AI Pragmatism: Choosing the Right Tool for the Right Problem

AI Pragmatism: Choosing the Right Tool for the Right Problem

It's easy to get swept up in the allure of the newest, shiniest tool on the market. Generative AI, with its ability to create content that mimics human output, has captured the imagination of many. But is jumping on the Generative AI bandwagon the best move for every organization? Perhaps not. The key to successful technology adoption, especially in the nuanced field of AI, is not to chase after the latest trend but to have a clear understanding of the problem you're aiming to solve and then selecting the tool that best addresses that issue.

Consider the approach of Sysco 's Chief Information and Digital Officer, Tom Peck, who said, "I don’t need a Generative AI strategy. What I need is an automation strategy." This perspective underscores a critical point: the solution should fit the problem, not the other way around. It's akin to using a Swiss Army knife when all you need is a screwdriver. The elegance lies in simplicity and precision, not in the complexity or novelty of the tool.

This principle is exemplified by organizations like KLM Royal Dutch Airlines and Boeing , which integrate AI into their operations but do so judiciously, applying it where it makes sense and sticking with more traditional methods where those are proven to be effective. For instance, KLM employs AI to predict flight no-shows, enhancing operational efficiency, while Boeing utilizes AI for route optimization. Yet, both companies understand that not every problem is an AI problem, and sometimes, the old ways—like direct voice communication in air traffic control—remain the best ways.

Airbus , in its quest to expedite the production of the A350 aircraft, chose to employ AI not because it was the most avant-garde option available but because it was the right tool for the job at hand. The AI-based system Airbus developed was designed to recommend solutions for production disruptions, a clear case of technology serving a precise, practical need. Matthew Evans, Airbus's Vice President of digital transformation during the A350's production launch, encapsulated this approach when he remarked, "Strictly speaking, we don’t invest in AI... We’re always investing in a business problem."

This brings us to the crux of the matter: understanding the landscape of advanced analytics and AI, not just in breadth but in depth. While it's impractical for leaders to be experts in every analytic technique, having a grasp of the different categories of analytics tools—Generative AI, traditional deep learning, econometrics, and rule-based automation—enables informed decision-making. Each of these tools has its domain where it shines brightest, and recognizing this can prevent the costly mistake of deploying a complex AI system where a simple linear regression would suffice.

It's vital to pose pertinent questions: What are the consequences of inaccuracies? Is it necessary for decisions to be transparent? Does consistency and repeatability matter? Contemplating these factors can assist you in choosing the analytic tool that best addresses your organization's distinct challenges.

The relentless pursuit of AI for AI's sake is a fool's errand. The true measure of an organization's technological maturity is not how many cutting-edge AI solutions it has deployed but how well it understands and addresses its core problems with the most effective tools, AI or otherwise.

I call this AI Pragmatism, defined as the judicious application of AI technologies where they are genuinely needed and can add real value, avoiding the trap of using AI as a hammer seeking nails. This concept emphasizes the importance of a problem-first, technology-second approach in the realm of AI and analytics.

While the allure of Generative AI and other advanced analytics tools is undeniable, the key to harnessing the true power of these technologies lies in a clear-eyed assessment of what your organization needs to achieve. By starting with the problem and asking the right questions, you can ensure that your technology strategy is not just about keeping up with the latest trends but about driving real, tangible outcomes for your business.

Andrew Peterson

Independent Civic & Social Organization Professional

1 年

thanks for sharing. There have been a lot of cool developments in the AI field lately. This one platform in particular is very interactive and customizable, you can give it a go too if you want: it's called textcortex

回复

要查看或添加评论,请登录

Edward Lewis?的更多文章

社区洞察

其他会员也浏览了