Yasu Sasaki of Dentsu
Are you aiming high enough?
"FEARLESS CREATIVE LEADERSHIP" PODCAST - TRANSCRIPT
Episode 257: Yasu Sasaki
Here's a question. Are you aiming high enough?
I'm Charles Day. I work with creative and innovative companies. I'm asked to help their leaders discover what they're capable of, and then to maximize their impact. Helping them to unlock their own creativity, as well as the creativity of the people around them.
Welcome to the intersection of strategy and humanity.
This episode is the fourth in a series of conversations that I'm having in partnership with the Cannes Lions Festival of Creativity.
For the weeks leading up to Cannes, we're focusing our study of leadership through a single lens. The impact of Artificial Intelligence on the Creative Industries.
Are we moving fast enough? Are we going far enough? Is this an opportunity to fundamentally redesign the creative industries, or should we adjust and iterate slowly and carefully? Do we follow the puck or skate to where it's going? There are opportunities and risks around every corner.
This episode is a conversation with Yasu Sasaki. He's the Global Chief Creative Officer of Dentsu. His company has a presence in over 145 countries and regions, and they've been working with artificial intelligence, in one form or another, since 2011.
One of the main questions facing the creative industries is which companies will see AI as an opportunity to stretch the boundaries of human creativity. To recognize that the limitations in our creativity are no longer practical, financial, or time-based, but are, in fact, our own confidence in our capacity to think originally.
We can also use this opportunity to create greater efficiency, a worthy ambition in a world in which resources are scarce, and a life in which the time available to us is finite.
In theory and in practice, we can both raise the bar and lower the cost.
But we must be careful how we do it.
“They are reactive, I think. Because some clients ask us to just provide a storyboard and they can customize using AI for their product. So they think it's efficient and cheaper to do that. But as you can imagine, the solution will, I think, become weak using that way.”
At the end of the series, I'll offer some thoughts on what we've heard and learned, and where we might go from here.
In the meantime, thanks for joining us.
Here's Yasu Sasaki.
Charles (02:39):
Yasu, welcome to Fearless. Thank you so much for joining me on the show.
Yasu Sasaki (02:43):
Thank you very much. It's a great honor to talk with you. Thank you very much.
Charles (02:47):
What's your relationship with AI? How do you see AI?
Yasu Sasaki (02:52):
Yeah, actually back more than 30 years ago, I actually, I studied computer science before entering the company. And at the time, this was a kind of second generation AI.
And I studied about the programming and AI. But at the time, of course, I didn't imagine that AI is connected to creativity. Now, finally, AI comes to our industry again as a creative partner.
Charles (03:22):
So I think you're the first person I've talked to so far that's described it as a partner.
A lot of people still refer to it as a tool, but you see it as more than that. You see it as an ally?
Yasu Sasaki (03:33):
Yes. So, of course, some people are talking about AI as an enemy or a tool. But, we now understood that AI can be a strong partner for us as creatives.
Charles (03:50):
And how does that manifest when it's at its best?
Yasu Sasaki (03:54):
Yeah. For example, we have some cases like scrolling selfie before EuroPharma to aid Parkinson's patients. So it actually won the Pharma Grand Prix last year in Cannes. So we utilize AI to help the patient. And so together with our idea, and AI can be a powerful partner to help those people.
So we, of course, we learn a lot from AI and we also put a lot of information to AI to make them stronger. So it's a kind of a two way collaboration between AI and human to create a great output together.
Charles (04:39):
And do you find when you're providing guidance to the people that work for you, that you have to encourage them to that point of view? Do you have to give them awareness of the fact that this is not simply a tool? This is not a threat, that this can really help us? How do you approach the mentorship, the guidance, the leadership of the people around you, in terms of their relationship with AI?
Yasu Sasaki (05:03):
It depends on the level of understanding about AI. And actually, Dentsu has a lot of, how can I say, AI experts, yes, who are creative but know much about AI, because of the science background or because of technologists, creative technologists who can understand technology. I'm encouraging those kind of people to use AI more to create a great idea together.
