The current discourse within the technology ecosystem assumes that Artificial Intelligence (AI) and Machine Learning (ML) are miracle solutions to all data-related problems. However, voices from across industries are beginning to express concern that AI, under the influence of hostile actors, could threaten our very existence, from mass unemployment to nuclear war.
However, the reality of AI capabilities is much more nuanced than manufactured headlines allow. This lack of nuance is affecting companies and boardrooms across all industries, whether they use AI technology or not.
The problem for many companies looking to implement AI is not fully understanding the problem they need to solve before turning to technology to solve it. This not only creates complications, but also encourages companies to implement AI as a strategic objective, rather than as an application that will truly revolutionize their business. Therefore, avoiding the dangers of poorly articulated strategies for solving data problems is a fundamental challenge.
This checklist is a starting point to help identify problem-solving methods and ensure that companies prioritize organizational impact over solution technology.
Director of Government and Advocacy at Mind Foundry.
Prioritize the human over the system
Optimizing only with the system in mind and prioritizing only technology instead of focusing on human effectiveness is a common mistake. Conventional wisdom assumes that system optimization will speed up processing times and computing capabilities, or that the data architecture will become more orderly. While this makes sense from a systems engineering perspective, if they are not usable or do not provide significant impact to the end user, the overall system will not improve. Increasing efficiency in non-bottleneck areas is a wasted effort, and improving system efficiency at the expense of usability can have a negative impact. Efforts should always focus on designing with the human being in mind, rather than the system.
Don't limit the solution within an existing system
Designing within the limits of existing systems introduces assumptions and creates problems that artificially inflate complexity. Instead, breaking out of an existing system allows for a curiosity-driven approach to figuring out how best to use the system and identifying what the problem really is. A restricted starting point only leads to immediate inefficiencies, since the system is optimized for some mistaken notion of the true goal.
A blank slate is not always the best
Balancing what exists and what could be is a challenge. It's easy to talk about starting completely from scratch and implementing an AI-powered backbone, but the reality of implementation is more challenging. The “blank slate” approach risks losing years of hard-earned knowledge, understanding, and best practices. Replacing this with a generic, poorly designed AI system only confuses existing users. Combining old systems with new improvements will allow for more significant gains in productivity. However, this combination requires a strategic approach with collaboration between AI vendors, systems engineers and end users.
Automation is not the same as AI
The term “AI” can sometimes be misleading. The goal is often to implement AI, which means we sometimes get distracted by the question of whether or not a technology is “AI.” The priority, however, should be finding the best tool for the job, not the best job for a tool. The most technologically advanced solution may not be optimal. That is why it is vitally important that AI is understood along with a variety of other tools in order to select the best one for any given scenario. The result of this should allow for more simplified implementation, governance and maintenance.
The data itself could be the most valuable
The assumption that all data is valuable leads to long journeys in search of this value, which is ultimately a waste of time. Instead, companies should consider how extracting value from data increases the likelihood of achieving a business or operational objective. We often collect data simply because we can or because it is a byproduct of “business as usual.” Many organizations waste time speculatively refining data in areas of the business that are not at the top of the list simply because they can. The priority should be collecting the right data to solve important problems.
These recommendations provide a starting point for companies looking to pursue data-driven, impact-focused projects. Asking the right questions is a critical first step and should be followed with an approach that designs solutions with implementation in mind from the beginning.
We have introduced the best data migration tool.
This article was produced as part of TechRadarPro's Expert Insights channel, where we feature the best and brightest minds in today's tech industry. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing, find out more here: