PODCAST: Generative AI in Comms - Grayling

Views (blogs)

PODCAST: Generative AI in Comms

November 28th, 2024

/ Tags: AI

The rise of generative AI is transforming how we create, communicate, and consume content, but it’s not without challenges. In this episode of Vantage Point, Arne Mosselman (Ainigma & Grayling), Nathan Kemp (Grayling), and Scott Langham dive into AI’s potential to revolutionise communications while tackling big questions about authenticity, responsible use, and strategy.

From practical frameworks for implementation to myth-busting and bold predictions for the future, this conversation offers valuable insights for anyone navigating the intersection of technology, creativity, and trust.

Listen to the episode to discover how to harness AI’s power without losing the human touch.

Transcript

Hello and welcome to the Vantage Point podcast. I’m Scott Langham, senior counsel at Grayling. And today we’re going to delve into the world of artificial intelligence. Barely a day goes by where there’s not a new development in AI and like it or loathe it, there’s no question that it’s here to stay.

In 2024 over 400,000 articles mentioned articles mentioned the word AI and according to government research, around one in six UK organisations have embraced at least one AI technology this year.

That figure rises to 68% in large companies, and it has long been trailed in popular culture. You only need to think of 1980s movies like war games or the Terminator, that unchecked reliance on machine learning and AI could lead to disastrous circumstances. But away from Hollywood, it could be argued that advancements in AI, do represent the greatest risk reward ratio in the corporate world going into next year.

And that year will continue to be dominated by questions around responsible use of AI and in particular how it’s deployment will change the way we work today.

So where are we on the AI journey and what promise and pitfalls can we see as we go into 2025? I’m joined today by two very lovely humans we have Arne Mosselman, founder and CEO at Enigma, which is a people first generative AI consultancy, which creates AI strategies and delivers change programmes to upskill every employee.

And our very own Nathan Kemp, Chief Innovation Officer at Grayling who, when we attended an AI conference together just a few weeks ago, admitted that he simply loves to play with shiny new things.

So let’s get into it, Arne, I’m gonna come to you first. Talk to us about what inspired you to start a generative AI consultancy.

I think from the very first moment I used Dalle or ChatGPT. I just realised this is going to transform at least our industry. I’ve been working in the world of comms for 15 years and the things that I always thought only humans could do. Suddenly a machine could do too and that means it’s going to change business models. It’s going to change roles. It’s going to change jobs. It’s going to change the things we read.

And I realised organisations and businesses will need to transform too, and that’s where we wanna help with.

But it’s gonna change businesses, Nathan. I mean, how do you see it transforming the business and creative processes?

00:02:30 Nathan

I think hugely and I think we’re at a really early stage of that. But I think to break it down perhaps in really simple terms and I speak for agencies here, all I know, as opposed to, much broader businesses.

00:02:43 Nathan

I think you can break what agencies do into 4 buckets and this is super simplified, but it’s thinking creating, delivering and measuring and obviously, every agency has loads of different products and services that sit within those buckets, but broadly those four areas and if you look at the potential of AI, I think it has huge potential across all of those different areas.

00:03:03 Nathan

So if you look at, your ability to mine insight and data to inform that early strategy phase AI’s already in a lot of the tools that we use. Creatively, I think that’s an interesting debate we had.

00:03:20 Nathan

Will you get to Cannes off the back of a brainstorm using AI – to be confirmed. But I think you have to accept that there’s no such thing as a like a totally original idea. I think every idea is a build on an idea that’s come before. And in that respect, platforms like Chat GPT are the world’s biggest ideas library.

00:03:39 Nathan

So I think it’s how you use that tool in the ideation creative process where it can start to become really interesting. I think it has lots of uses across helping account teams to do what they do.

00:03:53 Nathan

And of course, in terms of measurement, you’re back into the start point in terms of our ability to compute and track and monitor data, and again, AI as a role there and it’s already in a lot of the measurement tools that we use. So that’s perhaps a rather long winded way of saying it has, I think, huge potential, will be really interesting to see how it starts to play out across those different areas.

So you would say that there’s this, untapped opportunities within the space, maybe Arne you could talk about how generative AI is currently being used in comms and PR and then what those untapped opportunities might be?

So I think we’ll we’ll see two things.

So if we talk about generative AI, we all think of ChatGPT, but now also you can create images.

You can create videos, you can create podcasts. So what we’ll see is that we will all just have an endless ability to create content. Our LinkedIn feed. I’m sure yours too, the bakery around the corner can craft a beautiful LinkedIn post about the Christmas party last week.

