MOLLY WOOD: Today I’m talking to James Thomas, global head of technology at Dentsu Creative, about his organization’s early experience with generative AI at work. Dentsu is one of the world’s largest global creative agencies and was also one of the first to try out Copilot for Microsoft 365 as part of Microsoft’s Early Access Program. James has a report from the front lines. He shares how incorporating generative AI into everyday operations has transformed his team’s workflows. And he talks about what business leaders should keep in mind as they begin getting their hands on this technology too. James, thanks so much for joining us.
JAMES THOMAS: Thank you for having me. It’s great to be here.
MOLLY WOOD: Okay, so Dentsu was one of the first companies to try out Microsoft’s Copilot as part of its Early Access Program. Tell us about it. What has the experience been like for the teams that took it out for a test drive?
JAMES THOMAS: Yeah, so, a huge amount of people wanted to get their hands on it. Unfortunately, we didn’t have that many licenses. But the lucky 300 or so that got access is really a real game changer. A lot of people can’t see themselves going back to how they used to work before. People see the efficiencies. They see the quality of their output being improved and the quality of their collaboration, particularly in creative roles. You know, a lot of time is spent planning—complex production meetings and creative sessions where a lot of notes and a lot of action and next steps are needed to be created, a lot of complex budgetary Excel documents to manage, even just summarizing a long email chain. You know, those savings and efficiencies are huge. It means people can get straight onto the tasks that they’re supposed to do.
MOLLY WOOD: How has AI impacted the creative process at Dentsu?
JAMES THOMAS: So in the actual creative process itself, you know, ideation, we would typically work with a client in what we call a chemistry session, which is, you know, we get everyone in a room, we’ll really just kind of collaborate, come up with some great ideas over a few hours. Typically then what happens is we would go away as an agency and we would come back, you know, days, weeks, sometimes even months later with some creative treatment, visualizations of these ideas, right? And, uh, it’s taken months to even get to that stage. By using Copilot in Teams and DALL-E, we can ideate while we’re in the room with a client. We can sketch down ideas into Word and have ChatGPT expand that out into thought out ideas. We can create visualizations in DALL-E. The hard bit with creative is trying to bring other people on board with the idea you have in your head. And, uh, the more creative the idea, the harder it is to try and get it to people. So the idea of being able to really visualize things very quickly or just take a few lines of copy and suddenly turn that into some kind of proposal is huge. And again, it just means we can get going fast with the client. We know where the dead ends are and we know, right, this is the right idea. And it just helped us to iterate a lot faster. I think a lot of these use cases are ones we hadn’t thought of before. You know, creatives, I think it’s fair to say, were quite apprehensive about this technology when it was becoming more widespread, you know, like, like a lot of us—what does it mean for the future, their work, their jobs, humanity—but, you know, now I think creative people have seen that it’s actually just a tool in their toolkit and it means they can get to their ideas a lot faster. In fact, as the Copilot deployment at Dentsu has shown, yeah, we’ve got a waiting list of thousands. We can’t get these tools in people’s hands fast enough.
MOLLY WOOD: What was that preparation like? Because I do think the creative realm is sort of a space where the easy answer is, we will get all this efficiency and that will allow us more time to be creative. But talk a little bit more about the evolution, like as you started to discover, oh, this works in the process of being creative as well.
JAMES THOMAS: I think everyone is using it in different ways. What we have done really well, thanks to Dominic Shine, our CIO, and Brian Klochkoff, who is enabling all these services at Dentsu in a scaled way, we have really great office hour meetings, town halls, where all of the users are coming together, sharing their own tips and tricks they’ve found. Um, so we’ve really put a lot of work into helping people know how to use it.
MOLLY WOOD: What do you imagine as you go forward and eventually everybody has this, and these efficiencies start to pile on top of each other? What do you imagine that will allow?
JAMES THOMAS: Well, that’s a great question. I think particularly, and I can only really speak for creative, but what we’ve been battling against for quite a few years is that need from CMOs for almost unlimited content—just non-stop content, new versions, new iterations for new channels and everything else. Because of that, and because a lot of the workflows were manual, you could automate some things, but what it meant was creatives were really up against it, overloaded, right? And they were just churning out lots of different iterations and not having the time to spend on the upfront creative idea. The idea of being able to use AI and technology to generate content, generate copy, generate images, it frees up the time that creatives need to be creative. Creativity is not a fast process, usually. You need time, you need space to think through these ideas. But if you know, oh, I’ve got to create a thousand versions of this idea, I don’t have too long to think about the idea in itself. I have to get on and make all these versions. But if I know I can come up with a really great idea, campaign, set of assets that I can take a lot longer to do, and then use generative AI and other technology to help scale that, infinitely almost, then that’s a great place to be. And that’s what everyone wants. Everyone wants the best possible idea.
