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.

AI overreliance versus AI skepticism: Balancing the risks

While AI still has much room to grow and stabilize, forgoing it completely is no longer a viable option.

AI overreliance versus AI skepticism: Balancing the risks

[Adobe Stock / Konstantin Yuganov]

Profile photo of Steven Kawasumi

Product Management Executive · Steven Kawasumi AI

As artificial intelligence grows from a novel concept into a tangible, fully-fledged corporate asset, its implementation has created many implications, questions, and debates related to its countless practical and technical nuances. One such conversation revolves around how much—if at all—a company should automate its existing infrastructure. In this sense, today’s businesses face a critical balancing act between two AI extremes: relying too much on an emergent technology and denying its potential benefits outright. 

To properly navigate this process, companies must assess the unique risks associated with both outcomes. This informs potential automated initiatives and ensures the company has a proactive yet vigilant AI mindset. 

RISKS OF AI OVERRELIANCE

AI is already revolutionizing numerous aspects of everyday business and culture by reducing needless tasks and helping human personnel reallocate focus to higher-level, more nuanced obligations. These prospects are exciting and increasingly fruitful, but they come with an important caveat: AI systems, while game-changing, are only as good as their foundational training data. At present, despite emergent capabilities shown by the likes of large language models, AI resources cannot fully replace complex human thought and focus, as they remain prone to shortcomings warranting persistent retraining and model retuning.

Along the way, companies should avoid the temptation to fully give AI the keys to their operations, so to speak. The technology can serve as a crucial complementary tool, but its data requires diligence and forward-thinking. This includes finding a clearer balance between AI and human intelligence to mitigate overreliance while leveraging AI’s benefits.

In aviation, pilot training often addresses the “startle factor,” which occurs when an autopilot disengages unexpectedly due to an anomaly, leaving pilots momentarily unable to respond appropriately—even though the corrective action might be simple and would have been routine if they had been hand-flying the aircraft. This phenomenon highlights how skill erosion and overdependence on automation can hinder human response during moments of stress or surprise. 

Translating this concept to the workplace, companies must ensure employees remain confident and capable when technology fails to avoid the operational paralysis when workers are caught off guard by systems they have become too reliant on. Should a workplace find its technological infrastructure compromised, this balanced approach will ensure human personnel can keep operations moving without knowledge gaps created by excessive AI dependence.

RISKS OF AI SKEPTICISM

At the same time, aversion to AI resources can lead businesses astray by disconnecting corporate culture from modern methods and creating needless skepticism toward AI’s fully realized and objectively dependable uses. AI now defines the business landscape and sets the pace for corporate competition; this is an inevitable fact demanding adaptation rather than suspicion. 

Many business leaders are taking initiative in this regard, but others remain fearful, citing AI proliferation as an insurmountable challenge that has reached critical mass. These tendencies can quickly create a hollow and potentially anachronistic experience for consumers, who have grown to expect nuanced, AI-infused service as an emergent new norm. For instance, leaders may cling tightly to rigidly curated troubleshooting and customer service flows in their automated systems—an attempt to needlessly maintain control—which often results in unscalable, rudimentary experiences stunting AI’s practical potential. 

However, so long as AI presents ever-changing solutions to ever-changing issues, it will be the best way to navigate the often nebulous environment it has helped create—and leaders can find comfort in ongoing AI education and training opportunities for themselves and their subordinate teams. 

While AI still has much room to grow and stabilize, forgoing it completely is no longer a viable option; companies doing so will undoubtedly fail to grow their products and services within a dynamic, competitive environment centered on disruption. As data and signals increase, companies with a strong, well-thought-out AI infrastructure will have an easier time utilizing vast data when urgency demands quicker, more efficient decision-making.