The bakery around the corner can craft a beautiful LinkedIn post about the Christmas party last week, so we’ll see.

So we’ll see a world where there’s so much content out there and people are gonna use AI to find the things that they want. They might not go to Google, they might use another AI tool to find the things that they want. So the world will change in terms of what is being produced and how people people consume it.

And the real untapped opportunities is about agencies or brands or clients thinking, how do I stand out? How do I use these tools to actually get the attention of the public, to do something different, to do something interesting? So not looking at just AI as replacing the content you produce yesterday but actually thinking, how do I use these tools to craft and engage customers in a different way?

And I think authenticity is very key in all of this and ethical use of AI generated content. I think it’s a bit of a legal minefield, its certainly something that people are gonna have to grapple with next year in particular. How do you address those concerns around authenticity? You said that an idea is never entirely new and is always building on something else.

But how do you address that and use AI in a responsible and ethical way?

00:06:04 Nathan

I think you have to be really open and honest and transparent about how you’re using it across the business and I think that’s being transparent at all levels, internally and externally.

00:06:12 Nathan

So transparent with colleagues as well as with clients, having frank conversations about the role for AI in that process, I think it’s important you understand the legal framework and ramifications around that and to quote the event we’re at last week, he talked about lawyering up.

00:06:32 Nathan

I think that’s maybe a little bit extreme, but I think, understanding the legal parameters within which we operate. I think not relying on the platforms is another really important one. I think it’s generally accepted that they will probably do the least they need to do in terms of copyright protection and that thing, so not necessarily just relying on them.

00:06:47 Nathan

And I think, the ethics point is, super important. I think thinking about developing a really strong code of ethics as a framework right in the early start of your AI journey. And I think looking externally.

00:07:07 Nathan

You get the point of view and perspective of individuals outside the organisation to give you that external kind of benchmark and check and balance is important as well.

Thank you. I mean you guys have created a report recently.

It’s got a really snappy title which I’ll have to read cause it’s so long, ‘how generative AI is revolutionising the communication industry by empowering everyone from the ground up.’

I mean this outlines lots of different organisational responses to AI.

Can you just take us back into why you wanted to create this report in the 1st place and any key insights that you’ve taken from the process of of creating a draft?

What I really like about it is we were at the Dutch Embassy as part of the AI – the trade mission of the Netherlands into the AI Summit in London. And they said, well, actually, if you’re interested in in engaging in with one of your clients, do you wanna write a white paper? So we got the opportunity to do that. And I thought actually one of the things we discovered together in our conversation, but I’ve seen it in lots of organisations, is that the place where innovation happens, where people discover new things, are from unexpected places.

So it’s not per say the innovation departments or the places where the biggest tech specialists are, but it’s just people who happen to quite often open ChatGPT, open Dalle, open mid journey and said ‘hey, look about this. I tinkered over the weekend and I found a new way to do public affairs.’

Or ‘I just discovered a different way of doing a podcast.’

So I think the core insight to me is that this technology is available to everyone. You don’t need to be an expert.

You don’t need to know how technology works. You don’t need to have affinity with it. All you need is an idea and a customer need that you want to fulfil that you think, ‘hey, this is an interesting shortcut’ in a way and that shortcuts are in the organisation.

So the core message, I think of the of the White Paper is to say you need to give everybody in your organisation access to the tools and let them play. Let them reinvent ways of doing, because then you will see innovation. Don’t start at the typical place where you would do it.

00:09:13 Nathan

I think the other thing perhaps to build on that was at the time we wrote that report there were all these headlines being banded around about how it was promising 30% saving on time efficiencies.

00:09:16 Nathan

And it was just really interesting from a practical point of view in Grayling where we started to trial it and we ran the AI training at the start of the year that you don’t necessarily see that like immediate transformation, right?

00:09:33 Nathan

So I think it felt like you know, hang on a minute.

00:09:38 Nathan

There’s this slight difference between what we’re seeing, the reality of that and all of the promised delivery that it can do. And it felt like it made sense to just take a step back and look at actually, what’s happening? Why is that happening and what is the right approach?

And you talked in the report about the different and in brackets, ‘wrong’ organisational responses to generative AI. Can you talk us through some common missteps you can see companies making in terms of their use?

I think one is the wish to do it right and to pause until all questions are answered and the reality is we are in a very much unknown territory, unknown from a legal perspective, ethical perspective, value, perspective, technology perspective. We don’t know what the world will look like, but you can’t wait for the answers to be there at the start.

The second one people is people say, ‘oh, I’m going to rely on the experts’ so we’re going to start in this corner with our tech people, our head of AI, head of data and treat it like you would any other tech revolution before.

