Transcript for Cutting Through podcast Episode 1

Return to the podcast episode page to listen and find additional resources

Jonathan Holt   Welcome to this first ever episode of Cutting Through, the podcast for corporate digital communicators. I'm Jonathan Holt, Head of Strategic Insights at Bowen Craggs and we're hoping this will be the first of many conversational forays into the issues and challenges facing corporate digital communicators today. In this episode of the podcast, we're going deep into the promise and potential pitfalls of generative AI. 

AI promises to catapult efficiencies in ways the world has never seen, but does the step change in productivity come with a cost to things like trust, believability, and human connections? Efficiency goes head-to-head with authenticity in this episode of Cutting Through. 

With me in the virtual podcasting studio are Scott Payton, CEO of Bowen Craggs, and George Barrett, Vice President, USA. Hello to you both. 

Scott Payton   Hello. 

Georgia Barrett   Hello. 

Jonathan   So, since this is the very first episode of the podcast, and there might be some people listening who aren't familiar with Bowen Craggs, Scott, can you give us an efficient and authentic quick snapshot of what Bowen Craggs is and does? And feel free to consult ChatGPT if you need to. 

Scott   Well, do you know what, Jonathan, I did actually ask ChatGPT this question. So, I typed in yesterday the question, who is Bowen Craggs? And should I tell you what ChatGPT said?  

Jonathan   Please do. 

Scott   OK, so it says Bowen Craggs is a specialized consultancy firm that focuses on corporate digital communications, which is absolutely right. It says that we were founded in 2002 and based in London and New York. That's also completely true. 

The company helps large organizations to optimise their online communication channels, including websites and social media. ChatGPT is on fire. Can't fault any of this. We are well known for our unique benchmarking methodology, which evaluates the effectiveness of corporate digital communications across various industries. Completely bang on the money. Apart from we actually work across all industries, not just various ones. I'm being picky. 

And then it continues. Bowen Craggs clients include some of the world's most prominent companies, including BP, Estee Lauder, GSK and Target. ChatGPT is correct. And Bowen Craggs also fosters a strong community among corporate communication professionals, offering opportunities for networking, peer support and sharing best practice. So, it's all correct. It then goes into more detail, which is a bit wrong. So, I won't read that out. 

Jonathan   That sounds like Bowen Craggs has a fairly decent rapport with AI so far. So that bodes well for our conversation today. We've built this conversation as efficiency versus authenticity as two different sides of the AI experience in the corporate context. 

Let's start with efficiency. Since there's clearly a strong case to be made around AI and productivity, Georgia, what are some of the ways that corporate digital communicators you're talking with are using generative AI to save time or cost or both? 

Georgia   It's a great question. We recently wrote a report which is called Secrets of Corporate Digital Leadership and that's where we spoke to some of the teams who manage the world's best corporate websites, and we asked them how are they using AI in their day -to -day jobs. So, some of the things we heard were producing first drafts of articles and blogs, especially if they're not written in the first person.   

So, as we all know, AI is very useful as a research assistant, especially for highly technical sectors such as energy. It's also really great for generating ideas so that you're not faced with just a blank page when you want to start writing. So similarly, producing draft summaries of long documents like an annual report or a sustainability report is a good efficiency use case. Also creating graphs and charts from raw data. 

Proofreading, grammar checking, checking that your text is compliant with your style guide or is in the tone of voice of the company, doing first drafts of language translation, and adjusting images to fit specific dimensions or getting rid of background clutter. And then there are other tasks which are more specific to corporate digital communicators, such as writing metadata and alt text, all of which improve accessibility and SEO. 

So yeah, at the moment we're finding that AI and corporate communications isn't actively generating revenue for companies, but it is making those very time consuming and arduous tasks that come with managing a corporate digital estate more efficient, which frees up more time to do the high-level thinking and the creative work. 

Jonathan   So quite a lot of uses really, even at this early stage. Scott, where would you say that AI is making the biggest impact in corporate digital communications? 

Scott   Well, as well as all the examples that Georgia has given, which many of our clients are using, as she says, I think it's important to remember that lots and lots of software tools that have long been used in corporate communications are being all the time being infused with all sorts of AI features that are automating tasks that used to be just people having to do often boring tasks that are being made quicker and easier. This morning, I was on Instagram, and I saw an advert from Photoshop that was highlighting all of the fancy new AI image alteration tools that it offers. And it had a picture of a mountain. And you could click an AI button to add Northern Lights above the mountain. So, it's sort of creeping in everywhere. 

Jonathan   So, without necessarily naming any names, can either of you cite any specific use cases for AI among companies? 

