Gleb Tsipursky, an AI consultant and author of eight books, argues in an article in next month’s Textile Services that many AI initiatives fail because laundry operators introduce the technology too broadly and too quickly.
Leaders often purchase sophisticated software, showcase impressive features and expect immediate productivity gains. Instead, employees become skeptical. Supervisors question the accuracy of AI-generated information, frontline workers worry about job security and experienced staff doubt that AI can match their practical knowledge gained from years of operational challenges.
Tsipursky recommends a different approach: begin with a specific, recurring problem that employees already recognize. Rather than transforming an entire operation, companies should identify time-consuming tasks that create daily frustrations. Examples include compiling production reports, searching maintenance records or drafting customer communications. When AI addresses a clearly understood problem, employees can easily evaluate whether it provides real value.
One of AI’s most effective early applications is helping employees locate and use information more efficiently. In commercial laundry operations, workers often need quick access to standard operating procedures (SOPs), maintenance records and quality-control documentation. During busy shifts, locating the correct information can be difficult and time-consuming.
A secure AI assistant trained on approved internal documents can quickly answer questions about procedures, quality exceptions or required documentation. However, the technology should not replace human decision-making. Its role is to help employees find the right information faster, not to make operational judgments. By shortening the gap between workers and approved procedures, AI improves efficiency while maintaining accountability.
The article emphasizes that strong oversight is essential, especially in areas involving safety and compliance. Commercial laundries face hazards related to machinery, chemicals and workplace safety regulations. Because of these risks, employees should never treat a general-purpose chatbot as the final authority on critical safety matters.
Instead, organizations should establish clear AI guardrails. Leaders must define what information can be entered into AI systems, what tools employees may use, when human approval is required and which tasks remain off-limits. Sensitive information such as customer data, employee records, pricing information and proprietary processes requires particular protection.
The author proposes a simple guiding principle: the greater the consequences of an error, the greater the need for human review. AI can assist with tasks such as translating procedures, drafting safety materials or creating training quizzes, but qualified professionals must verify all outputs before implementation.
Tsipursky recommends testing AI through small, controlled pilot programs. Organizations should choose a frequently performed task, establish performance benchmarks and measure outcomes against specific goals. For example, a company might examine whether an AI tool can reduce report preparation time from 30 minutes to 10 minutes without increasing errors.
The focus should remain on measurable business results, rather than activity metrics such as logins, prompts or training participation. Valuable indicators include faster customer responses, reduced administrative burdens, improved troubleshooting and more consistent documentation. Maintenance and training functions offer especially promising low-risk opportunities, as AI can organize historical records, summarize technical information and create training materials while leaving final decisions to skilled employees.
The author repeatedly stresses that AI should support human expertise rather than replace it. Maintenance technicians must still diagnose equipment issues based on physical conditions such as vibration, heat and wear. Quality teams can use AI to summarize reports and identify patterns, but they should never allow software to determine compliance or product quality.
Operators should view employee skepticism of AI as valuable operational insight rather than resistance to change. By involving frontline workers, requiring evidence of success and maintaining human accountability, organizations can implement AI in ways that enhance productivity without undermining safety, quality or trust. Ultimately, the most effective AI deployments are not the most ambitious, but those that eliminate friction in specific workflows, while preserving the skilled judgment on which operations depend.
Tsipursky concludes that successful AI adoption resembles a continuous improvement process, rather than a major technology rollout. Organizations should identify a bottleneck, test a limited solution, measure results, refine the process and gradually expand successful applications.
Watch for the full text of the article, titled “Artificial Intelligence: Win With Measured Steps,” in September’s Textile Services magazine. Click here for more information or to subscribe to Textile Services.
Sign Up For Our Newsletter
Receive the latest updates on the linen, uniform and facility services industry from TRSA delivered straight to your inbox.
