As previously published on 1/16/25 in MarTech
An interesting piece of research was released last week but may have been lost in the busy holiday season.
Previsible, an SEO consultancy, announced that traditional Google search has “basically plateaued and has begun to have its search dominance degraded.” Why? People are using AI assisted search because it has gotten more capable and accessible.
ChatGPT, Claude, Co-pilot, and even Google, offer an AI search version available to most users. Compared to the traditional search, which relies mostly on keyword matching, AI search uses advanced algorithms to understand the context and intent behind the query. As a result, at least in theory, it should provide more relevant and personalized results.
The new capabilities and changing user behaviors are creating a potential warning about the risk of relying on AI. Because of AI’s ability to draw upon vast amounts of information, users often default to trusting that the query output is most likely to be the right answer, solution, recommendation, etc.
As opposed to the traditional search which returns the links to what are the most likely options to answering the query, the user has to make the effort of analyzing the results, reading and filtering information, and drawing conclusions.
And here lies the potential problem.
Nvidia’s CEO, Jensen Huang, stated on the company’s most recent earnings call that we are “in the beginning of a new generation of foundation models that are able to do reasoning and long term thinking.”
Cognitive scientist Gary Marcus says the AI we are currently building is basically like “System 1 thinking,” a reference to Noble Prize winning psychologist, Daniel Kahneman’s, “Thinking Fast and Slow” book.
In his book, Kahneman explains the dichotomy of human thought. System 1 being intuitive and fast with no voluntary control. This being one of the reasons he concluded that humans are bad at making decisions. System 2 thinking allocates attention to the effort that demands focus, oftentimes because of its complexity and/or need for computations. Think of the two systems as instinct or “gut feel” and critical thinking.
If, as the AI experts state, we are building System 1 AI models then users are at risk of making the same mistakes using AI as they might make in day-to-day decision making. And, as an observer of younger generations of marketers using AI, they may be particularly vulnerable.
My son, home for the holidays from grad school, mentioned that classmates are not only using ChatGPT to summarize course work, but also to write their presentations. And…they’re not questioning it, they follow the recommendation completely because it “saves time.”
B2B marketers are using AI tools profusely for research, writing, and recommending actions because they are “quick and accurate.” They have grown up in an environment that has emphasized scale and speed, and they lack the experience or interest to question the accuracy of what is being outputted by AI tools.
Where is this headed? Combine all of these factors and it could point to a massive wave of “group thinking” marketers that either lose the ability to think creatively and/or strategically, or eliminate it completely because they are wired to trust AI.
Generative AI has already come for the creative department as witnessed by Omnicom’s recent acquisition of IPG. If marketing executives don’t act now to create a plan to manage AI, “Hal” could become your CMO in a few years.
How should marketing executives respond to this threat? Daniel Kahneman might suggest focusing on skill development that emphasizes System 2 thinking. Teach your team how to do long term, critical and strategic thinking.
Combine the strength of using AI System 1 thinking to enable your staff with training on higher level System 2 type efforts like competitive intelligence (which I rarely see any more), market intelligence and strategy.
There is good reason to get back to these core strategic marketing building blocks. Marketing performance in 2024 was significantly down across channels and activities. It’s time to dig in on strategy. There are significant challenges to address. Going faster and creating more noise in the market is not a strategy that will win.
In 2017, I wrote an article on how Amazon had become the default search engine for buyers who knew what they wanted based on our research on buying behavior. In that post, I predicted that because of that trend, Amazon would soon eat away at Google’s advertising monopoly. At that time, Amazon only had 1% of the global advertising market. By 2020, it had grown to over 10%. This year it will be 14%, and by 2026, it’s estimated to become over 17%.
I see a similar trend with AI eating away at the marketing department, not because of the tools themselves, but by how behaviors are changing because of them (similar to what I observed with consumers and Amazon). To be clear, it’s not necessarily the technology that is the threat, but rather the behavior change caused by it.
If marketers want to remain valuable inside their organizations they will have to learn how to use AI tools to enable better decision making and not default to them as the decision maker. Or as my son’s professor said; “use them to become a better student, not to be the student” and remember, they’re only System 1 thinkers 😉.