Scott   Yeah, we're talking about the Photoshop example. We know there's a big client in the US that we have that is using AI to alter images. So, for example, tweaking a landscape photo so it works well in portrait mode on the website.   

Georgia   I can also think of a few examples. So, one of the most interesting ones to me is that some clients have been talking to us about AI powered crisis toolkits. So, they have an archive of past statements, announcements, materials that have basically been pre-approved with the correct content and phrasing so that if another crisis that is similar in nature was hit, then the AI tool can use the past crisis materials. 

So, when you're responding, it's a lot quicker to get that out the door. And I think at this point, it's good to mention that a lot of companies are developing their own version of ChatGPT, which is trained on company data and tone of voice. So, this really avoids the risk of disclosing company secrets. 

Also, a lot of clients, basically all of them that we spoke to, they're using AI for team meetings, creating notes and agendas, finding those insights and summaries. Yeah, and I think there's a lot of excitement about reducing the workload of running the corporate website day to day, but... with stuff like this it just introduces additional challenges and errors and issues, and I think there are huge reputation risks involved so if we're automating some of the kind of quote -unquote boring tasks then we do need to be careful about what that could mean for the corporate website. 

Jonathan   That seems almost like a natural segue for us to move on to talking about authenticity. But before we do that, I just wanted to pick up on the fact that Scott mentioned that AI is really seeping into everything. We are now starting to use it even in our work in Bowen Craggs. I think most people who have office -based jobs are. In my own personal experience, for example, trialing out Google's Notebook LM as a kind of productivity aid to help me draft things and whatnot. Sometimes it seems to save me time, sometimes it seems to cost me time. And it really seems to take a number of attempts for it to figure me out and me to figure it out. So, I guess I'm just wondering, how are you both using AI? What are you finding to be useful? 

Scott   Well, for me, we often produce very long benchmark reports for our clients and analyzing the strengths and weaknesses of their web presence. And they’ve kind of been tens of thousands of words long. And then we have to produce an executive summary that pulls out the key points. So, I think Notebook LM can be very helpful for that. And it sometimes pulls out points and trends in the report that I might otherwise have overlooked. Talking about of ChatGPT itself. I'm not obsessed with it, by the way, not on it all the time. But it can be a really good research assistant. But I think it's important to remember that it's a good research assistant. So long as you're not looking for very recent information, because it hasn't been trained on bang up to date news. So yesterday I did ask ChatGPT who is running in the 2024 US presidential election. And it listed loads of people, including Joe Biden, but not Kamala Harris, because its data is out of date. So, I think that's a kind of one health warning for ChatGPT. But it can be, you know, it can be a really, really useful, powerful research assistant. 

Georgia   Yeah, I think the grain of salt thing applies to how we've been using Read AI to record internal meetings and only internal meetings because the results have been quite mixed, and the goal is to deliver summaries and recordings and insights from meetings but it's just not quite perfect yet. 

Scott   Yeah, and I think Read AI is an example of a tool that's probably going to be amazing in a year or two, but at the moment is kind of less than less than amazing, shall we say. 

Jonathan   Presumably all of the AI tools that we're using now could potentially be quite different within a few months. And not to get off topic, and I think this sort of segues almost into our next topic, but I sometimes wonder why these tools have to rephrase things and then the tone is not what I would use, not what the user would necessarily use, but I guess that's just an added warning that we have to not just go with what the AI says, but look closely at it and use it as a collaboration tool rather than as an actual author in and of itself. So, let's move on then to authenticity. And Georgia made the case earlier for efficiency. So, Scott, would you be willing to make another pitch for authenticity as the aspect of AI use that digital communicators need to take seriously?   

I guess the question is why is authenticity important in corporate digital comms and how does AI threaten it? 

Scott   Yeah, I will do that. I think I feel kind of quite strongly about this. Our research, some of the research that Georgia referred to and research we did specifically on what Generation Z, Generation Z are looking for from corporate communications clearly shows that from job seekers to customers to investors, journalists, particularly younger members of those groups, they have a stronger thirst and demand for authentic messaging from companies and from everywhere else than ever before.  

People are tired of fake news, empty promises and corporate cliches. And in a world where more content than ever is being artificially generated, I do think that handcrafted messaging and material will stand out more and cut through more if it genuinely feels real. And I think it is interesting that there is rightly a lot of debate going on at the moment about the dangers of AI in terms of misinformation and disinformation. But I think in corporate communications, another risk of often underappreciated risk of AI is overuse and abuse of AI is the increase of dead-eyed, vanilla, same-same content that doesn't stand out from the crowd or emotionally-engage people. So, I think there's a real risk here with overuse of AI because companies trying to kind of cut through and get their message across, maybe shooting themselves in the foot if they overuse AI. But there's a great opportunity for people to be creative and to stand out from an increasingly vanilla crowd. 