But at the same time, of course, there is some people who don't much about AI. So in this case, we provided a guideline for ethical use of AI. So we actually have a guideline for that, for AI, for internally, how to use AI in an ethical way. And also we are making a guideline for the creative industry to share that together, as an industry standard to utilize AI in a proper way.
Charles (06:05):
So AI is interwoven into your organization, it is threaded through the organization, and it is also built as a separate capability. So you have an area of expertise around AI.
Yasu Sasaki (06:17):
Yes, that's correct. So we're using AI in a, yeah, different way. I can say we have two directions. And one is to increase marketing effectiveness, and the other is to extend creativity. Formalize including the automatic generation of advertising, like discovering insight, generating copy, visuals, predicting the advertising effectiveness, or customize the expression for each individual.
So it's a kind of a marketing tool. But the other is more, and kind to create views of AI and create a new solution to solve the bigger issue around the world into using AI and working together with AI. So there's some direct, defined direction for using AI.
Charles (07:10):
From your perspective, how does AI impact human creativity? And how are we going to draw the line between human creativity, and I guess what we might describe as AI reflection of us?
Yasu Sasaki (07:22):
Yes. So, AI vision is that people and AI enhance each other. And AI is enhanced by people, and people are enhanced by AI.
And there's a change in the way people and AI work in harmony. I think this will greatly improve both efficiency and creativity. And for human, is very important to utilize AI. So not for efficiency, not for the cost or economical reason, but if we can utilize for human, then I think we and AI can co-work together
Charles (08:03):
Do you see AI expanding human creativity?
Yasu Sasaki (08:08):
Yeah, I think so. So for example thinking idea is a kind of, I can say, lonely process, to think about the great idea, just one creative. But if we can utilize AI as a partner, then I think a person and AI can interchange idea, and the idea will, of course, make us stronger together.
Charles (08:34):
Can you see a future in which we can't distinguish between human creativity and AI generated ideas?
Yasu Sasaki (08:42):
Yes, for example, if you use AI just for efficiency, or get more clicks, or get more people interested in the short term, then I think the result is similar or more boring, because AI will just put the right answer or average answer, because of the data they collected. But if we only focus on the human way and emotional way, and use AI to strengthen those thinking, then I think output will be really innovative.
Charles (09:22):
Yeah, I mean that feels right to me, too. I'm interested in all of these conversations that I've been having, that there has been real conviction from some of the people that I've talked to, that we will always be able to tell human creativity and human originated ideas from AI generated ideas, content, output.
I don't think we even have the language yet, to talk about this. It's a hard thing to talk about, because we want to talk about creativity, but by definition, AI is not creative. It is simply reflecting us back. But I have talked to a lot of people who think that we will always be able to distinguish that if something is made by a human, we'll be able to tell.
And I'm just not sure that's going to be true. I'd love to get your thoughts on this question, which is, kids who are growing up today, it's widely expected, will have AI generated friends. And those friends in many cases will feel more meaningful, because they will actually at least provide the artifice of understanding them better, and caring about them more, and being more empathetic.
And so it's not very hard to imagine a world in which the next generation teenagers grow into young adults being completely comfortable, and in fact in some cases preferring, their AI generated relationships. And so, therefore, they see the world through that lens. And their idea of beauty or the things that move them will be at least significantly informed by those relationships.
Obviously we don't know if it's going to be true, but could you conceive a world in which we see the creative output generated by AI as being of equal emotional significance to human beings, as human content?
Yasu Sasaki (11:12):
Technically possible, I think. But yeah, AI is just learning from the data, actually. And if people is watching the same content, of course, people will be bored with that content.