Of course those people are core to actually making it successful, but don’t put it in the corner. So those are the two biggest mistakes that we see.

And from your perspective, Nathan?

00:11:00 Nathan

I think it’s focusing on purely the tech side of it and we’ve observed through our own experience at Grayling and it’s something that we’ve talked about a lot and it was something that one of the external consultants that we were talking to at the beginning of our process, at Grayling, he had bought and sold a whole number of huge tech businesses and his immediate perspective on it is it’s about 30% tech and 70% change management.

00:11:29 Nathan

And I remember coming away from that meeting and it completely reframed our perspective of that. I think it’s really easy to make all of the focus around the tech, for very obvious reasons and you get really, embroiled in conversations with IT teams. And you deliberate loads between what platforms to use and all of that.

00:11:50 Nathan

And actually you forget that you know the tech’s fine, but it’s only ever gonna be helpful, and you’re gonna be able to see the value in that, If you get a whole agency of people properly using and embracing it, and that bit is a much harder and bigger and longer job to do than making a relatively quick decision on whether you stick with your legacy IT provider or go with a new platform.

And are there any sectors or industries that you think are leading the way in fostering that bottom-up innovation, you talk about from the ground up with AI, is there in any particular industries or areas you think are really providing or leading the way in that space?

I don’t know of an industry or sector. I think one place where you see it is in SME startups, small companies, can be any type of company who just say, hey, I’m gonna give my people the capability to do something. Sometimes being able with 5 or 10 or 15 people to create something that was unimaginable before.

And then the second bit where we see it is being embraced is in the developing world, where for example, we see in the healthcare sector, doctors who weren’t able to say serve 50,000 people in in the area where they live, suddenly say, hey, wait a minute, all these people might have access to a smartphone.

Can I build a very simple tool to help with diagnosing certain diseases amongst the population who would never be able to see a doctor.

So in a lot of places where you just see that the legacy system isn’t good, where people aren’t being schooled, where people don’t have access to things that suddenly AI provides, that’s where you see people embracing it because they realise it provides something meaningful.

Let’s talk a little about the intersection of human creativity and AI. Obviously it’s that mix of human and machine. Are there any kind of myths or misconceptions you think have taken hold about generative AI that you think is worthwhile dispelling?

00:13:51 Nathan

I think to jump in here, and this again came from someone else outside the agency and he talks about, don’t think that AI can think.

00:14:04 Nathan

And I’ll probably do a very bad job of explaining this, but it’s massively complicated maths. Making a prediction on the probability of what word best comes after the word that’s just landed before. And that’s based on, everything it’s just pulled from the Internet or from Reddit as we learned in the early version of ChatGPT.

00:14:25 Nathan

And I think once you get that you aren’t working with this sentient thing that is thinking, there’s just a reality around that, I think if anything it makes you realise the importance of keeping a human in the loop and that you can’t expect too much, at least at the stage we’re currently at with these platforms when it is just kind of mass and probability.

The other bit of what I would add is this idea that AI is gonna help us become more efficient and more productive because it only does things that we don’t want to do, so treat it as an automation assistant. Oh, it’s great. There’s a lot of bunch of stuff in my work that I don’t like to do and AI can do that. I think that’s a really limiting view.

It’s a worldview where people predict we’ll have less jobs. A much more interesting and powerful aspect of AI is that it can you allow you to do things you thought were impossible before? So you can say I write press releases every day, AI can help me write that press release faster. Great. Gonna take Friday off. Or you can think, wait a minute.

With AI, I might be able to create videos. I might be able to create podcasts. I might be able to turn a story into a game.

And if you allow that aspirational thought into your brain, you’re like, wait a minute, where will I be in 10 years? Where’s my company gonna be in 10 years? What can we do? You suddenly plug the idea of creativity of, of innovation, of possibilities. And I think as a society, that’s much nicer thoughts to brainstorm together.

It’s way more interesting to sit, for example, together with Nathan and think, hey, if we use AI, how can we empower everyone to do more? So I think the efficiency routes might be helpful, might be good for business case, but it’s a misperception that can also leads to people just not looking at this aspirational aspect.

So using it functionally rather than using the possibilities for innovation. I mean I’ve talked to you before about this actually Arne, but I’m a little bit of an AI sceptic. You are winning me over every time we meet.

The thing for me is around the risk.

And I think there’s a lot of risk in terms of training the models.

Biases within the AI. You know it’s only as good as the information that’s been fed into it. So can you to talk us about how you would advise people to manage those risks and respond to the inherent bias in AI tools and outlets?

