Decision making is still hard. We're not really able to do it any better than before email. Or perhaps it did, but there were some unintended consequences. So let us know in the chat, and we'll move on. So I really wanna focus the rest of our conversation on, cutting through the hype and focusing on what you as a product builder can do today, ignoring what advancements may come in the future. And those are the three areas of ramping up on key skill sets that you need in this time.

Looking at negotiating the different trade offs with using AI for product and how to transform your ways of working, again, based on what currently is happening, not something that that is promised by any of these tools. And some great comments in the chat, which if we have time, we'll get to. If not, I'll, I'll stay a little bit extra and respond. So critical skills. The first one is specialization. So as you've probably already seen, there are situations and use cases where AI has outperformed human at specific tasks. So, things like image recognition, speech recognition, and there are benchmarks out there that continue to be surpassed and achieved by AI tools on a regular basis. Very, very difficult to keep up with all of the advancement. But research shows us that the people who get the most benefit out of using AI tools are those of us who are more skilled to begin with, those who have specialized skills.

This tracks with if you've been following the job market today, some of you may be looking for work. The postings that exist right now is employers are looking for specialized skills. They want skills in a particular industry. They want skills in a particular tech stack or type of product. So we're leaning into specialization. Another example is industry skills. Automotive, airline, oil and gas, some of the industries that we work with. Without so so these industries are all in need of AI tooling. But if you don't understand the industry, you can't build AI or use AI for it. So my advice to everyone is focus in and immerse yourself in the real problems and indeed that exist in these industries before you try to tackle using AI in them.

Next key as critical skill and and is currently underrated is communication skills. We don't talk about them enough, but when your main way of interacting with AI is through prompting, then you're going to need to be really good at expressing what you need. Lovable is an app I mentioned. If you can build an entire product using a simple prompt, you need to make sure that you're able to articulate exactly what you need, and this will continue to be important throughout. And the last one I'll touch on is information or data. I'm currently working with a client who's on a mission to organize, model, and serve up all of the data in their organization.

And the key driver for this is to be able to use AI tools, the latest AI technology to gain competitive advantage over, all of their competitors using their proprietary data. So what this tells you is to realize the promise of AI, you need to start with understanding and organizing your data. So learn. If you're not already doing this, become as data literate as you can. So we talked a little bit about critical skills. Let's talk about trade off. So as you can imagine and you've seen in practice, we can now generate way more content, more prototypes, more copy, than ever before faster than before. So similar to the rise of email, who's actually going to look at all of this content? And will it help us build a better product? So I love this quote by Yuval Noah Harari. Power today is really knowing what to ignore.

So, looking at signal versus noise, it's it's being able to think critically and cut through all of the noise that exist, when people are are using AI to generate to build products and to generate content. The second trade off that you'll need to get good at is we already know that AI significantly negatively impacts decision making of those who use it. So when you're thinking about, efficiency using AI tools versus depth, you have to learn how to balance the two. So an example is looking at maybe synthesizing all of your user research that you're doing with customer, using AI. You're losing the ability to your trading speed and scale, perhaps you can do more interviews, but you're losing your ability to build your intuition, to build your product sense. So I would encourage you to keep doing some things manually and take time to really absorb, especially when it comes to your customer.

Not efficient, but it is suited to our very human brain. And one more trade off that I'll quickly mention is AI tools are not the best solution for everything. Seems to be everywhere, but, I'll give you a quick example of Klarna, a Swedish fintech. They've been in the news for replacing humans with AI for customer support and also for reversing that decision. And I'll quote them quickly. In a world of automation, nothing is more valuable than a really great human interaction. That's why we're doubling down, investing in the human side of service, empathy, expertise, and real conversation. And, of course, there are many nuances to each of these guidelines, So review them as tools continue to improve. This is just directional.

The last bit that we'll focus on is, changing our ways of working. So I will just wrap up by saying the organizations that have the highest performance are the ones that have been able to integrate both human and technology and and more broadly, AI tools into their workflows. Humans aren't going away, but some of our ways of working will. So what does this mean for product leaders? I want you to think about looking beyond savings and productivity. AI, if you're only using AI to make yourself more efficient, you're missing the point. What are some things that you can do now that you haven't done before? Things like rapid prototyping and talking through examples rather than just in theory, aligning more with your cross functional teams earlier on in the life cycle.

So these are things that become, possible now that weren't possible before. And then looking at evaluating result, not just through traditional QA, but also taking into account the nondeterministic nature of, AI solution. Again, they may not be the best solution for everything, but we need to learn how to evaluate them so that we can deploy the best customer solutions possible. And then the last one is rules are merging on traditional teams. The traditional product managers, designers, engineers, data scientists, data engineers, all of these different product roles are broadening and merging into each other. So you may be required to be technology literate much more so now as a designer. Designers are building full on product.

So keep track of these trends and make sure that you're looking at what are the demands in the industry of your specific role and how do you level up in some of the skill sets? So where does this all leave us? It's very clear that it's critical for product builders to use AI to prepare for the future, and that these tools are continuing to evolve. So as a result, it's really hard to keep up keep up. But some things are not changing, like the overall flow of discovery and delivery, specialized skills, communication. Those things continue to be important as we transition into a new era. So I would encourage you to remember these three things in order to deliver the best product possible for your customers and for your business.

Remember that the best outcomes with AI are when you already have specialized skills and you're an expert in your area. Remember that great products come from an ability to make the right trade off. No different for AI tools and their use and building in AI features into your product. So be intentional about when you use it and when you choose not to and what some of those trade offs are. And then, of course, experiment now with new ways of working so that you can be ready when, as our ways of working are shifting. So I wanna thank you all for joining me, for for listening, and for sharing your experiences. And if you'd like to learn more about some of the work that we do in the AI space, you'd like to download our product thinking playbook.

Feel free to scan the code, get in touch with me. I would love to continue this conversation, and connect outside. Enjoy the rest of your conference, and thank you so much.