Jonathan   Well, just the other day the New Yorker ran an article titled "In the Age of AI, What Makes People Unique?" And it sort of highlighted for me that the role out of AI has brought up some existential questions that we used to think of as sort of the preserve of philosophy department at universities and things like that. And now it's being brought into everyday life. And one answer to that question of what makes humans unique, and the world of AI is that they can feel, you know, machines can be programmed to behave like they have feelings, but they don't actually have feelings, at least not so far. So, you know, when it comes to cutting through in the age of AI, how important do you think things like humor and wit and making an emotive connection are? 

Scott   As I say, I think with more and more AI content swishing around the world, messaging is much more likely to resonate if it genuinely feels like it's coming from a living, thinking person rather than a robot. Going to try to draw and I think I came up with a metaphor when I was walking the dog, which may or may not work, but we can hit. Are you ready for my metaphor?  

Jonathan   Go for it. 

Scott   OK, so and I feel slightly awkward talking about this because I'm not in this person's target market, but there's a musician called Charlie XCX who this summer, a few months ago, released an amazingly successful album called Brat and it was marketed and I think anybody under 30 who will listen to this will know what I'm talking about and no one else will, but I'll carry on. So, the Brat album was marketed via a slime green colour and a kind of an aerial based font that was deliberately designed to look a bit wrong and jarringly imperfect. It looked like it was kind of thrown together in a bedroom. It was kind of handmade. In actual fact, it was carefully designed over five months by a New York designer called Brent David Freeney and who looked at 500 different shades of green before he and Charlie settled on this deliberately horrible final one.  

It's kind of, I think, it's about as far away from AI-generated content as you can get. And I think it's quite an interesting kind of cultural rebellion against this kind of photoshopped corporate material that's increasingly pervasive and to hammer home the realness of Brat, the marketing team launched a Banksy kind of style green wall mural in Brooklyn that had different, I think was every week or whatever, different messages painted on it from I think it was lyrics from the album. And a fun fact, the wall is two blocks from Bowen Craggs' New York office. But I think, you know, I think I think the whole Brat thing, which has been kind of hijacked or exploited by the Kamala Harris campaign. It's everywhere like Barbie Pink was last year. And it's very, very un-AI generated. And I do think it illustrates the fact that a big risk with AI contents that unlike Brat kind of resonates emotionally and psychologically with its target audience. 

But your content, if you overuse AI, it will feel dead to your target audiences because there's no real life behind it. 

Jonathan   So, Georgia, what are your thoughts on the Brat example? Presumably we're not saying that large companies should start to use that term in all their social media postings. 

Georgia   I think there are huge dangers of jumping on a trend just because it's viral and it just not being appropriate for you as a company. And yeah, it's one of those examples of corporate humor going very, very wrong. I was just reading a Guardian article about if there's anything that is not Brat, it's the bank industrial complex. And that is in response to Deutsche Bank basically getting on this trend in a way that is just not funny at all. So yeah, there's a health warning there. 

Jonathan   It might be funny in German. 

Georgia   Maybe. 

Scott   What did they, how did they refer to it? What was the... 

Georgia   So, they posted on their Instagram account, we're looking for a “Brat in Finance,” which is obviously two different cultural moments coming together on their corporate channels and there's been quite a big backlash just saying it's just not the spirit of the trend. 

Scott   That's the most tragic thing I've heard this week. 

Jonathan   Well, it comes back to the need for having people who are savvy about the cultural trends if you're going to try to become part of that moment. 

Georgia   Especially if you're a company, I don't think you can get involved in every latest trend that's happening in the world of culture. 

Scott   Yeah, and I think I think even the Kamala Harris example, she was invited in because I think Charlie XCX did a tweet, didn't she? Kind of saying, “Kamala is so Brat.” So, she was she was almost given kind of permission to run with it. And I wouldn't imagine Charlie XCX giving Deutsche Bank permission. 

I think jokes are important. Jokes are important. Laughing is important. And I think it's kind of interesting to see how the Kamala -Walz campaign, as you say, has harnessed that. So, it's taken to kind of issuing these statements on social media rather than kind of traditional press releases and kind of pithy funny statements. 