So one of the things we advise organisations on, is the first thing you should train your people on is do a responsible AI usage training. Should be the very first thing you do. Because in very simple terms, I fully agree with you, it is risky.

It’s Risky on on a personal level with your own data, with all the biases, with the quality that goes in there, how it can be used against people by criminals, by malicious actors.

And the trouble is, there is no solution. I think there should be better regulation to force companies to adhere to those things because I don’t believe you can leave the tech sector to its own in that way. But in the end of the day, the tools are out there and the only way to uncover how they actually work, where the limits are and the opportunities are, is to work with them.

But to treat them as a tool, as a very powerful tool. Sometimes I tell people almost envision, like you’re suddenly have Superman, Superwoman powers, and then I think it was Superman that correct me if I’m wrong.

With great power comes great responsibility, right.

Sorry, Arne it’s Spiderman.

I should have ChatGPT, AI’d my line.

Yeah, that’s the misinformation that we worry about.

So I think we should all realise that and to give a very practical example, if you’re a designer, you’ve been trained to always when you’re making images to think about, do these images reflect the view that I wanna portray. I wanna portray six families.

Do they match the inclusive approach that I have, that it needs to, so designers have trained to think about that.

If now suddenly you give somebody else the ability to create images in a presentation, or report that person who never created images before might just not think about that. Be aware of that. So if you give people these tools, or if you have those tools yourself you should really understand the risks and devices in the system in order to use them to right in the right way,

But it’s very challenging.

It is challenging. Nathan?

00:19:16 Nathan

I think just to add to that, I think it’s also acknowledging that there’s risk within your organisation and outside of it as well. So I think you know internally there’s risk as Arne’s just talked about in terms of how employees are using AI Platforms.

00:19:29 Nathan

You might have one department that’s using it in one way that is at odds with the external comms that that business is putting out and believes how they’re actually using it. So I think there’s risk within inside the organisation and then, there’s misinformation as we’ve talked about last week, the threat of that coming externally from from bad actors. We had examples of chat bots going wrong.

00:19:50 Nathan

So I think organisations need to have a risk strategy whether that may come from internal or indeed from external misinformation challenge.

And those chat rooms were quite funny as well.

00:20:05 Nathan

They were, and I think actually the interesting thing around them was how businesses responded to them. The ones that were very open around the failure of it, but also used it to talk about the fact that they were embracing AI as a new technology, I think from a from a corporate reputational brand point of view was actually quite interesting.

00:20:24 Nathan

I think being a business that looks like it’s on the front foot, it’s leaning into that technology. It’s experimenting, which we talked about in the White Paper, the importance and role of experimentation to unlock that innovation. I don’t think there’s anything wrong there with being transparent where that’s worked and where that hasn’t.

00:20:41 Nathan

And there’s any number of other innovations from loads of businesses, you only need to look at Amazon un-AI related, where you know a part of innovation is accepting that everything’s gonna work out. I think being transparent and open about that is an important part of that process and the comms around how we’re using as a business.

And just for those listening, we’re talking specifically about Car dealerships.

So if someone goes into a car dealership and says I’m looking for a four seater family saloon and they recommend them to go down to a competitor down the road because that’s a better choice or an airline customer service agent offering free flights to someone because they’re trained to be empathetic and something goes wrong or sweary delivery channels.

Just before we finish, I think we should look back to look forward.

You talked us through, Arne, why you decided to create a generative AI consultancy, but clearly the launch of ChatGPT was a pivotal moment in the development of it.

That was the moment where the world kind of stood up and took notice. It’s a turning point. So what was the seminal moment when you realised that, actually this could have a revolutionary impact on the PR and comms industry?

I think for me it was the moment, six months? I think five or six months before ChatGPT launched, they launched Dalle, the image generator and although the images were, especially if you look back, but also in that day they were horrible.

They didn’t look good. You were like, what am I looking at? But you just could see that you type in a couple of words and you press a button and suddenly an image is generated that at least tries to look somehow like the first input.

Suddenly gave me in my own world this idea like, wait a minute. There is now a machine, a tool an intelligence, that can or will be able to quite soon to do a apart of my job that I never would have envisioned, never thought about even.

Suddenly a technology would would come and enter.

How about for you, Nathan? What was the moment? The light bulb moment for you?

00:22:46 Nathan

Perhaps I was a bit late to the party. I think for me it was when ChatGPT launched. I am no tech fanatic or expert. You’ll never find me in the front of the queue for an iPhone launch. Ever. But it was something about just the accessibility and the ease of it.

00:23:05 Nathan

Just being able to go on this website and access this platform and ask it a question.