Human needs are more different and surprising and something that, we cannot imagine that. So I think at the beginning, I think it is okay to work with AI. So if AI can give a lot of information, that is very surprising and interesting, but it's the same as the human. If we are only talking with the same person, maybe someday, maybe people will be bored and not interested about that. But human is more than that, I think. So meeting with different people will make you more stimulated, and have a different feeling, but maybe AI cannot do that in this current technology. Maybe it's possible to simulate the human encounter, but currently it's not possible.
Charles (12:19):
So the unexpected and the original, which humans uniquely can provide, will be a determining factor, at least a distinguishing factor, as far as you can see.
Yasu Sasaki (12:29):
I think so. So actually Dentsu has a product of conversational partner using stuffed animals, and we put AI inside the animals, and AI can be a conversational partner for mainly for elder people. It's okay for them to have a general conversation, and it's easy to talk, and the stuffed animal will listen to him or her, and answer in a good way.
But, think about the young people, children, they need a more defined point of view, so not just listen to, or not just giving a comfortable answer, sometimes need a strong defined opinion or very different point of view. But AI itself doesn't have those kind of intentional will to do that.
So it depend on who is using AI or why they're using AI. So it's good as a, just comfortable conversation partner, but I'm not sure it was like educational or, how can I say, emotional training or something like that.
Charles (13:36):
Do you think the industries that you and I spend our time in, do you think that they are moving fast enough to meet this opportunity, moment, threat, however it's perceived?
I know Dentsu has made big investments into AI, and you mentioned the fact that you have AI expertise threaded through the organization. But human beings have a tough time with understanding exponential speed and exponential growth. I made this point at the opening of one of the earlier episodes, that people don't see how quickly things can evolve and change. We're not very good at that, as a species. So do you think in general the industry is moving fast enough, with enough urgency, to structure itself in a way that allows us to fully take advantage of and also manage this evolution?
Yasu Sasaki (14:24):
Yeah, thinking as a whole industry, we are still, the industry is just following the latest technology. generative AI or conversational AI. But we are still not sure how to utilize that AI in a proper way. Think about Dentsu. We have been working on AI since early on. For example, in 2011, we collaborated with the MIT Media Lab about AI. And 2017, we launched AI copyright, the first AI copyright at the time, before ChatGPT. And we also started working on virtual humans in 2019. So like that, we started earlier to think about AI. So that's why some people in our creative division are okay to use AI as a partner, not just following the technology. But still, yeah, changing is a bit so quickly, so some of the people in the creative industry might not be able to use in a good way.
Charles (15:35):
So, it sounds like you're approaching AI from a strategic standpoint as opposed to a reactive standpoint, and it feels to me that there are still quite a lot of people in the industry, maybe even the majority, who are reacting to AI, whereas you are not only using it deliberately and with intention, but you are also looking into the next stage of evolution and planning for that.
Is that a fair assessment?
Yasu Sasaki (16:02):
Yeah, I think so. Of course, we first just reacting with AI, but now together is a lot of diverse and creative people like technologists or scientists in our agency. Now, we can use AI in a strategical way. And then it is easy for creative people to use other AI as a partner and come up with a great solution using the AI, not just for making visuals, or not just for a chatbot or something like that.
Charles (16:36):
So do you have a vision of any kind for, say, I don't know, two years from now, about what you think the industry will look like and how you need to be using AI proactively? Are you that evolved in the way you think about it, that you're seeing a different future and working back from that?
Yasu Sasaki (16:52):
Yes. Our vision is that people and AI enhance each other, and in that way, for example, AI can focus on more efficient tasks, make a more efficient advertising, or efficient way to engage with people. Then, we humans can more focusing on non efficiency, means more emotional or sometimes it looks like useless, but something that people cannot forget.
So yeah, AI and people can go off together, but in a different direction. And then we people can more focusing on the more exciting and engaging and emotional direction to create a bigger idea than before.
Charles (17:42):
And you've invested significantly in AI capabilities and you've got partnerships with a number of different tech companies.
How are those helping you to support clients in a different way?