And I think there has been recently a lot of discussion about whether AI content, whether it's video images or text, can break through or has already broken through the uncanny valley, where there's if something doesn't feel quite right, there's a kind of emotional reaction. So, you have this kind of instinctive revulsion to it. And there's been a discussion about whether a press release that's been AI-generated, whether it is distinguishable or not from something that's been written by a human. But I think even if it has AI text, AI press releases have broken through the uncanny valley in the sense that it could have been written by a human. I think AI content is still in danger of falling into the kind of boring valley or the tedium valley, because even if it doesn't feel creepily artificial, it won't have any heart to it. And people can sense that. It doesn't matter if it's an investor relations press release, doesn't need to have a heart to it, but it does matter if it's a message from the chief executive about climate change or a message from the HR department about company culture. I will shut up in a minute, so Jonathan. 

Jonathan   You referenced the uncanny valley and arguably what's happened is that AI generated things have become so real, deep fakes, images of people who aren't real but appear to be real, that in a way cast a shadow of uncanniness across reality. You also... lot of the way that a lot of us... 

Scott   Mm.

Jonathan   Certainly, I became conscious of Brat was through the tweet that was directed at the Harris campaign saying that Kamala is Brat speaking a language that, you know, many people like me didn't know yet how to speak. But it seems to have sparked something really interesting in terms of political dynamics, where, you know, the Harris campaign has really embraced the witty version of humanity. So quite relevantly, in a for this conversation, recently Donald Trump accused the Harris campaign of faking the people at one of her rallies and said it was AI, which I don't think it was. But how does one respond to that? Well, one way that the Harris campaign has responded is through humor, through using social media to say things of Donald Trump's campaign, such as, well, we don't know if it's artificial intelligence, but there's certainly no intelligence there, which would be funny or not depending on your point of view, politics is fraught. But it is a really interesting example of how just being real and human and doing things that the technology can’t do can help to cut through. 

Scott   Recently Elon Musk interviewed, had a conversation with Donald Trump on X. Afterwards, Harris campaign issued a statement saying, Donald Trump's extremism and dangerous Project 2025 agenda is a feature, not a glitch of his campaign, which was on full display for those unlucky enough to listen in tonight during whatever that was. Trump's entire campaign is in service of people like Elon Musk and himself, self-obsessed rich guys who will sell out the middle class and cannot run a live stream in the year 2024. And that got 80 million views in two days on X alone. So, yeah, I think it's another thing that I suspect AI can write jokes. 

I don't know if they're funny or not, but I think wit and humour is absolutely another powerful way of cutting through. 

Jonathan   Georgia, you and I did some investigation into the role of humour in corporate communications when we looked at, we talked to younger corporate stakeholders a couple of years ago. And one of the takeaways from that, I think, was that it needs to be handled with care in a corporate context. You know, the political examples that just been given are examples of political people sort of throwing caution to the wind and embracing the brave new world. But that has not happened by accident or without care. As I understand it, there's a whole team of very young and social media savvy people in the Harris -Walz campaign, know, sending out those funny ha-ha tweets.   

What's your view on why the companies should be doing this now? 

Georgia   Yeah, I mean, it needs to be appropriate and your point about having younger people on the team, I think is a really valid one if you're going to be making jokes as a company. It's like with the use of corporate TikTok, for example, not every company, I think it needs to be treated with care. Although we're recently seeing more companies start to use TikTok as a corporate channel, which, yeah, I think is very... interesting and it's authentic and the best thing about something like TikTok is it's very selfie style, it's very speaking directly to the camera, and we see that in the best, most authentic forms of employee communications on a corporate website. They tend to be selfie style videos or employee profiles and features, so TikTok is really kind of made for that medium. Yeah. 

Jonathan   It is really interesting that we're seeing more companies who are getting on TikTok, companies that we wouldn't have initially expected to like Chevron, for example, Sanofi, quite corporate entities, but who are daring to go there. Well, so also Georgia, you've been involved for a while now with the Content Authenticity Initiative, and that's alongside people from all sorts of different organizations around the world who are looking at how to maintain authenticity, given partly the role of AI. What have you learned through that about the role of authenticity in corporate digital communications specifically? 

Georgia   Yeah, so the Content Authenticity Initiative is this open-source coalition, I guess it's all about promoting content transparency and provenance. So, the idea is to create a secure end-end system so that you can track the creation of content so you can see where it came from and if it's been edited or modified by AI or by anything, all in a tamper evident and transparent way. 

And I think this is really interesting because it's kind of the, it's like the solution to a world where we don't know what's real and what isn't is to have things like this. And Verizon has actually been doing this for quite a while, but using blockchain technology. So, they've incorporated that into their corporate newsroom on their press releases. So, it means that all of their official news releases are permanently logged via blockchain technology. So, any edits are visually tracked. So, I think the takeaway for corporate communicators is to embrace things like the Content Authenticity Initiative and to make transparency and accountability a business priority if it isn't already.  

