The pendulum that makes up the AI hype cycle is in full swing. It’s just that no one has told PR and marketing yet. Please indulge me while I change that.
Welcome to the global correction, in which AI is repositioned as a normal technology.
Academics and tech experts have already articulated the AI hype backlash. PR and marketing have just been slow to catch on to the responsible AI messaging pitch.
I get it. It’s our job to play up and amplify capabilities, using key tools to get the message out, including AI chat platforms like ChatGPT and Claude.
We’re now at a point where we have to make sure that we right-size messaging so that it aligns with what our vendor clients provide. If that means dialing down some of the gaudy language and narrowing our apertures for enterprise AI messaging, so be it. Think AI vendor messaging discipline and rigor.
To understand why, let’s look at how we got here.
Over the past three years, wave after wave of hype has crashed on the beach of AI, first with generative AI and, more recently, agents. In 2023, we collectively proclaimed GenAI as magic for its ability to generate reams of reasonably decent content in mere seconds.
The postmortem? We came for bottomless productivity and left with something that resembled Google auto-suggest search on steroids.
Next, AI agents were going to change everything from booking our vacations and business trips to executing a lot of the mundane knowledge work we loathe. Populating spreadsheets and shipping them, or generating reasonably good research reports and shipping them to stakeholders.
They’re not much harder to use than GenAI, as thousands of people who don’t code for a living and have built their own CustomGPTs and other agentic tools can attest. But it turns out that while they can do some amazing things, they also crash early and often, as The New York Times detailed. And real governance seems illusory, an oasis mirage in an intelligent software desert.
The latest craze is the hunt for the holy grail that is AGI, which some tech bros claim is already here in the form of recent models from Anthropic and OpenAI, others have described as an ill-defined nothingburger and third contingent claims will remain elusive until real-world models are mastered and diffused as products. Try as we might, we can’t help ourselves from moving the goalposts.
As to the promotion of non-AGI solutions, I like how Forrester Research analyst Mark Moccia couched the current AI froth: “The AI market right now looks less like a strategy discussion and more like a land grab with better branding. Every company is sprinting, every board is asking the same question and every vendor is waving a map that conveniently leads right back to their product. This is not new territory. We have seen this movie before in cloud, in digital, in mobile, in the early days of the internet.”
This seems like a smart play to coach vendor clients to demystify AI. I also encourage tech companies, particularly AI startups navigating the terrain for the first time, to not drown in your own Kool-Aid. It won’t serve you well.
This means PR professionals should make sure that if they are promoting a product upgrade that it is positioned as such, not as a game-changing release. And if they are promising something revolutionary, that it really is. If the client can’t get passionate about describing it, then maybe they pull back and wait until they have something important to say.
Why risk skewering AI marketing ethics and diluting your brand for the sake of a SEO/AEO traffic bump? We don’t want to be shouting from the rooftops for something that is the next metaverse. At the same time, we can’t all be versed in every nuance of technical detail.
So try this: Talk to analysts, academics or even journalists and tech content experts to get your finger on the pulse to understand whether what you are selling is really magic or a Magic 8 Ball.
This will help you duck AI messaging red flags.
It’s okay to have frank conversations about refining messaging so that vendors can be sure to deliver what they promise, and that they promise only what they can deliver. In other words, ensure the messaging is right-sized for the actual capabilities, not just aspirational capabilities.
As my colleague John Ward elegantly put it: “The strongest positioning in AI is achieved when the messaging deliberately separates the category’s aspiration from the product’s evidence.”
