Skip to content
Fast Company Executive Board

The Fast Company Executive Board is a private, fee-based network of influential leaders, experts, executives, and entrepreneurs who share their insights with our audience.

Is that company just licking the AI cookie?

Staking a claim in the enterprise market is one thing. Delivering something people can deploy and adopt is another matter entirely. Get ready as AI enters the lick-the-cookie era.

AI cookie

[faithie / Adobe Stock]

Profile photo of Rahul Roy-Chowdhury

AI is being overhyped and oversold. Companies are cramming their products full of new features and integrations, but how many people are finding these new tools useful—or even usable?

Staking a claim in the enterprise market is one thing. Delivering something people can deploy and adopt is another matter entirely. Get ready as AI enters the lick-the-cookie era.

If you’ve ever seen kids eating cookies, this mindset may look familiar: When you want to save the last one for later, you lick it so nobody else takes it. For AI companies, that means rushing a product to market to put dibs on a space before competitors can get there—selling experimentation, not implementation. Great enterprise AI has to be proven, usable, and measurable. Too many of today’s offerings fall far short of that mark. 

Tech industry observers are already recognizing the dynamic I’m describing. Amid premature hype and rampant AI-washing, Gartner has declared generative AI to be headed for the “Trough of Disillusionment,” while investors are growing increasingly skeptical of the technology’s prospects. Expectations of bubble-popping have reached a breathless pitch. 

But it’s important not to get carried away by the AI backlash. While there are plenty of soggy cookies on the market, there are many other platforms that are ready to eat—and truly satisfying. For customers, the key is to know how to tell them apart. 

WHEN AI COMPANIES EXPERIMENT ON CUSTOMERS

Shipping faster instead of better is a time-honored Silicon Valley tradition. The name of the game isn’t to deliver perfection—it’s to deliver first. Your early adopters will tell you how you did, and you can iterate from there while fending off later arrivals.

But this approach can come at a cost to credibility. Consider the media merriment that’s greeted generative AI stumbles like marriage-undermining chatbots and Onion-ready Google AI overviews. Such high-profile goofs are great for laughs, but they risk turning a truly transformative technology into a punch line.

Within the enterprise, users are handed hastily developed tools that are hard to adopt and use effectively and not specialized or deeply trained enough to drive real value. Product-centric experiences force people to change the way they do things, disrupting work instead of enhancing it. Adoption stalls and regrets mount. There’s often no way to measure impact, but even without metrics, it’s all too clear that another AI promise has fallen short. That’s bad for the market as a whole. 

FROM HALF-BAKED HYPE TO REAL IMPACT 

To regain customer trust, AI companies have to go beyond buzzwords and be able to prove day-to-day value. That starts with making AI highly usable by integrating it naturally into the way people already work, without forcing them to “learn AI”—or even be aware that they’re using it.

Think of the way Zoom has seamlessly woven AI capabilities throughout its familiar experience or the way my own company, Grammarly, helps people express themselves better through the communication channels and platforms they’re already using.  

For customers, choosing an AI solution they won’t regret depends on practical criteria and realistic expectations. As you develop an AI strategy for your organization, stay focused on:

  • Employee needs and priorities. Identify your AI power users and work with them to find the use cases that make the most sense for your business. Make sure they’re broadly adoptable, scalable, and drive real value for key operational or strategic priorities.
  • Measurable impact. AI can do many things—some of them well—but are they really worth doing? Clear ROI is far more important than impressive specs and flashy features with uncertain benefits. Make sure you have relevant metrics to track so you know your investment is paying off. 
  • Evolutionary progress. Start small with a limited set of high-value, easily adopted use cases with immediate relevance. As you gain experience and knowledge, you can build on that success and make informed decisions about longer-term AI investments down the road. 

No kid wants a cookie that’s been licked, and no business should get stuck with a disappointing investment. Before reaching for that AI product, make sure that you trust its intentions and practices, and that you’re confident in the experience it can deliver for your users and your organization.