Yasu Sasaki (17:55):
Yes. For example, to enhance the efficiency of the marketing activity, we have Mugen AI. Mugen is “infinity” in Japanese. So Mugen AI's ecosystem, and it can find the insight, and also it can find the direction which message is better for the client.
And it can generate a copy, or visual, or movie now. And of course it can predict the result, and also it will introduce us how to improve the advertising. So we have that system. And we develop that in house, actually, and we're providing to more than 100 clients.
And the other way, to create the more groundbreaking solutions, we co-work with different AI provider like OpenAI or Google or other platform, to think about the innovative AI solution together.
Charles (18:53):
So with all of the experience you've got both individually and as a company, of using AI to expand the possibilities for how you can support the growth of your clients, do you have a sense yet of whether all of this change will result in the industry needing fewer people, the same number of people but doing different things, or more people doing different things?
Yasu Sasaki (19:20):
I believe that AI can give creative more time, so especially young people, more time, and they will be able to leave the unimportant, boring work to AI and spend their time thinking about more human things that only human can do. And also we found that using AI makes defined creative job. For example, prototype engineering, or the planner who can connect AI with solutions.
So I think from now on, we can make defined creative jobs using AI.
Charles (19:58):
So do you think that the economic model of the industry will need to change to reflect the fundamental change in the way that AI allows us to do work and the kinds of things that allows us to do? Or do you think, to your earlier point, that there'll simply be different kinds of job titles and job descriptions that can be still charged using the same economic model?
Yasu Sasaki (20:19):
Yeah, it's a very difficult question. But yeah, definitely, if we can use AI in a proper way, our time spent will be decreasing to create a greater idea. So it means that the charging fee in the time base, I think, it's changing.
So maybe we can more sell our idea in the quality based, not quantity or time based. So maybe we need a different economic system to charge that to the clients.
Charles (20:51):
Have you found clients who are receptive to that evolution?
Yasu Sasaki (20:56):
Still they are reactive, I think. Because some clients ask us to just provide a storyboard and they can customize using AI for their product. So they think it's efficient and cheaper to do that. But as you can imagine, the solution will, I think, become weak using that way. We need to have more proper understanding about our creativity and the effect of using AI.
Then I think we can work together with client and AI to come up with a greater idea.
Charles (21:35):
Do you think this is an opportunity for the value of creativity to be reimagined, to be re-understood? Is this a moment whereby the impact of genuinely original thinking can suddenly be put out in the world in a completely different way than this industry has been able to for the last several years?
Yasu Sasaki (21:56):
Yes. Of course, if the creative people are just average, AI can give us an average idea. Then I think if people can come up with a more outstanding, groundbreaking idea, then that idea will stand out more in the AI age. So it's been tough age for creative people, but at the same time, it's very exciting for talented creative people. Yes.
Charles (22:28):
And have you found that you're looking for people with different kinds of talents, different kinds of aptitudes now than you were, say, five years ago? Has it changed the talent supply chain from your perspective?
Yasu Sasaki (22:42):
Yeah, I think so. So, actually, Dentsu have a lot of creative technologists in the office. So we started hiring those kind of people more than 10 years ago, thinking about the digital advertising or digital experiential campaign. But it's very successful because they’re good at utilizing AI. So I think in this industry, creative industry, if we can have more those kinds of people, tech people or scientists people or data people, then I think it will be more exciting to come up with the idea together with AI.
Charles (23:25):
Back when I first got into the advertising industry, one of the things that stood out was the amount of training and development.
I mean, I remember going through training programs—I worked at Ogilvy, and they sent me through two or three different training programs as an assistant account executive, and as an assistant media planner. It was pretty remarkable. And then the industry really lost that component. Is it hard to find people who are fully trained or have developed the skill sets that you need? Or are you finding that you are now back investing in developing talent into different kinds of skillsets?
Yasu Sasaki (24:02):
Yeah, it's a little bit difficult to, yeah, train the talent, because the capability and talent and skills are really diverse now.
