top of page
  • Writer's pictureBrian Baecker

Generative AI is Speeding toward the “Trough of Disillusionment”: And I Couldn’t be More Excited

I know, the onslaught of Generative AI posts is a lot, and trust me, the irony is not lost that this is yet another one, but it’s been interesting to see a rapid shift in tone related to GenAI content in the past couple of weeks and I think it’s an important callout to recognize where we are and how we think about what’s next.


So, what the hell am I even talking about?


There are many adaptations to the business S-Curve, but I really like Gartner’s Hype Cycle methodology outlined below:

graphic of hype cycle
Gartner’s Hype Cycle

For those of us who have seen countless technologies and trends come and go, this trend line shouldn’t be surprising. 


For all the ways that GenAI is revolutionary, it still seems to fit this familiar innovation pattern, albeit, on steroids.  It’s understandable that everyone may see themselves at different points of this curve. So to keep us together, I’ll take you through what I think we’ve seen thus far and where we’re headed.

Technology Trigger Phase

It’s hard to believe that just 18 months ago, GenAI posts were barely making it into mainstream media and business.  Then OpenAI took the world by storm with ChatGPT. 

In 5 days the ChatGPT platform to a million monthly active users In just 2 months, they had over a hundred million

In a few short months, countless AI start-ups emerged, new roles and titles began to form, and it felt like every company started jumping on the hype train to figure out this next frontier in tech.

For those of us who have seen countless technologies and trends come and go, this trend line shouldn’t be surprising. 

Peak of Inflated Expectations

Over the past year or so, AI content and offers are like drinking from a damn firehose.  It’s coming for your job! It’s integrated into your toothbrush!  It’s going to revolutionize (insert whatever here)!

collage of AI news headlines
AI Everything Everywhere All at Once

Then came the bloated evaluations, leading many to think that GenAI will invade every part of our lives in a matter of months, not years. 

chart showing added economic impact of AI

Naturally, every business under the sun has had to have an opinion about it, offer products that feature it, and many, unfortunately, overpromise and underdeliver in hopes of getting a sliver of that market potential. 


In our market analysis here at F’inn, we are continually taken aback by the lack of knowledge, chronic use of buzzwords without substance, and general lack of benchmarks and security in offers being rushed to market.


This whole period gives off heavy “Peak of Inflated Expectations” vibes, but there is another side to this dynamic, a “glass half full” of rich experimentation, everywhere.


This period is really when use cases start to take shape and hypotheses start to be proven vs. disproven providing viable pathways to utilization (though many fail or don’t live up to stated expectations).  

The Trough of Disillusionment

Over the past few weeks, I’ve seen an influx of media questioning the efficacy and ROI of GenAI.  Companies are starting to pull back on AI experiments and take a more conservative approach to integration.  After running at full speed for the better part of a year, it seems many are trying to catch their breath.

pessimistic post about AI
State of Generative AI in the Enterprise: Getting Real about GenAI

What does it all mean?  Are expectations too high? It’s quite possible that we’re riding this roller coaster downhill to the “trough of disillusionment” and I couldn’t be more excited. 


We’re seeing a classic case of demanding the shiny new thing and then bailing on it when it doesn’t meet expectations out of the gate.  The reality is innovative technologies rarely generate immediate ROI easily.  There are challenges and friction to end users that require collaborative thinking in how to scale / get it right.  It also requires the right people willing to challenge stakeholders on the promise of what can be achieved with continued investment and experimentation even when the current output is not totally there yet. 


This is the time in the innovation hype cycle when experts start to separate themselves clearly from the pack.  It’s a time that forces companies who truly believe in the power of what GenAI can achieve to hunker down and stay focused on its development, with little to no immediate ROI, knowing that the true potential will come with better education, realistic expectations, and more transparency around benefits and limitations.  It separates truly valuable applications from hype and buzzwords. 

This is why F’inn is actively testing an array of open- and closed-source models. We are establishing benchmarks against to:

  • determine the variables that matter

  • rate their efficacy

  • find the strategic implementations

We are constantly experimenting with use cases to see what is and isn’t possible. All this as we, collectively, start to progress up the slope of enlightenment.


Now is the time for companies to partner with agencies that thrive in uncertainty. You want to find teams that demonstrate thought leadership through direct experimentation and are willing to be transparent with you on the process and learnings.


While it seems self-serving for me to say that companies shouldn’t press pause, coming from an agency actively creating GenAI solutions. I’m really saying more that companies should be smarter about partnerships. Ask yourself:

  • What is the methodology that influences outcomes?

  • Is security & privacy being taken seriously?

  • How transparent is the process?

  • What role does consumer intervention play in gut checking the outputs?


For those who act now, the reward will be an accelerated path to the Plateau of Productivity.


There is so much potential to be tapped with GenAI. Don’t let this short blip of noise dissuade you. We’re just getting started with the real, deep impacts to many industries and you won’t want to miss out.

button saying let's stay connected


bottom of page