Stansons_’s Post

Transcript

The other interesting 1 is about agents and what we call agents and agentic frameworks. So if you think of a model like Gemini, it's our family of large language models within Google. The most basic usage of that is that you ask it a question, it gives you an answer and the answer gives is based on the training data that it has seen the training time, right? And these models have been trained on millions and millions of data and that's why they are. O smart, let's say agents are models that actually interact with other systems. O imagine right now you ask Gemini a question, it gives you an answer. That's it. You use your answer to do other things, clearly. So it's very helpful in that sense. But agents would mean you can actually ask it to do something for you. For example, based on the question you gave me, send an e-mail. Or add something to my calendar at the memo interacts with all of the other tools that you would use on your day-to-day basis. O this idea of agents. Actually, it's the real augmentation of it's no longer just something that we brainstorm with. For example, that's how I use LMS a lot. It's like I use it to brainstorm ideas to search for things within a repository of knowledge and knowledge repo. And the next step is really, now that you gave me an answer, I actually want to do something about it. Send an e-mail to my colleague. Remind me to do this later on so yeah, I think these two there are many more but these two I think are very interesting and worse thinking about yeah, I love that sort of setting the bigger picture there and many of the actually is already touching the design industry. But really I think if you think about it this way, I like to kind of set the timer for most people to have kind of been exposed to AI as when ChatGPT launched or when they saw Dali or mid journey image generation tools and really I think that. Has set the that sort of set the timeline for people to go ohh like ohh, AI is really a thing and it's really here. But as we as we know AI has been around for arguably decades and and really it's it's it's because of the consumer level kind of impact that ChatGPT had. We really saw it take take take hold of the world and everyone's attention. So for in the design world, I would almost bifurcate there's there's two real interesting applications and one which is getting. Both are getting a lot of attention, but one definitely seems to have be having a lot more, you could say aggressive development a lot more. You know, there's a lot more development going on and that is in image generation technology. Now you have image generation technology, but you also have, as Lucrezia pointed out there with LLM technology, right, which is, so think of it simply images or creating text. Creating text is, is insane because that's information and knowledge. And so you can create knowledge and that's another. Crazy thing because if you think about what ideation means, ideation as designers and there's a lot of designers in this room and maybe listening, but ideation is the ability to come up with novel ideas, right? And we haven't really seen that being possible before generative AI. But now you're in a position where you you can ideas can be generated that somewhat external to you, but still there's a level of integrated to what you wanted it to do in the 1st place. So there's this massive almost tension of agency like. If we use this tool, does it really, do the ideas really belong to us? Can we really claim agency over that? And so image generation tools are really interesting as well because as designers were visually minded and when you want to come up with a concept, you can now simply type that in. Or what Gensler has really been pushing is what we call the hybrid process, where it's not just about typing stuff in and getting outcomes, but it's about one, creating and using the tool that is that we've embedded in. Information into and we can have a lot more control over but what happens if the AI process starts with the human sketch or starts with a human collage or a model or information so that idea of integrating human workflows within the AI process is certainly what what we're interested in when it comes to generative AI 11 aspect of generative AI. But obviously Productivity Tools are really being craved for in in the industry and. Things that just make organization easier. Is there a creative change within Gensler on the process integrating with AI or is it still early days? So we were very, we were early adopters, you could say an early experimenting certainly industry wide, I would say. And that's allowed us to be proactive now because as the as the. It's no longer a trend as this new era kind of deepens and our people becoming more and more cognizant of of the importance of of integrating AI in an intentional way. We had already taken some very key steps and and people ask people have asked me like, you know, why images, what you know, I can do so much far more than you know what about data data. Look, ultimately data is the key to everything in this essential in one way and we certainly. You know, we have a very robust strategy when it comes to data implementation and strategy and that will definitely form the basis of how these tools evolve and how we implement them in intentional ways. But really if you think about it with the power of designs being generated in a visually, you know, with image generation tools with that power put in everyone's hands, which it is, it kind of raises a bit of a question of, well, what then gives us as designers? A unique edge if everyone and anyone can create these beautiful designs that are on there that have been sourced from all across the world like how do we use it intentionally? I think so that's why we paid a lot of close attention to using this in a very intentional way and so we have already made. A lot of a lot of I would say we're, we're, we're, we're the furthest in the industry. OK, nice.