AI Mindset: Adopting the Right Framework for AI Implementation

By on

Click to learn more about author Andriy Latysh.

Businesses today are facing a key turning point in AI adoption and, as such, are perfectly positioned to reach new heights with AI, by gleaning actionable insights, improving operational efficiency, deciphering customer behaviors, and reducing costs. That said, just because an organization is willing to adopt AI does not always mean the deployment process will be trouble-free. The fact is, in many cases, non-IT management may not understand the AI process, which can lead to friction. Adopting AI means also embracing the right mindset. Below are five ways that management in any organization can help facilitate AI implementation.

1. Start at the Top

One of the first obstacles to AI adoption is that executives often don’t have a grasp of the profound changes that AI can have on the company. It’s important to present – and implement – AI as it truly is: a means for cost reduction, improved efficiency, better products, and better services. Another common mistake is to think of AI as a kind of tool that comes out of a box.

Even when demand for AI comes from the executive level, it’s important to make sure the first AI projects show results in order to maintain executive backing. However, results from AI can take some time, so it’s important to create realistic expectations of how the results will manifest and the time period in which the executives can expect to see those results.

Whether or not the executive team is on board, the most important element in the AI strategy is to correlate it directly to the business strategy and goals. Implementing an AI strategy offers an opportunity for the team to take a step back from day-to-day operations and address the company’s greatest challenges. Is the business environment changing? Is the operations department overly complex? Are customers receiving mediocre service? AI can address many of these problems – but it’s important to prioritize the actual challenges.

So, AI projects should involve a detailed study and playbook that details the objectives, timeline, data sources, cost, key performance indicators, and next steps based on the discoveries.

2. Expertise and Education

While data scientists are the experts in how to implement AI, they should also be the experts in the company process, goals of each department, and company strategy. In any organization, it’s important for the experts on the ground to articulate the challenges and opportunities in their specific department or areas of responsibility to the data scientists as soon as possible. Deep discussion between the domain experts and the data scientists will lead to clarity around the opportunities that AI can address in every realm of the company’s operation. From that perspective, it’s important to reinforce to the employees that the more they share with the AI project managers, the better their needs can be met. A well-trained AI team or consultant will know how to ask the right questions and build trust to discover the challenges and develop solutions.

For example, the GoCheck app consulted an AI company with a problem that came from their user experience team. The app is used to diagnose children with lazy eye and other curable vision impairments. Nurses and practitioners could take a picture of the child and the app sends the images for processing, which diagnoses a variety of conditions. The processing didn’t happen in real time, and if a child didn’t look at the camera, the algorithm would return “no diagnosis.” Using AI, the team was able to integrate an instant notification when the child was not looking at the camera, so the practitioner could take a new photograph right away. This is a straightforward case, but one that could never be discovered by data scientists unless they were working together with the product team.

Another approach that companies take is to educate their staff on how to think about AI. Online courses such as “AI for Everyone” are designed specifically so that the staff members of any organization can identify and communicate the potential applications of AI in their realm of the business.

3. Avoid Data Scientist Silos

While every AI project needs a project manager or champion, it’s important to ensure integration with the rest of the enterprise. When data scientists are siloed into a separate team, they encounter difficulty in implementing the projects – usually because they are either misaligned with company strategy, or they are unable to get the support of the business and IT teams. Data scientists should be embedded into the teams they are serving. This allows the data scientists to understand the full context of the work, iterate together with the team, and produce software that aligns with the outcomes the business teams want.

AI takes time to evolve and requires constant creative and material investment. Working within a business team allows collaboration and multiple perspectives throughout the iterations of the technology. Combining technological resources with the creative people within the business provides the best outcomes.

4. Multi-Level Strategy

Deciding whether to hire in-house or outsourced AI consultants depends on the size of the company and how they approach different problems. Generally, a multi-level strategy works best for enterprises. The project champion needs to be an in-house employee, but a combination of outsourced consultants and internal AI professionals often provides the best combination of expert advice with dedicated employees. Partnering with an AI company opens up a wider range of options than most internal teams can provide in terms of range of solutions, industry benchmarking, and the variety of AI technologies.

5. Next Steps and Cultural Transitions

Implementing AI solutions often requires a rethinking of internal processes. For example, a company may use AI to identify customers who are at risk of switching to another vendor. This introduces a new procedure in the company – handling these customers. The company will need to figure out who gets the information as well as how to communicate with these at-risk customers. Similarly, if it turns out that a large number of at-risk customers belong to a particular account manager, there may be a problem with that specific employee, and the company will need a process for either retraining the employee or transitioning the customers to a new associate. Similarly, in working with a customer service application, AI can determine which client queries can be answered with automated messages, and which ones need to go to human service representatives, but that means the organization needs to prepare the infrastructure for the automated responses, as well as have more highly trained staff to answer complex questions.

In other words, AI can indicate areas where action is necessary, but the results will only occur if the company implements new procedures for taking those actions.

AI can have a profound positive impact on enterprises, but it’s important to take into account the cultural and process changes along with the technical implementation. The key to adoption is to adopt a phased approach, implementing projects within individual departments and getting a feel for both the scope of the benefits and the scope of change the company can handle. After successful smaller-scale implementations, enterprises can scale up their AI efforts to organization-wide implementations.

We use technologies such as cookies to understand how you use our site and to provide a better user experience. This includes personalizing content, using analytics and improving site operations. We may share your information about your use of our site with third parties in accordance with our Privacy Policy. You can change your cookie settings as described here at any time, but parts of our site may not function correctly without them. By continuing to use our site, you agree that we can save cookies on your device, unless you have disabled cookies.
I Accept