And we're seeing with AI search, for example, so we have generative AI search overviews on Google. Google is now becoming a single version of the truth. So, if you type into Google, "What is Nestle's stance on climate change?" it will serve you a sentence that has been AI generated. And if you are Nestlé, you really want that sentence to come from Nestle.com as your single version of the truth. And I think the corporate website is the place to be the authority and to be the, well, the authentic version of the truth. I feel like Scott has some better thoughts on that than what I just said. 

Scott   No, at all. Not at all. No, I thought that was very good. That was great. And I think it's an interesting, I think everybody does need to be kind of open minded about all of this and just start and kind of there are lots of the world is changing, audiences' expectations are changing. The way that people are looking for information about companies is changing. The way that information is being delivered is changing, all amid a world where it's increasingly difficult to distinguish fact from fiction. 

Jonathan   Well, absolutely. And I think the Google AI overviews is a really interesting case. I've been looking at those myself recently. And what I observed there is that it brings the trust element out more than other types of search ad, even featured snippets. It makes it very obvious where the algorithm trusts the information and doesn't. 

And you can see that by this, well, they say they're not citations, but by the links that accompany each overview, sometimes it may be just the organization in question, but other times it's critical voices. And the way, there are a lot of things that undoubtedly go into that and very technical SEO type things. But one of the fundamental ones is whether the information seems rounded, whether it's actually addressing the topic at hand, head on and answering questions. 

Those sorts of things. You know, that's one example of how companies can cut through as the world rapidly changes to be more AI-powered. Now, you know, historically, even long before AI companies still had trouble cutting through still had trouble seeming or being authentic. And we've seen that, know, at Bowen Craggs over and over again, whether it's, you know, pictures that don't look quite real or that you can't tell or words that sort of lack any sort of personality or substance to them. So, and, you know, even the word corporate to many people conjures up something that's inauthentic, that's kind of bland or faceless and anonymous. Is AI changing the stakes on this? And what should companies be thinking about if they want either generate content using AI or make sure their content stands out in a world where others are doing it. 

Scott   I do think that an over reliance on AI can amplify what, as you say, are the kind of perennial pitfalls of corporate communications, which is bland statements written by committees being stayed in impersonal. I think that is kind of there's a risk of that being kind of turbocharged by AI. 

I think if, as Georgia says, if those tools are harnessed to automate the boring stuff and free up time for more human creativity within communications teams and other parts of business, there's actually an opportunity for corporate content and messaging to be more human and authentic than before. If the robots are kind of doing the alt text and lots of the other kinds of time-consuming things and writing the financial press release or at least the first draft of it. 

So, I think I think it's and I think also there's an opportunity for companies that are bold and brave and funny, human to really kind of stand out from the crowd because the crowd is going to get more vanilla. I don't quite like vanilla as a taste flavour, but you know what I mean? You know, there's samey. So, I think there's if you look at it, it's interesting if you look at LinkedIn, which was never the most charismatic and emotive of social channels, you know, it's quite corporate. But there has been a kind of a disease of AI, AI generated posts. And you can get a sort of feel for it is a bit like the uncanny valley, actually. You know, you can kind of there's a certain type of bullet point, emoji, kind of corporate, often quite sales and marketing posts, which is AI generated. So, you can increasingly see how AI is altering the kind of tone of corporate messaging broadly. And I think, you know, just like I'm going to stand by my Brat analogy, even though I'm 48 and I, you know, because I do think it is interesting to kind of see what really does kind of cut through. And I think there is, you know, multinational corporations aren't 32-year-old pop musicians and multinational corporations are not, you know, they're not politicians either, they're not political parties. But I do, I genuinely think it's kind of, it's well worth companies looking outside the corporate world to see what sort of messaging, you know, is cutting through at the moment. 

It's not just a kind of case of it being real, but it needs to kind of feel, it needs to feel real and it needs to kind of say something and it needs to say something that genuinely feels like it's been written or designed or thought of by a real person with feelings. 

Jonathan   Which companies historically have not been necessarily brilliant at. I mean, it doesn't take an AI bot to write something that’s bland.  

Georgia   Yeah. 

Scott   No. 

No, humans are brilliant at writing bland things as well. Humans are great at writing bland things, particularly if they have an enormous committee of people to approve it, particularly in a kind of in a risk averse and volatile environment like that we're in now. I mean, think it's very easy. I think there's a lot of pressure on companies to write things. We can kind of sit here and say what it's important to be bold and brave and cut through. 

Jonathan   We're very good at that. 

