John O’Brien on Why Comms Needs to Go All In on AI

Mixing Board Studio Session

John O’Brien is the Co-Founder of Strange Brew Strategies, a tech focused communications consultancy that has partnered with innovative companies, institutions and venture firms, including: American Express, Arweave, Cloudflare, Deel, Flexport, GitHub, Lightmatter, Paradigm, Plaid, Relativity Space, Rivian, Shopify, Tezos, Twitter, Yuga Labs and many others. Before starting SBS, John has spent most of his career at agencies like Weber Shandwick, Horn Group and The Outcast Agency.   

Strange Brew has been helpful to the Mixing Board community so we thought it would be fun to show our appreciation with a Studio Session. Plus, John has a lot of interesting takes.

In this conversation, John and Mixing Board founder Sean Garrett talk about why aspiring comms practitioners should start their career at an agency; making the comms industry more innovative; how to ask for equity; and, lots on how AI is going to change comms and what needs to be done about this.

SG: What drove you to create Strange Brew in the first place? 

JO: I’m one of the few people who have been in the industry for almost 20 years and have never worked in-house. I’ve been all agency side partly because I just really enjoy the variety of working with new clients. Back in 2016 a lot of agencies bleeding-edge was enterprise cloud when I was personally interested in things like artificial intelligence, blockchain, space, the new space economy – those types of things. When we first started, we positioned ourselves as a next frontier company because we felt like that was an underserved market in the comms world. We had a lot of really deep tech clients early on, like The Allen Institute for AI and Relativity Space. Back in 2017, 3D printing a full scale rocket was just complete science fiction. Two days ago they launched it. 

SG: Part of why I love talking to you is that you're both a tech enthusiast and you’re a comms geek. You get into tech and you are super interested in what's behind it, how it's built, what the ramifications are. And you approach comms in the same way. You want to disassemble it, to reassemble it again. Not saying anything shocking here, but comms has been an industry that’s been somewhat bereft of innovation. If we’re going to be working with the most innovative companies in the world that are changing paradigms, what do we need to be doing to change our own paradigms?

JO: In some ways, you need to throw spaghetti at the wall and see what sticks. That's how a lot of innovation starts. And for one reason or another, the comms industry has been really conservative and slow to change. That's probably the reason why there's a lot of negativity around the industry. Journalists frequently complain about comms people because they have not changed the way they do business to adapt to the modern world. 

For us, we're just willing to try new things. Over the last few years, our clients have moved from email to Slack. Slack is much more interested in communications. So we decided, let's start a fully Slack-based comm service called Onboard, to service seed stage or early stage startups who may not need the very formal approach that comms agencies usually have to justify large budgets. That's been working out really well. 

Everyone, especially in the media relations world, needs to adapt to the fact that journalists don't want to have formal information transactions. 15 years ago, the industry was very formal – you were to sit down with a journalist, you were to deliver a news announcement, and not talk much outside of that. Now it’s much more like a  conversation over beers. You sit down, you talk about the industry broadly, you have unique perspectives on things outside of just your company. 

The table stakes information is the stuff that's on your website. With clients, I don't worry about how you describe your product. I want to focus on why you matter.

SG: The dynamic that exists now is fluid, where information is relatively porous between a company and those on the outside – that includes journalists, but certainly they are not the only audience. But there's still a lot of founders (and a lot of VCs who are coaching the founders) who are still operating in a world where, it's about, “Hey, when do we do the press release?” How do you help founders adapt to the current realities versus the way things worked previously?

JO: We tell founders that they need to be frequently communicating and commenting on how macroeconomic events or macro political events are going to affect their company, but also their industry broadly. I tell clients that comms isn’t successful as a campaign by campaign approach anymore. You can’t just go news announcement by news announcement. You have to weave news announcements throughout a broader thought leadership campaign.

And thought leadership can be one big idea with lots of threads, or it could be lots of separate things. But it's important to have conversations about everything, especially now, in the age of AI. In the age of synthesized information – people are going to be searching for the best companies in X sector – whether it's cybersecurity, cloud, or hyper spectral satellites. AI and Chat GPT are going to change how people find information and thus change how they make buying decisions. It's never been more important to thread your conversations throughout everything that people are going to be searching. Making sure that all aspects of your business and technology are represented for AI systems to find and deliver.

SG: So basically what you're talking about is AI SEO. So the question is, what becomes the feedstock for these AI systems? What I'm hearing you say is that a lot of it's going to be journalism and news reporting. So how will AI measure The New York Times versus someone's blog post on Medium versus Wikipedia? How is that going to shake out?

