First, they came for the rote tasks: assembly lines and appointment scheduling.
And I said nothing, for I don’t do rote tasks.
Then they came for our blog posts. And still, I said nothing.

And then they came for me.

I was shocked recently when, in a mentoring session, a VP of Marketing I work with flat-out asked if he should be worried for his job because of AI. I really didn’t think so. 

Then came ChatGPT. 

Suddenly, folks in the marketing and creative industries felt threatened; their sentiments reflected in Reddit threads like this one:

Now, AI-powered creative output still has a long way to go—AI-written content can be dry and lackluster; AI-drawn people tend to have too many fingers—but the possibility of machines taking over our jobs is definitely there. Still, humans need not fear being driven out of their jobs. According to this Future of Jobs report, 75 million current jobs will be displaced by the shift in the division of labor between humans, machines and algorithms, but 133 million new jobs will be created as well. 

With that, I decided to do a deep dive into the impact I see AI having on marketing to see which jobs do have a chance of being replaced by machines, and which jobs don’t. Here’s our fearless forecast for the future of AI in marketing:

10 Predictions for how AI will Reshape Marketing

1. AI will be the inflection point that ends the iPhone’s dominance.

Imagine asking your phone to update a Salesforce contact. Or to check a Looker dashboard.

Adept AI is working on a tool like that for desktop: its ACT-1 Google Chrome extension. I wouldn't be surprised if an AI program like this also gets launched for Android.

Now, will this happen for the iPhone?

Apple’s proven to be excellent at hardware but abysmal at software (see Apple Maps, iTunes, and of course, Siri). Do you think Apple will allow a voice assistant besides Siri to sit on top of iOS? iOS is so locked-down and Siri herself is so shitty, that this “voice-updating-Salesforce” will never happen on an iPhone—but it can definitely happen on an Android device (or a forked version of Android created by someone else).

And as AI becomes a bigger part of our lives, I predict it will eventually give consumers a big enough reason to switch to another phone OS.

2. We’ll witness the death of SEO.

Think about the time you googled that cocktail recipe, only to land on a 1000-word article covering who invented it and its history through the ages. That’s because most of the content out there is written for Google, not for getting us to the answer.

Google faces a classic innovator’s dilemma: 57% of its revenue comes from selling ads against pages of search results. But a powerful enough disruptor such as Bing, Amazon, or TikTok could easily leapfrog the search results page to give us the best answer. The result would be that rankings will stop mattering.

If this happens, I predict Google will be slow to adapt: there’s too much revenue at stake. But over time, this means the value of pumping out SEO-optimized text-based content will decline to zero.

3. The balance of power will shift from channel management to marketing ops.

For most of my career, the highest-paid marketers on the team were the performance marketers who managed paid search and paid social. But the action isn’t really within the channels themselves —t’s in implementing the tech stack to get data flowing properly between ad platforms and tools.

According to Reddit user dan_obrien, “Digital marketing nowadays is probably closer to financial trading than "traditional marketing". Algorithms already play a significant role and we'll continue to see innovation there, but I doubt ‘AI’ comes in and owns the process end-to-end.”

One of my biggest performance marketing wins of last year was an experiment where we started feeding lower-funnel conversion events (i.e. what happens in Salesforce) back up into the ad platforms so we could train Facebook’s AI model to hunt for SQLs instead of Leads. Today there are AI tools to pick the right ad text/image/CTA combination, and AI tools to predict which customer is most likely to churn. But configuring and testing them—I predict that’s where the action will be. And I predict that’s where we humans will continue to be indispensable.

4. Personalization will be more sophisticated—but will also require more finesse.

More recently I’ve seen tools being used for cold email outreach that, similar to a deep fake video of Hillary Clinton, create a personalized “video from the CEO” for each prospect.

But I predict that as the complexity of the tools landscape increases, so does the need for humans to stitch all the plumbing between tools and systems together to package up the insights and then iterate on them. And sure, you can program “Deepfake CEO” to say anything, but knowing what to say that is remarkable but not creepy still requires intuition, judgment, and taste.

5. Content will be ruled by point of view and personality.

I’ve seen content farms that crank out 2000-word “pillar page” articles optimized to rank on Google search. They do this by repackaging the same content that’s already out there—and the resulting content is predictably bland.

But now, there’s no barrier to entry: ChatGPT can do the same thing, and it’s completely upended the SEO content landscape. I predict that the result will be that people either want “the answer,” or they’ll want “an argument:” a unique point of view. And as information itself becomes commoditized, I predict we’ll see the rise of content that’s more polarized and more emotive—this will be the new way to distinguish between AI-generated copy and copy that’s written by real folks.

6. Data accuracy will soon cease being an issue.

Today, AI chatbots like ChatGPT are prediction models: their objective isn’t necessarily to answer questions accurately but to predict a response to a query. As a result, they often share inaccurate information. Just a few days ago, Google’s brand new AI chatbot Bard drew flak for regurgitating inaccurate information in one of its demo ads. 

However, these issues around accuracy are surmountable. I predict that, as Bing rolls out a new search bar powered by ChatGPT and Google refines Bard, we’ll see these accuracy issues disappearing over time. 

7. Chatbots will continue to NOT be a thing.

Real talk: inexperienced marketers tend to overestimate the value of chatbots. A 2016 study from Oracle said 80% of businesses already use or plan to use chatbots by 2020—and I get the appeal, it’s a shiny, promising new toy.

However, as a performance marketer, I can tell you that if I just spent $200 to acquire a lead, the last thing I want to do is to funnel that lead into a chatbot. I don’t have very high confidence that most businesses would be able to train even future AI-powered chatbots in such a way as to make them good at persuading a customer to buy. I’d much rather get a high-value lead directly in front of a real person.

There is still a place for chatbots, but it’s not in marketing to new customers. I predict chatbots will be most successful in customer support. Chatbots may not make sense for a website’s homepage, but they’re great for AT&T’s support page: they’ve already locked me into a 4-line family plan, and they know I’m not going anywhere. I’d love for this chatbot to be more useful with the help of AI.

8. Knowledge is commoditized, so arguments are the future.

With Google and now ChatGPT at our fingertips, we can get any detail of any information we need—but actually using this data to prove a point is a different ballgame. We believe that no matter how smart AI gets, it won’t be able to replace the unique way the human brain approaches reason and processes information.

9. Decision-making will be augmented by AI.

We expect AI to be good for games with fixed guidelines, such as chess. But recently, AI has also shown promise in situations with no rules, such as securing jobs and negotiating salaries. And though AI’s impact is most obvious in these big, life-changing scenarios, it actually influences almost every aspect of our daily online behavior.

Lots of the apps we use day-to-day, such as Yelp and Uber, are fine-tuned through AI to give the best recommendations. I predict that, in the future, every aspect of our day-to-day lives will be powered by Yelp-like reviews, from the customers you work with to the places you plan to visit, to the clothes you’re thinking of buying. That’s the algorithm hard at work.

And though hyper-personalized recommendations (that even come with rationale) will push businesses to keep improving their customer service, it might also deal a hard blow to small and newly opened businesses, the latter of which would have a difficult time catching up to their more established peers.

Do y’all have more to add? Drop them below!