So, for example, the copywriter is not just a copywriter. So digital copywriter, or AI copywriter, or the content copywriter is different. And also, the art director is also different because to make a bigger key visual or prompting to the AI director or… So I think the skills will be really diverse.
So we need a defined educational system, training system, to have those kind of diverse talent in the network. Of course, AI can help those kinds of training program, but yeah, we need to have the good training system for the coming age.
Charles (24:59):
And one of the issues the industry struggles with is talent retention.
Do you think that the expansion, the evolution of AI and the development of different skills, as you've just described, do you think, is that going to improve talent retention? Are you already seeing any change in talent retention, that people decide we want to stay longer at this particular company because we've grown up here, or they've helped us develop skills that we wouldn't have developed elsewhere?
Yasu Sasaki (25:25):
So, retention might be also difficult, because if there is a similar AI system in a different company, I think people can easily move around, and the system is almost the same. So agency will be commoditized, so any place, any agency is the same now. So it might be more difficult to keep talent in the company for a long time, longer time, unless we have the very different AI system or education system, or different network of people that is attractive to the talent.
Charles (26:04):
I read somewhere that Dentsu, that had done a study with CMOs who talked about the fact that they were very conscious of the value of human creativity, that they believed there was always going to be a distinction between human creativity and non-human, AI generated, and that customers are prepared to pay a premium for human created content.
Do you feel the same way? Have you seen evidence that's true?
Yasu Sasaki (26:36):
Yes, so currently, yes, as you said, thanks to Creative 2023 CMO Survey told us that marketers value human creativity. And 75% believe that it will never fully replace the essence of human creativity. And 81% of CMOs agree that customers will be prepared to pay a premium for human created content.
And I agree that in this situation, yes, that if we can make a more attractive AI solution with human creativity, maybe CMO will change their thinking. But currently, yes, clients and we believe that human creativity can have more impact.
Charles (27:24):
So with that as a building block, that's a really powerful reference point for the entire industry, if that remains true. And to your point, we have to make sure that the value of human creativity is perpetually enhanced, that it's applied in the most meaningful ways possible. I know that you have built what you describe as a Tech Center of Excellence.
Can you talk a little bit about that and why is it important to build a Tech Center of Excellence and how does that relate to the value proposition that human creativity offers?
Yasu Sasaki (27:54):
Yes, so it's important to have a tech team, dedicated tech team for AI, because AI technology is so different and very quickly changing. So just creative people is not enough. So we need those Center of Excellence power to go for the AI creativity so that they are very helpful for us to create a solution for the clients.
Charles (28:23):
How has the advent and the expansion of AI changed your leadership of creative people? What's different today than, say, even two or three years ago for you in terms of how you lead?
Yasu Sasaki (28:35):
Yes. Mainly, there's not so much difference, I think, but our possibilities are getting bigger because of the AI age, so now together with the AI, we can create more than advertising, but we can create a more human or societal solution using the AI.
Now I'm encouraged the older creative to use AI more, and with a higher perspective to think about a bigger solution, not just an advertising campaign, but we can make an impact for people, impact to business, impact to society, using the power of idea and power of AI together. So, our field is expanding, and maybe our leadership and our creative direction is getting bigger and wider.
Charles (29:29):
Are you finding that is more inspirational to talent? That the fact that you can, to your point, you can have a bigger impact on society, you can have a different impact on the human race in certain situations? Are you finding that's more inspirational to the people that work for you?
Yasu Sasaki (29:44):
Yes, definitely, yes. So now we have a very different types of creative people in the office, and it's really inspiring, and also we can easily make an innovative solution together with those people. So I think it's very exciting and inspiring.
Charles (30:03):
Obviously there are some clients who are even more nervous than others about the risks attached to AI, and things like trademark infringement is probably the most visible one.
