It’s essential to understand the details, implications, and strategies when machines, software, computers, and/or microchips use information purposefully to manage, affect, and perform cleaning tasks and related operations.

Information, purpose, and control

When we supply the information and purpose to machines, integrated circuits, and/or software, we largely control the decisions and outcomes.

Examples include current robotic floor cleaning, pollutant and contaminant sensors that validate cleaning or prompt remediation, and “smart” systems that integrate data from cameras, sensors, autonomous equipment, and humans to optimize cleaning operations in predetermined ways.

Artificial intelligence (AI) has advanced rapidly in recent years due to more computing power, large language models in systems such as ChatGPT, and better algorithms, prompting the question: “Who’s in control, and what does it all mean?”

Interestingly, ChatGPT calls itself a language model, not a reasoning machine.  Humans have supplied the information and, to a large extent, its purpose, and hence, have some degree of control over outcomes.

“Language models encode what is reflected in human text rather than offering a deep understanding of it, although they may sometimes project the appearance of such deep understanding,” notes the book The Age of AI: And Our Human Future, authored by Henry A. Kissinger, Eric Schmidt, and Daniel Huttenlocher.

So, in many ways, humans still control AI, but with advancing technology, AI has more ability to “think” at least within certain limits.

A new chess master

For example, until 2017, Stockfish was the most powerful chess-playing program in the world, pre-programmed with all known moves of human chess masters. That is, until AlphaZero, an AI program developed by Google DeepMind, soundly beat Stockfish 155 times by being given just the rules of chess, the processing power to review every possible move, and no pre-programmed human chess moves. AlphaZero had learned to “think” in a sense, and no human has ever beaten it.

Still, is life as simple as a game of chess?

Limits and possibilities

Of course not, but—as a Google or AI search can attest—with copious amounts of data, powerful computing resources, and innovative algorithms applied, AI has advanced healthcare, education, entertainment, and manufacturing to improve efficiency, quality, and even customer satisfaction.

Currently, AI is not good at non-repetitive tasks; however, AGI (human-like artificial general intelligence) will be if/when it comes to our sector.

AI is potentially good at repetitive tasks in professional cleaning, such as emptying of trash, dusting, floor care, and more, with limits that relate mainly to financial considerations; for example, building the perfect dusting robot would be an expensive undertaking, one most useful where the size of an operation justifies the cost of development.

AR (augmented reality) and VR (virtual reality) applications in aviation demonstrate this and help define AI in commercial cleaning as a “helper technology.”

Helper technology—follow the money

Just as airline pilots train on simulators, workers can receive training using AR and VR.

The question is: Will they be so trained, and if yes, when?

Will autonomous equipment replace or complement workers, as they currently replace or complement workers in repetitive, tedious, dangerous, and highly precise industrial tasks?

In the current environment (2024), with the relatively low cost of entry-level labor (janitors) and the modest needs of most jobs, technology solutions will not be top-of-mind in most operations, at least as it relates to the labor pertaining to commercial cleaning endeavors.

As helper technologies (e.g., AR, VR) become less costly to access, either through supply and demand, market pressures, or rental or leasing, helper or service tech will gradually become a part of the daily lives of many workers.

This has already begun in operations where the cost of tapping these innovations makes financial sense.

Scope-of-work driven AI

AI technology will essentially emerge in a matrix linked to a scope of work unique to each cleaning operation.

Scope of work relates directly to the purpose of cleaning. Making a personal application: Our cleaning, because company is coming over in 15 minutes, is a different scope of work than if we clean to prevent our kids from getting the flu.

The selection of AI cleaning “tools,” including those with AR and VR technology, will depend on where they make sense. It will also depend on whether and to what extent identifying specific contaminants or pollutants matters and which measurements pivot to the customer and affected business(es).

AI’s future is in our hands

While the future of AI is potentially bright and exciting (see the sidebar, “AI Applications in Professional Cleaning”), it can also be largely a self-fulfilling prophecy.

“Don’t ask what computers can do; ask what they should do,” said Brad Smith of Microsoft.

“Humans still control it. We must shape it with our values,” noted “The Age of AI: And Our Human Future.”

You may have heard the story of a wise old man who could answer any question posed to him. It goes much like this:

A village boy wanted to trick him, so the boy captured a small bird, took it to the man, but held it behind his back. “Tell me, is the bird in my hand dead or alive?” he asked. Knowing the boy would crush the bird if the man said it was alive and release it alive if he said it was dead, the wise man replied, “The answer is in your hands.”

Likewise, the future of AI for commercial cleaning is mainly in our hands.

AI Applications in Professional Cleaning

We can apply AI to professional cleaning in various ways, such as:

• Autonomous cleaning equipment: AI can enable the development of smart robotic cleaning equipment that can navigate complex environments, avoid obstacles, and perform various cleaning tasks, such as vacuuming, mopping, disinfecting, and sanitizing. These innovations can reduce the need for human labor, increase productivity, and improve quality and consistency. They can also collect and analyze data on cleaning performance, the condition of the facilities, and user feedback to optimize cleaning schedules, routes, and methods.
• Occupancy sensors detect people’s presence or movement and can help determine how often areas should be cleaned and where lights and HVAC can be turned on or off to save energy.
• Data analysis can help cleaning companies improve efficiency in deploying labor. For example, if an area is frequently more trafficked, cleaning operations can adjust schedules to focus more attention there.
• Intelligent cleaning products: AI can enable the development of intelligent cleaning products that can adapt to different types of surfaces, stains, and dirt to deliver the optimal amount of cleaning agents, water, and energy. These products can improve cleaning efficiency, quality, and safety, reducing environmental impact and operational costs. They can also monitor and report the usage and performance of the cleaning products and provide recommendations for maintenance and replenishment.
• Smart cleaning management: AI can enable the development of smart cleaning management systems that can integrate and analyze data from various sources, such as sensors, cameras, autonomous equipment, products, and users, to provide real-time insights and actionable suggestions for cleaning operations. These systems can help optimize the cleaning resources, workflows, and outcomes and enhance communication and collaboration among the cleaning staff, managers, and clients. They can also provide feedback and training for the cleaning staff and improve their skills and satisfaction.
• Content creation: AI can create and translate original training content, including procedural guides and checklists.
• Business strategy: AI tools such as ChatGPT and MSCopilot can provide marketing and business growth strategies based on the criteria provided.
• Research: Large language models such as those used by ChatGPT and Microsoft Copilot enable market analyses, determining customer preferences, and more, though the results are not in real time as the information models used may be several years old.
• Customer service: ChatGPT or AI-driven chatbots can help create customer service responses, answer FAQs, direct questions to the right person or department, book appointments, and more.

Top Priority: Educating the Workforce

One definition of intelligence is that it is the purposeful ability to capture, adapt and use information.

As concerns arise regarding artificial intelligence or AI taking over, causing mischief, or worse, it’s wise to remember that AI arose from human intelligence (HI), not vice versa.

In principle, then, improving human potential through the practical application of knowledge should precede improving AI, and expanding our workers’ ability to capture, retain, and build on human knowledge and expand their skill set is a top priority.

In turn, workers imbued with a growth mindset through expanded knowledge can help inform, develop, and maintain related AI for better grassroots-driven, customer-centric, and financially attractive cleaning.