The more a technology or concept permeates and becomes normalized in our daily lives, the more we expect of it. About two decades ago, a dial-up internet connection of less than 56 kbps seemed like a miracle. Today, with internet speeds as high as 2000 Mbps becoming normal, the 56 Kbps connection would be seen as something of a failure, at least in the developed world. This expectation change also applies to AI. After seeing many practical applications of AI contributing to human comfort and progress, the general population and the AI research community now expect each new breakthrough in the field to be more earth-shattering than the last. Likewise, what qualifies as an AI failure has also seen massive change in recent years, especially from the perspective of the problem owner.
What counts as AI failure today
The mere fact that an AI model performs a specific function with the expected levels of efficiency is no longer the only requirement for its applications to be considered successful. These systems must also be able to provide meaningful real-world gains in the form of time savings or earned revenue. For example, an intelligent parking system capable of predicting parking space availability with 99.7% accuracy – while undoubtedly effective – cannot be considered a success if its adoption in the real world does not lead to tangible gains. Even with such a system installed, parking lot managers or smart city administrators may not be able to make optimal use of their parking spaces for a number of reasons. These can vary from simple causes, such as parking operators not being able to use the software interface optimally, to complicated causes, such as customers and drivers finding it difficult or reluctant to adapt to the new system. . For these and many other reasons, only a fraction of AI projects succeed. Estimates of the total percentage of AI projects that fail to deliver real value range from 85% to 90%.
In most of these cases, the lack of tangible results obtained by AI systems has much less to do with the technological aspect than with the human aspect of these systems. The success and failure of these projects depends on how people interact with technologies to achieve the intended goals.
Why Most AI Initiatives Fail
As researchers continue to work and enrich the body of AI research, the effectiveness of AI and AI-based systems is constantly increasing. However powerful it is, any AI-driven tool is just a tool. The success and failure of AI initiatives, more often than not, is determined by how users – primary and secondary – perceive, receive and operate these AI systems.
Lack of buy-in from management
Business leaders — such as owners, directors, and senior executives — often end up being just secondary users of AI, or any other technology application for that matter. However, they are among the biggest beneficiaries as well as the biggest enablers of these initiatives. After all, it is often their will and their means that count in driving AI initiatives. Thus, the most common reasons why AI initiatives fail to deliver real value often involve a lack of buy-in from business leaders. Membership does not necessarily simply mean a willingness to distribute funds for AI initiatives. Either way, a growing number of companies are investing in AI initiatives, which means AI failure isn’t necessarily the result of a lack of investment.
Today, buy-in is represented by complete belief in the ability of a technology or investment to have an impact. This belief translates into a commitment to making these technological endeavors successful in ways that involve more than the technology itself. For example, a company that is truly committed to the success of its AI initiatives will also invest in non-core aspects of the initiatives, such as security and privacy, among others. Ultimately, it’s this commitment that ensures they take whatever steps are necessary to ensure AI’s success.
Inadequate user training
More often than not, AI-based applications do not fully automate manual processes. They automate only the most analysis-intensive tasks. This means that human operators are needed to operate and augment the data processing capabilities of AI. This makes the role of human users extremely important for these AI applications.
Even the best AI-based business intelligence tools will prove useless if the executives using them aren’t trained to navigate dashboards or understand data. This issue becomes even more pronounced when AI tools are involved at the operations level, such as computer vision-based handheld vehicle inspection tools or a mobile parking app that users can use to find and book. parking spaces. When users are not trained enough to navigate and use technology interfaces, applications may not deliver the expected results. Although a well-designed user experience (UX) can be very helpful in these circumstances, it is equally crucial that users are trained in these applications.
Prior to hands-on training on using new AI applications, users should receive awareness training on how the new technology will add value to their work. More importantly, they must be convinced that the goal of technology is not to replace them but to augment their efforts. Indeed, fear of obsolescence is one of the main underlying reasons for low user adoption.
Fear of human obsolescence
Whether consciously or unconsciously, many workers, most of whom are potential AI users, fear becoming obsolete as AI becomes more mainstream. This perception of threat often manifests as a reluctance to embrace the technology. Lack of enthusiasm then leads to lack of involvement in training, which ultimately hinders the results of AI initiatives.
AI initiatives will only be successful and deliver meaningful ROI when all users, from senior executives to blue collar workers, are trained not in the technology, but also in their role in .
How User Education Prevents AI Failure
Most AI applications are tailor-made solutions to problems specific to the companies and customers who use them. This means that there is no fixed manual on coexistence and the use of AI tools. Therefore, it is unreasonable to expect users of AI solutions to educate themselves about their organization’s AI initiatives. Companies, along with AI implementation partners, should come together to create case-specific user training strategies for the entire AI solution lifecycle. By creating and executing these user education strategies, companies can ensure their employees are facilitating AI initiatives in several ways.
Building and managing expectations
Before an AI project even starts, it is imperative to ensure that the upper management of the organization is in agreement with the project. And that’s exactly what high-level user training does. When business leaders and key investors are aware of the outcomes to be expected from proposed AI initiatives, they are more comfortable investing in the same. However, it is equally important to set expectations in terms of the input and support that will be required from management to make an AI initiative a success. Raising awareness of potential outcomes and expected support will ensure that AI projects have the structural support needed to be sustainable and successful. Sensitizing high-level decision-makers to the challenges will also reduce the chances that they will withdraw their support when projects run into obstacles.
Maximize user adoption
In addition to making secondary users aware of the potential benefits of AI initiatives, it is crucial to ensure that primary users not only accept but also are enthusiastic about the adoption of AI. Ultimately, if end users don’t use the technology the way it’s supposed to, the technology can never live up to expectations. Thus, part of the expectation of top-level leaders should be to convince lower-level managers and employees of the value of proposed AI initiatives. Leaders can do this by first establishing, through open and clear communication, that AI applications will not replace the human workforce, but augment it. Another way for management to accelerate user adoption is to provide adequate retraining opportunities for employees so they can be better operators of AI tools. Additionally, translating the broader benefits of new AI solutions into individual benefits for workers in different roles will ensure that workers welcome the infusion of AI into day-to-day operations.
Enable optimal use of AI tools
Hands-on training in the use of the technology should be the last step in the user education strategy. Once management and end users are sufficiently motivated to use new AI solutions, they will be more receptive to instructions for use. They will thus be able to better contribute to AI initiatives and participate in their success.
This user education process should not be viewed as a linear, one-time activity aimed at mitigating AI failure. It should be viewed as a cycle that begins with the discovery of new applications and ends when those applications become an integral, value-added part of regular business operations. Companies looking to implement AI in the near future can start now by educating their employees on why they should not look to the AI-driven future with fear but with hope.