MOLLY WOOD: Talk a little bit more about the assistance in the creative process. You know, I’ve spoken to people who are using ChatGPT, for example, to brainstorm ideas or, you know, please challenge me on this, or to kind of bat around concepts. I wonder how much of that is happening with you, that actual interaction in the creative process and iterating and maybe using these tools to continue to generate or improve on your ideas.
JAMES THOMAS: No one likes to start with a blank page. You know, it becomes very quick and easy to just throw down a few ideas, get them scaled out into, you know, from one line into a paragraph, from a paragraph into a proposal. But then it also allows you to—because you can get more ideas out, it means you can hopefully get to a better idea quicker. Again, all these things allow for more creativity, and it helps when you’re selling the idea to people because you can be like, look, this is the message, but look, here’s how we tell it to your 10 different audiences.
MOLLY WOOD: You’ve been using this for a while and it, you know, sort of sounds like it’s sunk in for you, but was there a moment when it blew your mind?
JAMES THOMAS: Yeah, I mean, a couple of things that keep wowing me every time is, you know, I spend my life in meetings, on Teams mostly, and the note taking, the transcribing, and the action steps—and also, if you join a meeting late then just kind of recapping what’s happened so far. The accuracy is incredible and I’m back to back in meetings, so unfortunately, quite often I am a little bit late for the next one. I can hit recap meeting and I’ll be up to speed straight away. It’s way more accurate than the meeting notes I take or anyone else I know takes. So that is always a huge one for me. And also, I travel a lot and I get a lot of emails. So catching up on emails is always a pretty daunting task while on the move or coming back from holiday. And now just being able to summarize long email chains in Outlook with Copilot is great. I don’t need to sit through these 30 emails on this one subject. I can get a two-paragraph summary on what’s happened and who’s doing what.
MOLLY WOOD: So let’s talk about advice to business leaders in terms of adopting this and integrating it into an organization, especially in the creative industry, because like you said, there is— [interruption] What happened?
JAMES THOMAS: So that’s my automatic cat timer. And it’s my voice telling my cat it’s dinner time. Quite embarrassing.
MOLLY WOOD: That’s incredible. You really have it all. You have all the toys.
JAMES THOMAS: Yeah, I didn’t know that would get picked up.
MOLLY WOOD: There’s a timer that tells your cat that it’s dinner time in your own voice? Oh my god.
JAMES THOMAS: Yeah.
MOLLY WOOD: Okay, so you’ve given us some really tactical examples so far on how Dentsu leaders are incorporating Copilot into their workflows. What would be your advice for other business leaders who haven’t gotten their hands on it yet? What should they be keeping in mind as they get started?
JAMES THOMAS: We’re a huge holding group. So, you know, there’s equal value in creative as there is in the finance team, as there is in the HR team. Everyone’s using it in a different way, and I think helping to bring people together. Help them to share their findings and their insight on how they’ve been using it, and also sell it to them in the use cases they understand. If you’re talking to a creative part of your business, talk to them about the creative use cases. You know, they might not be so interested in what you can do in an Excel document, but they’ll be very interested in some of the other use cases that are much more specific to their world. So I think maybe tailoring it for your audience will help drive the adoption.
MOLLY WOOD: As you have described it so far, it sort of sounds like these are tools that will allow you to keep up with and succeed in the media world that we live in now, right? Where there are, like you said, different models and different distribution outlets and different ways to get attention. Once you feel caught up, can you imagine how this would also help you create new business models and services?
JAMES THOMAS: Yeah, absolutely. And I think the technology with Microsoft and OpenAI, and what we can do with ChatGPT and DALL-E and how via APIs we can start to plug these things together and build out tools. One thing we launched this year, really to kind of meet this challenge, is what we’re calling the Technical Delivery Center of Excellence. It’s a combination of talent from our India, Brazil, and Poland teams, innovation and generative AI experts and specialists. And we’re really scaling these teams in order to meet the demand of what’s coming. And what’s coming is a lot of custom experiences, a lot of custom platforms and tools. You know, we’re already doing award-winning work. We had two awards at Cannes this year, the top two awards, the Grand Prix awards—one for Scrolling Therapy and one for The Inflation Cookbook, both AI-powered tools. There are so many different products. Also, building custom models and custom algorithms, you know, that’s where a lot of the IP and a lot of the differentiators will come. This technology is going to be available to everyone—brands, agencies. Everyone’s going to be able to get a hold of it, but it’s how you piece it together, how good you are at building a custom model. They say you can give anyone a guitar, but not everyone can play it the same, or play it well, right? Just because you have tools or have access to this stuff doesn’t mean you’re going to be effective with it. So yeah, I think there are so many opportunities if these technologies and experiences can enable more attention and more engagement and ultimately more sales, most of the time, then that’s a good thing for everyone.