Scott   But we're in an increasingly polarised world where it's actually quite easy to say something that could lead to some kind of backlash. 

Georgia   Yeah, we had an example of a company where we said to them that their imagery on the site is a bit cliché. It looks like it's taken from a stock photography library. And they turned to us and said, no, it's real. Like we, those are our real employees. This is our office. We just created these photos. 

So, I think it's important to have trust signals on content that has been created authentically. And a good example of that comes from Sanofi and they have, if you go to their corporate website, all of their images are captioned with the name of the person who is in the photo and their job description and where they are in the world. So having those named authors, photos of the author, photos of the person, I think all of this is increasingly important. 

Scott   I totally agree. a thread through this is, is people. You know, I think a lot of this is about having people, whether it's employees or the chief executive or, customers or, charities that people who running or benefiting from charities that a company is working with, it's about giving the space for real people to talk about things in their own words. And looking again at the Kamala Harris campaign. It's not just humor that they're using. It's also there was there was a video of Kamala Harris ringing Tim Walz to kind of say, would you be my running mate? And of course, it was completely staged. But actually, they did a blooper reel.   

They showed some outtakes at the start because apparently Kamala Harris rang as planned and Tim Walz' adviser didn't recognise the number and thought it was a scammer. So, Kamala Harris went straight to voicemail, and they filmed it and showed it. And think it's that kind of... “And there was laughter!” So, it's showing the people, whether it's a presidential candidate or a chief executive, are real.   

They're real and they're flawed, and they find things funny. 

Jonathan   Well, that's a really interesting example because it kind of proves both points faked communication can be kind of cringy and counterproductive. But if you can then follow that up with something that's real, then you've kind of won, I guess. Within a corporate context, we see all the time companies that will be extolling the virtues of how wonderful their company is as a workplace, only to be showing you pictures that are clearly taken from Getty images of people in front of glass walls with colorful post-it notes and things. And that's just so the opposite of what we're talking about here in terms of building a reputation for authenticity, but at the same time, it gets harder and harder to tell. 

We used to say at Bowen Craggs think that people can always tell when an image isn't real people, but I don't think they can. And I was looking the other day at GSK's website on the homepage, and there are these quite large format pictures that are very clearly real people. And yet, there's no GSK branding. There's no attempt to color those images so that it fits the brand palette or anything like that. 

And it's hard for me to say exactly why I'm so sure that those are real pictures, but I can tell. And so, I think that's maybe an example of this reverse uncanny valley that we're now living in that there's something, there's a certain something.   

Scott   There's there is I think there's another thing I'm kind of I'm talking a lot about outside the corporate world, but I think it's kind of important because I think the corporate the world of corporate communications can be a bit of an echo chamber. And I think there's so many lessons that can be learned from journalism and politics and elsewhere. But I've noticed I'm not particularly in the Marvel multiverse or whatever it is. But there's a thing I understand there's a film Deadpool which is a new Marvel film. 

And frequently in this film, they're breaking the fourth wall. You know, they're kind of making in jokes about, the kind of ownership rights of who, which, whether Sony or Disney, own which characters in the film, you know, and again, it's very it's quite, you know, it's quite meta. It's postmodern. What is I'm not quite sure what it is, but it's an acknowledgement. It's a kind of nod wink to the audience that we kind of know this is actually, this is a pantomime. This is not real.  

And I think you're kind of seeing quite a lot of that in culture at the moment. It's almost like this kind of nod to authenticity, which is actually kind of being stretched further than I think it has in the past. Not quite sure. Georgia. 

Georgia   Yeah, in culture but also in corporate communications and I think the research that we did recently into what the next generation wants from online corporate comms, a lot of it was behind the scenes glimpses and if you're a jobseeker evaluating if you want to work for a company you do want to see that behind the scenes glimpse into the culture and we're seeing it in corporate governance videos that show the board at work doing their day-to-day activities and not just a statement written down about what the board does. You know, you see an interview with the chair, so I think these kind of real glimpses that acknowledge... 

Jonathan   Well, know, one another answer to that question of what makes humans humans versus robots or, you know, machines that are not is that humans make mistakes, their flaws, and arguably the machines are being programmed to be as perfect as possible and to reflect a kind of echo chamber of something close to perfection.   

Scott   Can I fact check myself Jonathan? Can I fact check myself? It's Deadpool and Wolverine. Deadpool and Wolverine I'm sorry. I just had to ask ChatGPT about that. I googled it. I googled it. Sorry. 

Jonathan   Sure. Yes. 

Well, so the question I was sort of leaning into was, you know, does that mean that in order to seem real companies need to sort of muck it up a bit, be a bit more rough and ready? 

