Inc.
AI hype is fading because most projects never scaled and money went to tools without purpose. The real issue is not AI but weak goals. Lasting results come when leaders treat AI like any tool with clear aims solid data strong rules and a plan that shifts as they learn.
Author: Louise K. Allen
The business world has been in the throes of AI’s hype cycle for just over three years now. Right on cue, the tides are beginning to turn. Headlines touting the technology’s potential to transform every aspect of daily life, both in the office and at home, have been replaced by stories about its failure to drive meaningful change or produce the promised returns.
Most recently, an MIT report on the state of AI implementation found that just 5 percent of AI pilots to date have gone on to full implementations. This reality has left the majority of the estimated $30 billion to $40 billion spent seemingly wasted on tools that never scaled. While it would be easy to attribute this startling statistic to faults in the technology itself, the reality may be both more complicated and harder for leaders to swallow.
Hammers and nails
When you have a hammer, everything looks like a nail. Cliche? Maybe. But it’s a succinct summary of the approach leaders have taken on AI thus far. AI companies handed business leaders a tool they claimed could do it all, and leaders took it at face value.
The problem is that—while versatile and highly impressive—today’s AIs are fairly fixed in their underlying functionality. Generative AIs and large language models do one thing: deliver predictive outputs based on reference data. This core function can be applied in near limitless ways, but that doesn’t change the process happening behind the scenes.
This critical fact was lost somewhere in the hype. The prevailing belief became that all AI investments would guarantee returns, and any implementation was better than being left behind. So, businesses sprinted to “do AI” without first asking what AI would do for them, specifically. To correct course and flip MIT’s genAI divide, leaders need to start seeing AI for what it is: a tool, like any other, that must be wielded with purpose.
Getting back on track
The good news is that leaders have extensive experience doing exactly that. It’s a story that’s played out countless times before. Virtually every transformative technology has ridden this wave before finding its natural place within operations. For many tools, the transition from shiny new toy to actual accelerator took far longer than a few years. It took decades and a historic crash for the internet to mature into the indispensable tool it is today.
Though high-profile hype and overstatements of today’s AI capabilities have played a role in getting us here, many leaders have forgotten that pursuing a technology, for its own sake, isn’t an effective strategy—no matter how capable that technology may be. To utilize AI to its full potential, businesses will have to start treating it like any other tool.
How to utilize AI to its full potential
- Knowing the goal. In the rush to win the race to AI, technology itself became the goal. Applications guided by the desire to use a specific technology are destined to fall short of expectations, because they are solutions without purpose. The first step in turning AI into measurable returns is identifying objectives and finding solutions that move the needle toward them.
- Identifying measures. These are the “key results” in OKRs: the indicators of progress that leaders reference to assess efficacy. They should be aligned with the objective of each initiative and aligned across all departments, so each contributor is working toward the same overall outcome, regardless of their role within the organization.
- Laying a foundation. Data drives AI solutions, not the other way around. Data infrastructure strategies must precede AI implementations if leaders hope to derive any value from the latter.
- Getting serious about governance. The buck has to stop somewhere. Let governance be that stop sign. Establishing clear, enterprise-wide guardrails around AI investments may feel like a hindrance to progress. However, it’s the key to making sure all efforts are moving in the right direction.
- Expecting to pivot. Inevitably, the data will reveal that a change in course is necessary to move forward. That’s the whole point. Rather than abandoning the initiative altogether, adjust. If needed, undo it. Roll things back and start again with new measures and methods in mind.
An outcome-based AI strategy
Just as the dot-com crash didn’t signal that the internet lacked utility, the underperformance of AI investments to date isn’t an indictment of its potential. Rather, it’s a call to change the approach to AI-driven transformation. If leaders are to derive value from AI solutions, they need to start treating them like any other investment. That means developing an outcome-focused strategy centered on core objectives and underpinned by clear guardrails and a strong strategic vision.
Credits: TCA, LLC.