JO: AI is obviously extremely data hungry. It's going to consider all of the publicly available sources on the internet, that's just a given. If you go and ask an AI engineer how neural networks work, how the information is weighted, they're going to likely say that obviously an article in The New York Times far outweighs a tweet thread from some unknown person who could potentially be a bot. 

All of that information would be considered, but the weights and balances are going to be heavily favoring journalism. Because journalism is a credible source of truth and unbiased (for the most part). So the information that exists on people's websites – that's going to be really important. That go-to-market messaging is going to be how AI systems describe it. If someone asks, “What is Mixing Board?” it's probably going to pull from your website. If somebody asks deeper contextual questions or competitive questions it's probably going to pull a lot from news articles that have been written about the company.

For a long time in the comms world, we have struggled with the measurement of ROI. I've always felt that ROI in comms has been extremely limited, especially within the context of a certain time frame. We measure the outcome of a news announcement based on the week or the month that it's in. But I've had so many people, reporters, and producers contact me about a client because of a story they read that happened years ago. There is inherent value in the history of journalism and comms on the internet. That's never been more true than today because AI is going to be synthesizing the totality of data to present who you are, what you do, why you matter, and how you stack up against competitors. 

SG: For the last eight years or so there’s been this push to “go direct,” because people can find your blog post on Google just as easily, if not even more easily, than they could find a news article from a year ago. The differentiation between those two is relatively minimal. But if the power dynamics change again, and AI is going to use The New York Times or the Wall Street Journal as the gold standard for information, the relationship to the builders and miners of this content will change. If we're living in a world where AI is measuring on a much more acute level, what you invest your time in, and what feeds the AI, is going to be different. 

JO: There's good and bad there. If you’ve played around with Chat GPT, you know that creating content is going to be easier than ever before. There's also going to be a lot of value to having detailed articles on every product that you offer, every service that you offer, so that AI can pull from that and represent it in the right way. The downside of it is, it could get really spammy. There was a period where every company in Silicon Valley thought they should become their own media company. I always thought that was a terrible idea because it just sounded like spam. There's very few companies that could do that effectively and they would have to be the top tier, clear leaders in their industry to actually pull that off. If everyone tried to do it, it would just resemble spam. 

It's going to be a very interesting time for content, and not just in the written word. Going direct with really cool animated videos and pictures is going to be really interesting. The ability to make an animated explainer video just by typing what you want is going to be something that's really helpful to brands. But as you democratize all of it, it’ll get a little spammy.

SG: How should leaders, comms teams, and executives start thinking about this stuff and managing it now? What are the steps that people should be taking?

JO: It's not just the emergence of AI. Every time there's a major moment – like Silicon Valley Bank and a potential recession – it's always a good time to reset and go, okay, what are our goals going forward? That should start with business goals, but it should also include, what do we want our brand image and brand tone to be over the next year? In bull markets the brand tone might be more playful. In bear markets that playfulness may not play as well. It's a good time to think about all of those things. 

But realizing that AI is going to be the gatekeeper to how people find information about your company is really important. There are going to be a lot of companies that end up doing this well. And certainly there's going to be a lot of companies that try to take advantage of this and get people to be a little bit spammy to try and increase SEO across AI systems. 

SG: What do you stack it up against? You have the advent of search, you have the advent of blogging and social media, and the advent of mobile devices in everybody's hands. And AI can potentially touch all these different elements. But all those different moments led to different ways that we interacted with technology and interacted with the outside world. But also as communications people, it completely re-shifted our jobs. Is this at that level, or is this one of those things? Is this one of those things where you could almost go too deep and lose the forest for the trees? Or do you think this is an all in thing that everyone needs to be super focused on?

JO: It's all in. I've been comparing it to the smartphone in everyone's pocket. That was a moment in which humans and machines were forever paired. This moment, and what's going to happen over the next couple years, is going to be the moment at which humans and our AI systems are forever paired. All of the software that runs on our smartphones, and down the line potentially glasses or brain interfaces or whatever, they are all going to be underpinned by these AI systems. Generative AI is going to underpin every application that we use. This is that moment where humans and AI are going to essentially coexist for the indefinite future.

SG: Do you have people on your team that you've assigned to focus on this all the time? Or are you just having everyone think about this? What are the internal conversations that you're having?