I've seen stories, I think I spoke about this on an earlier podcast about one CEO of an independent agency telling AdAge that they had won new pieces of business and were required in the contract to confirm that they would not use AI at all. Could you imagine today servicing a client without any use of AI whatsoever?
Yasu Sasaki (30:37):
If we don't have proper rules or guidelines or tools, then perhaps some clients don't want to use AI, and thinking about the risk of that. But if we create proper rules, guidelines, and safer tools for clients, then I think client will be happy to utilize AI. It depends on how we are ready to use AI in a safe way.
Charles (31:08):
So it's an educational process.
Yasu Sasaki (31:11):
Yes, I think so educational, and also ethical, legal.
Charles (31:15):
You touched earlier on how you are set up as an agency, in terms of integrating AI talent, capability, expertise into the organization, but also developing it and having it as a standalone area of expertise, as well. What's the division between those two? How do you determine who you need involved within the day-to-day functioning of the business, and which skillsets sit separately.
How do you divide those two?
Yasu Sasaki (31:43):
Yes, of course, we have a special tech team or AI team, but the important is have AI specialist in each division, like in media, or creative, or strategy, or account. Just having an AI special team is not enough. So important thing is to make each defined division, people, think about AI or use AI in a proper way.
So actually Dentsu has more than 5,000 people in AI connective community. So it means people in the division can think about AI or how to engage AI or, we share the idea together to make the whole division see AI. It’s a kind of collaboration, collaborative teams for AI.
Charles (32:34):
And as you are looking to hire talent for the organization, across the board, what are the attributes that you're looking for? Obviously the skillsets are changing all the time, but what are the attributes?
Yasu Sasaki (32:45):
Yes. Of course we need a tech AI specialist. But at the same time, we need a totally different talent, a copywriter or art director who totally don't understand about AI, but if they are curious enough to think about AI or use AI, that is enough. So please imagine, if everyone is just a AI engineer, we cannot make a innovative solution using that people. Important things that have a very different people, especially sound and totally non-digital designer, then I think that can make, I think, more innovative idea together.
Charles (33:28):
As you look at the future for AI within the creative industries, what are you optimistic about? What are you excited by?
Yasu Sasaki (33:37):
Yes. So most exciting thing is that currently, AI is just for making, generating visual or generating text or conversational partner, so I want to make AI more common partner for everyone, not just a AI engineer or geek or special tech people, but for everyone. Then I think AI can be a much more powerful partner for all the people to create innovative solutions.
Charles (34:10):
And as you look at the future of AI, what are you afraid of?
Yasu Sasaki (34:15):
Yeah, if AI is good enough to just make content or generate an idea, maybe people might become lazy and don't think in their own way. And if People become lazy using AI, then AI will lose their learning source, and it means that AI will become weaker together. So it is the worst scenario to make AI and people getting weaker together. So needs more very human creative power to avoid that.
Charles (34:56):
Yasu, I really want to thank you for coming on the show. I think it's fascinating to talk to somebody who's clearly been immersed in AI and the potential of AI and the ability of AI for such a long time. And to see how organic it is to you, to see how instinctive and I think almost intuitive it is to you.
And it feels to me like you are building a company that is trying to figure out where the puck is going, and you want to meet it in the future. So I really want to thank you for coming on today and for sharing your deep expertise and your very empathetic approach, I think, to how you partner with AI. I think it's a powerful reference point for all of us. So thank you.
Yasu Sasaki (35:33):
Thank you very much. Thank you.
—————
If you’d like to know more about the podcast, or about our leadership practice, go to fearlesscreativeleadership.com, where you’ll find the audio and transcript of every episode, and more information about our work with some of the world’s most creative and innovative businesses.
Don’t forget to share Fearless with your friends and colleagues, and to leave a rating and review.
Fearless is produced by Podfly. Frances Harlow is the show’s Executive Producer. Josh Suhy is our Editor. Sarah Pardoe is the show’s Producer.
Thanks for listening.
