JASPER, IN—More and more manufacturers are investing in artificial intelligence technology to increase productivity and decrease costs. Applications include everything from advanced data analytics to cybersecurity.

“AI tools can be leveraged to protect your data and get the most benefit out of what you already own,” says Tim Pritchett, engineer operations manager at Matrix Integration, a company that specializes in IT infrastructure. Because AI is becoming a more commonly built-in component of many managed software suites, there are three issues that manufacturers should consider:

• Data quality matters. “Whether businesses are using AI to generate content or analyze production efficiencies, high-quality data is necessary to train AI models,” claims Pritchett. “Already, biased inputs in large-language models like ChatGPT have led to biased outputs that could damage a company’s reputation on a great scale. In the case of data analysis, inaccurate or damaged data fed to an AI model will lead to unusable outputs.
• Data security isn’t guaranteed. Manufacturers need to consider how they secure their own data, as well as data supplied by customers and suppliers. “This requires asking questions and developing transparency and trust with cloud service providers, as well as AI vendors,” warns Pritchett. “For example, many businesses provide customer-facing chatbots run by AI. Is data safe once it gets uploaded into a cloud-based AI application?”
• Humans are key for AI to work properly. “Right now, much of AI seems to be a ‘black box,’” says Pritchett. “Most people understand the inputs and outputs, but are unfamiliar with how learning algorithms work and how they handle data. Experienced security professionals can play a key role by fine-tuning notifications and building automation for these tools.”