JO: We have talked about this a lot. We definitely want to be on the cutting edge of using systems like Chat GPT, but with strict guidelines. I've used it many times to help with writer's block, but I’m very careful about cutting and pasting. There's going to be a wave of people that get caught positioning generated responses as their own thoughts. That is not something that I ever want to be a part of. 

I also believe that we need some new standards, in terms of disclosure and what parts of content are AI assisted. It could spin out of control. It's really important that we understand what is real and what is not. I've talked a lot with clients and journalists about the end of reality, when we can no longer trust what we see or hear.

Just yesterday there were a whole bunch of pictures circulated of Trump being arrested, and that didn't happen [yet]. Creating those types of pictures and videos is going to be as easy as typing in what you want to see or hear, and it'll be pumped out. It's going to be a really interesting era of figuring out how we verify information. 

We're very used to real-time news, and over the next few years, we're going to be conditioned to not believe versus believe. Not believe what we see until a reputable news organization actually verifies that this thing happened – a war did start, an earthquake did happen, a political leader was shot. All of those things we believe right now because it trends on social media. Three years from now, we will all say this is probably not true, until it is true.

SG: It's like the opposite of trust but verify. Verify, then trust. Do you think that the AI industry, which includes tiny startups and then Google and Microsoft and others, has done a good job of getting ahead of these trust issues? 

JO: No, not really. Right now they're mostly focused on shipping. Because AI has obviously become very much an arms race for consumer attention. You saw how fast Google followed with Bard after Chat GTP lit the world on fire. I'm not sure that they see it as their responsibility to introduce standards for how people or companies should verify information. We're in that period where this is going to be released and then we all need to take a beat and figure out how this is actually going to integrate into our lives in a responsible way.

SG: We'll deal with the lawsuits and the regulations as they come. Well, we've all seen this movie before, so we know how it turns out.

JO: I can't overstate this enough, but this is extremely powerful technology. And the genie's out of the bottle. And if the United States shoots itself in the foot with an unbelievable wave of lawsuits and things like that, while countries like China unleash it and grow it and become more innovative in this world, then that is a real problem.

SG: There's a thing that we've said for a while, that over time, every company is going to become a tech company. Over time, I assume, every company is going to be an AI company in some way, shape, or form. Whether you're buying toys or you're going to a football game, there's going to be AI involved. This is not just some sort of discrete tech issue. This is an everybody issue.

JO: Essentially all software will become artificial intelligence. That's just the way it's going to progress. 

SG: One thing that you and I have talked about is how comms people can get more skin in the game and be elevated into the level into the role of a true trusted advisor. And then get compensated through investing or gettin advisory shares. You've been doing this on your own, pretty successfully for a while, and have integrated into the Strange Brew fabric. What can people learn from how you've gone about doing this? Because, a lot of times,  the biggest barrier for people is learning how to ask.

JO: The first step is just understanding that you can ask, and frankly, there's been times where I've begged. But asking becomes a lot easier after you show value. If you start working with a company, you prove that you can be valuable to the company, then interest alignment becomes a lot more attractive.

And more education needs to happen, this could happen from the VC level, but there just needs to be more awareness. If I were a founder today, obviously I'm biased, but I would find it extremely beneficial to find someone who is really plugged into the network that I want to operate in, or the journalists that cover my industry. I would want to lock down very early access and counsel from this person for years to come. That's a huge strategic benefit, and a lot of founders don’t do that. A lot of VCs don’t counsel their founders to do that. And frankly, it's a mistake.

SG: What typically happens is that, “Hey, we have an announcement. Let's find a comms person. Let's try to work with them for two weeks, and then the next time we'll try to find another person.” It's not smart. Founders – stop doing that.

Last question. If you were just starting off in your career in comms today, where would you begin? 

JO: I am biased, but I think even in-house people, career in-house people, would admit that starting in an agency is probably best. It's just so good to get exposure to a variety of companies. The biggest benefit that I can offer to my clients today is the things that I've done in the past. Being able to connect the dots between news announcements, stories I've worked on with journalists, or counsel that I gave in 2009 during an economic crisis that still may apply today. Getting a wide variety of experiences is really important.

Going in-house can also lead you on a path to specialization, which can be incredible if you're super into gaming and you just want to work with gaming companies the rest of your life. However, if you start an agency, you may find that you find the world of AI more fascinating than gaming. You find enterprise cybersecurity way cooler than you ever thought you might. What's important is just being exposed to different sets to figure out what it is that you're into.

###

Curious about Mixing Board? Visit us here and feel free to reach out.