00:23:10 Nathan

And you know, we work in comms. We’re used to briefing people. So I think instinctively we’re probably quite good at knowing how to frame a good prompt and getting a relatively decent answer back. And I think that moment of looking at that and then just thinking about the application of that across our jobs and what that means for the industry was pretty profound.

It’s predictions time before we finish.

I would like you to predict one wow moment, the moment that, in the next 10 years, in the next decade, we might see, a moment where it explodes into something beyond what we thought about from those early images into the early form of ChatGPT. Arne, please.

Can I choose two? I think as a society. It will probably be when we just realise. Wait a minute. This new technology can help solve parts of cancer, for example, or develop new proteins that really make us more healthier.

So I think as a society that will be the moment that we realise this turned from a toy into something magical to actually something radical. On an individual level, I think most of us will probably, every other month we’ll have a wild moment when you realise it can now do something that I never thought a computer was able to do.

And I think at least if I look back at the last year, every other month, I see suddenly wow, it is now able to do this and I just couldn’t envision that.

So It’s not one prediction, but there will be tonnes of wow moments in the coming year.

Ok Nathan now you’re only allowed one.

00:24:48 Nathan

Thanks. I’ve also got 2, but Arne stole both of mine, so maybe I’ll have none.

00:24:55 Nathan

I’m no expert at all, but I think when we start to see those real leaps forward in like health where it starts to really solve a problem that is critical and personal to a huge number of people I think will be really, really interesting.

00:25:13 Nathan

I think there has been some advancements already, talking about how the NHS are using it and others. But I think to see a really big scale would be fascinating. And I think on a personal level, I once heard someone talking about AI as being an imagination maker and I think that’s still true.

00:25:32 Nathan

If you go in and ask imagine ‘x’, then I always find the response you get back is super interesting and a bit thrilling and I think just continuing to view it through that lens and use it to unlock our imagination and see where that takes us is really interesting insight.

Final question. Oh sorry Arne were you gonna say something?

I was gonna say could I try in another wow moment that you can cut out if you want to.

We wouldn’t cut anything from you Arne, it’s gold.

On behalf of the non-native English speakers. I think what you’re now seeing in the ability of live translation, so I heard a couple of people speaking last week where they said that more and more on online forums they see that people are commenting on eachother all in their own language.

So there’s one conversation around the topic, but because the AI is already, translation is integrated into the platform, everybody can speak their minds and their thoughts, whether speak Czech or Mandarin or Swahili or Dutch or French. It’s one conversation where everybody speaks from the tongue that feels most natural to them and to be able to have one flow.

And I think we live in a world where there’s two or three languages, really dominant, and people who don’t speak those languages as natural as possible.

But there’s a billions of people around the world who cannot really take part in the same way in business conversations, organisational conversations, cultural conversations and that moment last week YouTube launched auto dubbing.

So you will be able to upload this video and suddenly have somebody watch this in French or in Greek and it sounds like our voices.

So who knows? In quite a short time we could do this interview again and I will be able to answer in Dutch. And you asked the questions to me in English and it would feel like the most natural conversation to all of us. And I think empowering, getting all the voices, all the cool creatives on the same level in the world, that’s the moment I look forward to.

And maybe your avatar will be sitting here instead of you.

Just a couple of final thoughts Nathan, just coming to you, you talk about things being thrilling.

What’s the most exciting thing for you about the future of innovation in comms in this AI space?

00:27:53 Nathan

I think it’s just the extent to which as an industry, we always innovate and I think if you think back to like 20 years ago what our day jobs looked like then, we were stuffing, printed press releases into envelopes and taking them to a franking room, and spending 20 minutes stamping them, Like look at our jobs now just in that time frame.

00:28:14 Nathan

Now introduce AI into that and think what that might look like in 20 years time. And I think as an industry we do a pretty good job of evolving and innovating, and just generally, progressing and I think AI will just be a huge catalyst to that and that’ll be super interesting to see where it takes us.

Great. And finally to Arne, as we wrap up, if there’s one piece of advice you’d like to give to businesses looking to integrate artificial intelligence.

What will it be?

Empower all your employees. Give them all access to an AI assistant. Make sure they all get trained on the risks, but also the opportunities.

And allow change to happen, this will change your organisation.

So allow people to experiment with the tools, to use them, but also maybe change your workflows, change your internal operations and change the services that you deliver.

Well, thank you very much. Look, it’s been a fascinating discussion. I’m sure we’ll continue afterwards as well, but thank you to Arne Mosselman and Nathan Kemp and this was the Grayling Vantage Point podcast. Thanks very much.

Thank you, Scott.