Scott   It's a really interesting question. don't think they should muck it up, but I think they should certainly show their... I think as Georgia mentioned, should show where there's still work to be done. They shouldn't pretend everything is perfect. They should be open about the fact that they are... the past hasn't been perfect. And they should, I think, you know, kind of, as I said before, kind of give, their employees a kind of platform to kind of talk about the business and what the business stands for and what its values are and what it's doing in their own words. So, I think they should definitely, yeah. And I think they should definitely show their human side, more. 

We worked with a big global bank, which has a very, very polished brand. And it's very, very polished in its marketing output. It's very professional and slick throughout on its corporate website and elsewhere. 

And actually, that kind of led to a bit of a challenge when it came to providing content in the careers section that feels authentic and real, that's going to resonate with job seekers. Because initially they kind of applied the same kind of heavily edited, highly polished approach to things like employee testimonials. And it kind of, was on brand and it was highly professional, but it undermined their attempts to come across as being authentic and real. We work with them about ways of getting around that. So, I think it is a challenge, but think companies kind of, know, lightening up a bit more and showing their human side is probably the kind of key, rather than actually mucking it up. 

Jonathan   There's a real parallel there, I think, between some of those highly polished but somehow not impactful, not believable videos that some companies make, and the AI generated social media posts that you mentioned earlier, Scott. I mean, the reason those don't work or don't really resonate, even though they may get lots of likes and shares and whatever, is because they're kind of dead behind the eyes. They're not conscious of the fact that whoever's reading that probably already knows the things that are so matter-of-factly being stated about whatever the subject is. Because in many cases, those things are being put out into groups that are for people who are aficionados and whatever that topic is. So, there's something here about being aware of the audience and letting the audience in. That seems to be a theme in the behind-the-scenes notion of authenticity.    

Scott   Mm hmm. Yeah. Yeah. Yeah. And I think not to be trying to be something you're not because I don't think it necessarily needs to be kind of being we've talked about humour, for example, about being kind of being different. But I think I think being boring can be appropriate, too, if it is genuine. And I think, you know, I'm to use another political analogy. Keir Starmer is the new prime minister of Great Britain. 

He's a very safe pair of hands and he's a kind of, you know, he has a background as a prosecutor and he neither tried, he didn't try to be kind of wacky or interesting for its own sake, but he kind of came across, I would argue, with a number of kind of swing votes as being someone who actually seems to be he is what he says he is. He is what he is. Someone said to me once that, you know, there's a lot of trends for companies to be innovative. A lot of customers don't necessarily want their bank to be innovative. They don't want the pilot flying their plane from New York to London to be innovative. 

But you want them to be kind of a safe pair of hands. I think if you can, think authenticity can manifest itself in lots of different ways. And I think a lot of it is about kind of being true to, true to the, you know, what the company, the company's culture and just to be open and honest. 

Jonathan   And true to the individual who happens to be speaking as well, presumably, you know, I'm sitting in North Carolina where our state motto is to be rather than to seem that those seem apt words for how to proceed in the age of AI, not to not try to be something that you're not. And as you say, if that means not being the witty, you know, jokemaker, then perhaps it's best to embrace that. Can we talk just a little? 

Scott   I'm sure William Shakespeare, was just going to say, it's quite almost Shakespearean state motto you have. But yeah. 

Jonathan   Well, it's “esse quam videri,” so it's taken from Latin. Very highbrow, which doesn't seem to match the behaviors politically or otherwise in our state or the world today. 

Scott   Mmm. Yeah. 

Is the state motto of Oklahoma, "Oklahoma, it's OK"? Or is that, it's not, was that the, there's nothing, is the state motto the same thing that's on the, it's not the same thing as on the car registration, that's different thing. Right. I don't know. It's not like the sunshine state. OK, right. Yeah. 

Jonathan   Well, I think that might be a catchphrase. 

Well, I don't know, it could be, but I think that might be an advertising slogan or something. It's quite clever. Can we talk for just a moment as we move towards wrapping up our discussion about some of the nuts and bolts of these things? And I guess maybe imagery is a really good one, a good example to dig into a little bit because we're aware, and you mentioned at the top, I think, Georgia, that some companies are starting to experiment with using AI tools, which are now being backed into virtually every bit of software to alter images in some way. I know I've had conversations with corporate communicators who are starting to do that, sometimes with abandoned, sometimes with more caution. Where are the boundaries? What should people be thinking about if they're considering taking that leap or even if they already have and just aren't quite sure where it's all leading? 

