Catch-22: Thoughts on “AI” in Marketing and Inevitability

I’m firmly on the record as not being a fan of “artificial intelligence” (and its various offshoots and derivatives), so it’s been disappointing to see people I respect sipping the Kool-Aid as it becomes a nexus for influence-chasers and scammers alike. Some have fully bought into the self-perpetuating “inevitability” argument as half-baked “AI” features are being shoved into every imaginable product and service, or existing features are opportunistically rebranded, while others are selfishly pushing “inevitability” because they have a vested interest in driving adoption as fast as possible before the bubble inevitably bursts.

What’s also inevitable is the short-term negative impact of force-feeding unproven technology into your own workflows because of peer pressure or investor demands. “AI” is nearing its “Peak of Inflated Expectations” and its main backers are desperately trying to ensure it’s as entrenched as possible before the “Trough of Disillusionment” kicks in so they can extract their profits and let everyone else deal with the fallout. It’s the same old Hype Cycle; we’ve seen this movie many times before, and yet here we are again.

Is it even possible to work in marketing in 2024 and avoid the siren call of AI tools, though? I think so, although it definitely depends on your specific role, industry, and past experience.

From Bad to Worse

Practically every notable marketing-related service is scrambling to add an AI component to their offerings, and then building content marketing efforts around them because they’re invested in the inevitability argument. LinkedIn is absolutely littered with free and paid marketing for it, as well as inadvertent examples of how bad it can be from wannabe influencers chasing the algorithm. (HubSpot has a relatively level-headed overview, although if the rumors of a Google acquisition are true, I suspect they’ll go all in shortly after that deal closes.)

The vast majority of use cases for “AI” in marketing are related to potentially tedious tasks like writing repetitive marketing copy, blog posts, and automated email campaigns; and creating social media content, digital ads, or “free” images. The iterative prompts required to generate anything remotely useful (that will still need to be edited if you actually care about good content and not being a plagiarist) can often take as much time as doing it from scratch — assuming you have the necessary experience and skill. This approach seems to be particularly appealing to non-marketers playing the thought leadership game, believing they can trade in their ghost writers for ChatGPT and no one will notice — and ironically, most of them are correct.

One of the biggest reasons “AI” is gaining traction in marketing circles so quickly is most marketing is actually pretty bad already. Sorry, not sorry. Of course, this is partly because marketing departments are often under-resourced to begin with and are typically one of the first areas to be cut when margins need padding, or an inept CEO needs a scapegoat.

I’ve spent significant chunks of my career as a marketing department of one, so you’d think tools like these would be appealing to me, but they’re not at all. In fact, I’m generally confident that my refusal to use them gives me an advantage over competitors tempted to embrace them because I’m almost always playing the long game.

Artificial vs. Augmented vs. Authentic

One of the most fascinating findings of the legitimate research on the impact of “AI” in different areas is that “productivity” is a double-edged sword.

Last year, I attended a provocatively titled discussion, “Do bodies still matter in the age of AI?” that included Vivienne Ming, a theoretical neuroscientist and self-proclaimed “Professional Mad Scientist.” I’d never heard of her (nor her co-presenter, Meta’s Richard Newcombe), but despite my skepticism about “AI,” I’m always open to perspectives that challenge my thinking and might change my mind. It ended up being an incredibly insightful discussion, mainly because Ming and Newcombe are experienced practitioners rather than business executives selling snake oil, and when it was over, Ming had joined Baldur Bjarnason (The Intelligence Illusion and his ongoing commentary) as one of my most valued voices on the “AI” front.

45 minutes into the video below, she breaks down some of the research into the productivity gains across various industries and activities that inform my continued skepticism about using these tools in my own work.

“First, in every single one of those studies, the improvement is almost entirely reserved for junior or lower-skilled workers. Experienced workers saw either no benefit whatsoever, or rather marginal benefit compared to the lower-skilled. Many of the authors said, ‘Look, we’re decreasing the skills gap for employees. Isn’t that great?'”

She discusses a few specific examples, including decades of research on “AI” in education, and perfectly sums up my primary concern about companies relying on “AI” tools to increase productivity:

“All the people that are getting the productivity boost out of generative AI are lower-skilled than their comparison group. To me that suggests that these people will show up on the job, they’ll have Co-Pilot, or they’ll have a system that reads contracts for them, or assist their diagnoses, and they’ll never learn how to do their job.

The person you got on Day 1 is the person you have on Day 1,000.“

ReD Dialogue NYC – Do bodies still matter in the age of AI? from ReD Associates on Vimeo.

Knowing When to Say, “No Thanks”

My entire career has predominantly been in marketing and was built on three key pillars: 1) understand how tools work; 2) understand how tools impact MY work; and 3) understand what colleagues and senior executives know and don’t know about those tools and who’s promoting their use.

“These are unreliable tools made by organisations we absolutely should not trust, coming out of a field (AI) with a long history of over-promising, snake oil, and even outright fraud. The burden of proof is on them, not us.”

Baldur Bjarnason

I’ve always been an early evaluator, cautious adopter, and 20+ years of parsing incredulous hype about the new shiny du jour has made me skeptical of everything — and a frequent annoyance to bosses past and present. “Doubt, but verify.” As a result, I’ve never missed out on or felt terribly late to anything that actually had value. I’ve frequently avoided wasting time and money on silly pivots that unnecessarily disrupted proven, foundational processes. Most importantly, when something new actually has legitimate value, I’m usually ready to take full advantage of it, and know when it makes sense for ME to do so.

It’s really easy to get swept up in the new shiny hype if you a) are relatively inexperienced, b) don’t have a firm grasp on the fundamentals, and/or c) are an active participant in and beneficiary of the attention economy.

The first two are arguably more difficult to overcome than they used to be because of the third, so I empathize with younger marketers who feel caught between a rock and a hard place. If nothing else, the red flags might be a little more obvious to them when they realize how little their work is actually valued. Even if you’re forced to use any of these tools because you can’t say no, use them with discretion, be mindful of the shortcuts they take, and actively focus on learning about what they’re allegedly making easier for you to skip over.

There’s measurable value in doing “tedious” processes like researching and writing and editing. It’s how we learn and improve, ensuring we’re not the same person on Day 1,000 that we were on Day 1.