MOLLY WOOD: You’re so far ahead of everybody else, right, in adopting these tools and having this early access. And so, I wonder, what do you say to AI skeptics, or people who are reluctant to integrate generative AI into their lives?
JAMES THOMAS: The biggest misconception is that it’s just going to take everybody’s jobs. I mean, obviously AI has been around for a long time, but generative AI in particular, obviously hit the big time because I guess it’s a bit more sexy. It’s a bit more fun. You can go online and turn yourself into Drake and make a song. You know, there are so many fun things that anyone can do. For people to be able to go on and just create content, it’s great. And as a creative, I’m always in this kind of slight battle with myself: I’m a creative first, then a technologist. Anything that can make people creative is a good thing, right? You know, if I can go online, and I might not be the best video editor or photo retoucher ever, but if I can create something and go online and AI can assist me in bringing it to a much higher level, then that’s great. I think everyone should be able to be creative and create the things they have in their mind without technology or skill, to a degree, becoming a blocker. I take advantage of that all the time. As I talked about before, it’s just finding ways for people to be more creative and augment their workflows and make better content, or faster content. But you’re always going to need humans. You’re always going to need the human feel and human touch over this work. And it opens up new opportunities for careers and skill sets and talents as we see people getting into prompt engineering.
MOLLY WOOD: What is, other than feeding your cat, the latest thing that AI has helped you with at work and not at work?
JAMES THOMAS: At work, it’s helped me to deliver incredible work to clients, work that we’d never been able to do without this technology. Personally, it’s enabled me, as I was talking earlier, to create my own content better. I also have a music career I’ve had for a long time, and that makes me have to release other content on a personal note. And yeah, the tools to get that content better, edit it quicker, faster, improve it is great.
MOLLY WOOD: What kind of music do you make?
JAMES THOMAS: All sorts, I work with a lot of big hip-hop artists in the US to film scores to brands. I’ve done a lot. We used to do it full time for a long time as well.
MOLLY WOOD: What do you do with any extra time that AI creates for you?
JAMES THOMAS: Learn about the next thing that’s coming with AI, usually. [Laughs] You know, everyone’s excited about generative AI and where we are right now. My job, and my team’s job, is always to be one step ahead.
MOLLY WOOD: Fast forwarding three to five years, what do you think will be the most profound change in the way we work?
JAMES THOMAS: Uh, wow, that’s a big question. It’s actually really hard to think about where things are going to go. You know, a year ago, we weren’t even thinking about generative AI, really, you know, on the most part, and look where we are now. So, it is hard to say, but I think almost everybody in almost anything they do career-wise, you know, professionally, and as a consumer, will be leveraging AI in their day-to-day work, whether it’s Copilot, whether it’s specialist tools. I think everyone is about to get a lot more efficient, hopefully a lot happier, taking some of the legwork out of some of the mundane tasks, but I can’t see a world now where people aren’t going to be leveraging AI to enhance the way they work and the way they live.
MOLLY WOOD: James Thomas is global head of technology at Dentsu Creative. Thanks so much for the time.
JAMES THOMAS: Thank you so much for the opportunity.
[Music]
MOLLY WOOD: Thank you again to James Thomas. And that’s it for this episode of WorkLab. Please subscribe and check back for the next episode, where I’ll be talking to Microsoft Chief Scientific Officer Eric Horvitz about how leaders can leverage the tremendous potential of AI to best serve humanity. If you’ve got a question or a comment, drop us an email at worklab@microsoft.com, and check out Microsoft’s Work Trend Indexes and the WorkLab digital publication. There you’ll find all of our episodes along with thoughtful stories that explore how business leaders are thriving in today’s new world of work. You can find all of that at microsoft.com/worklab. As for this podcast, please rate us, review, and follow us wherever you listen. It helps us out a ton. The WorkLab podcast is a place for experts to share their insights and opinions. As students of the future of work, Microsoft values inputs from a diverse set of voices. That said, the opinions and findings of our guests are their own and they may not necessarily reflect Microsoft’s own research or positions. WorkLab is produced by Microsoft with Godfrey Dadich Partners and Reasonable Volume. I’m your host, Molly Wood. Sharon Kallander and Matthew Duncan produced this podcast. Jessica Voelker is the WorkLab editor.