Georgia   Yeah, I mean, I think if something has been AI generated, especially if it's a photo or a video, it is very imperative that you need to say that it has been AI generated and you need to do that clearly and transparently. But then the gray area is, guess, if you are doing something very small scale, like almost like a Photoshop, you you're turning an image that is portrait landscape, those sort of more technical things. I think it depends on the context. 

But it doesn't seem like you should have to announce that on your corporate channels that you've used AI to make the photo a different size. But I think if it's to do with people, if it's a quote, for example, from an employee or an executive, then that should absolutely come from the real person. And I think the overarching message for both photos and text is that we should be using AI as an assistant, like a design and editorial assistant as opposed to an editor in and of itself. We always need that human oversight. 

Jonathan   A lot of the things that AI, some of the things that AI can do subtly play around with the colours in an image so that you look at different, the various images of the page and they have the corporate blue, corporate red, corporate orange or whatever, or have traces of that so that it sort of supports it as a sort of brand-building thing. 

I think, guess, that companies are not media outlets. Well, some are. Most of them are not media outlets. And so, maybe don't have to quite hold up to the same standards as the New York Times or the London Times or whoever would get into deep trouble if they started publishing AI-generated images and not acknowledging it. But can we say maybe that there's some sort of loose litmus test there that some of those basic things that where AI is simply saving you time versus less or more rudimentary software options that maybe it doesn't need to be declared. But presumably, if it's a case where the person is being generated whole cloth, it would be a good idea to declare it. Or maybe we can say, don't do it. The default should be not doing that, because in general, you'll be helping your company better if you can't manage to capture real pictures of real people. 

Georgia   Yeah, I guess it's the efficiency versus authenticity thing. If you're using it for efficiency, then I think that's okay with the kind of safeguarding of using common sense. And the overarching thing to bear in mind is authenticity and just using real people as much as possible. So yeah, it's kind of two different use cases. 

Jonathan   It really is.   

Scott, any thoughts on that, particularly maybe around text generation using ChatGPT or other tools? 

Scott  Yeah, I think we haven't talked to talk about the kind of the data privacy risks of inputting kind of company information into ChatGPT or these open, open large language models. Obviously, there's a kind of loads of security and data privacy risks around that. I think when it comes to creating the first draft of things, I think when it comes to text, I think it is generally going to be fine as long as the thing that is generated has been then edited and refined by real person. I think, as I agree with Georgia, that the no-no there is kind of generating quotes from real people like employees or the chief executives. I think there's already often been a dark heart. Statements from the CEO to not to be written by the CEO in many cases. They were hopefully signed off by the CEO, but not written. I mean, people have, you everyone has speech, there's speech writers and there's, you know, there's ghost writers. I think AI, some of this isn't kind of necessarily as new as maybe it appears because I think a lot of the old kind of it's common sense, about just to be kind of just to be open and clear about what is, who has written what or what has been, you know, where does something come from? But I think if something has been finalised, refined, edited, signed off by a human, then I think that that's really the key. Imagery, I think, is a different kettle of fish particularly images of people because I think there is a, if something has been substantively altered and you're not disclosing that, then I think by and large you need to kind of the default position. It needs to be, that's a problem and it needs to be disclosed.   

Jonathan   Well, this has been such an interesting conversation, and we've unearthed lots of issues that I think are going to continue to need to be unpacked. 

To wrap up, I'll ask you both for your top takeaway. And what should corporate digital communicators and really anyone who cares about corporate reputations and bottom lines be taking away from this? Georgia? 

Georgia   So, for me, I think AI makes being authentic and transparent more important than ever. And it's always been a huge priority, I think, for corporate digital communications to be authentic and transparent. But I think we're seeing a seismic shift where the corporate website and your corporate social media channels are becoming the place to do this and to be the authoritative and genuine authentic place to kind of say who you are as a company and what you're doing. So, I think the corporate website is going to be more important than ever in light of what's happening with AI and AI search in particular.   

Jonathan   And Scott, any parting words? 

Scott   I think there is research does show that as there's trust in politicians and traditional media is eroding, and people are increasingly looking to corporations to get information for factual correct information about the world. I think if you use my main advice is to use AI automation to free up your creative team to be more creative and engaging about the business and the world in a factual humorous witty slash deleted appropriate, but engaging way. 

Jonathan   Well, thank you both very much. 

Scott   Thank you. 

Jonathan   And thank you very much listener for being with us for this first foray into podcasting conversation. We hope it will be the first of many and that you'll join us along the way. Meantime, if today's conversation has sparked any questions or thoughts, ideas, world saving concepts that you'd like to discuss with Bowen Craggs you can find us on bowencraggs.com.  

Return to the podcast episode page to listen and find additional resources