Artificial Intelligence and Other Emerging Technologies

Should we be worried?

By Ed Butts, PE, CPI

Happy New Year! As I indicated in last month’s column, this year’s topics will be primarily dedicated to some of the favorable technological and industry improvements and updates as well as the not so favorable changes many of us have witnessed throughout the past 50 years.

As someone who began my career in 1974, I can certainly attest to, and in some cases lament, the changes I have observed over the past half-century. This month, in conjunction with this month’s issue theme of diversification and emerging technologies, we will kick things off with a discussion on artificial intelligence (AI), robotics, and smart systems—and whether we should worry about these new technologies or embrace them as additions to our arsenal.

Change Is Inevitable

Although predicting the exact changes and technological improvements that will occur in the future may not be possible, we all know change is undeniable and will undoubtedly occur. Change in our lives, whether personal or professional, is constant and inevitable.

This so-called progress is generally necessary, and in most cases, desirable and beneficial for most. Many of these changes are simply improvements to existing technology or methods while others are introductions of new equipment or methods intended to improve an existing task or system or facilitate the function of new science and technology.

During my lifetime, I have witnessed countless changes and massive improvements in technology. For example, the Soviet satellite Sputnik had been in space for less than a year when I was born. Most technological improvements have occurred during the past 50-plus years since 1970 with many directly tied to the development of integrated circuits needed to accommodate the explosion of computer technology during the space program of the 1960s.

Disregarding these many real-world and awe-inspiring benefits from the space program, most of the changes that have affected ordinary folks like you and me have been less dramatic but just as life and career altering.

In fact, many of the wonderous improvements for conducting business introduced during the 1970s and 1980s are now considered to be outdated if not totally obsolete. A good example of this is the fax machine.

The fax machine during the 1980s was a method of transmitting and receiving written correspondence using thermal paper in the beginning and plain paper later on. However, while positive, the business world’s need and desire for instantaneous and accurate transfer of data was still at the mercy of an often-cantankerous machine and unreliable and slow telephone connections.

The eventual evolution of the personal computer, email, and more efficient data transfer not only mostly eliminated the need for day-to-day use of a fax machine but a large part of the U.S. Postal Service as well. There’s no reason not to believe that future technological improvements will replace many of the systems we now use and consider advanced. Only time will tell.

What AI Is

Artificial intelligence (AI) is defined as technologies that enable computers to perform a variety of advanced functions including the ability to visualize, understand, and translate spoken and written language; analyze and interpret data; generate and submit recommendations; and many more tasks.

AI is not a substitute for human intelligence, but a field of science concerned with developing advanced computers and related machines that can reason, learn, and perform in such a way that would normally require human intelligence and intuition. AI can also be used for situations that involve the use of data when the scale and scope exceeds a human’s ability to analyze it within a given time.

AI is a broad field that encompasses many different disciplines including computer science, data analytics and statistics, hardware and software engineering, linguistics, neuroscience, and even philosophy and psychology.

Because of this, it is crucial, now more than ever, to know the basics of artificial intelligence and how it will ultimately fit into our daily lives. On an operational level for business use, AI is a set of technologies that are based primarily on machine and deep learning used for data analytics, predictions and forecasting, object categorizing, natural language processing, recommendations, and intelligent data retrieval.

Games such as poker and chess have been historically considered as games that use strategy and abstract thinking, traits typically possessed by humans. However, given the correct algorithms and programming, AI has recently been able to defeat chess grandmasters at their own game.

What AI Is Not

Many people misrepresent and exaggerate AI’s ability to surpass and replace human intelligence and think it will eventually develop control over humans and society. Consider the television series Westworld and the movie 2001: A Space Odyssey as two examples of terrifying Hollywood visions of AI.

The benign use of artificial intelligence is nothing to fear, although it can obviously be manipulated by people with bad intentions. As with almost any new science or technology, the potential danger exists with humans behind its development and use, not with the technology itself.

As Powerful as the Data It Is Fed

AI is a process of collecting information and data, while sorting and learning from what is input and stored. This enables computers to perform tasks that normally require fundamental human intelligence. The key to the appropriate use of AI is to mimic, but not surpass, human intelligence. This means there must be a level of human-input planning, learning, problem-solving, representation and use of knowledge, motion, and manipulation as well as some sociable skills and creativity to a lesser extent.

Breaking AI down to its fundamental elements is about understanding the importance of data and how machines can be trained to synthesize, analyze, and act on input data as a human being could. Consider AI as an engine fueled by data. It consumes data, recognizes patterns in that data, learns from those patterns, and can act based on those patterns.

Just as better fuel quality leads to stronger engine performance, leaders across industries are realizing that using good and complete data is crucial if they want more accurate and insightful AI predictions, which in turn strengthens their business operations.

Combined with machine learning (ML), a subdivision of AI that analyzes large volumes of data, AI and ML can find patterns via algorithms, enterprises, organizations, and governments and are able to perform impressive feats that ultimately drive innovation that result in better business.

AI is a collection of continually developing interactions between computers and data. It spans data integrity and prediction, data science, machine learning, and more and can be seen as both an application of technology and a school of thought, much like physics.

Generative AI

A recent addition to AI is generative artificial intelligence. Generative AI is a type of artificial intelligence technology that can produce models of various types of content including text, imagery, audio, and synthetic data. Generative AI can be considered as a machine-learning model that is trained to create new data, rather than making a prediction about a specific dataset.

Therefore, a generative AI system is one that learns to generate more objects that look like the data it was originally trained on.

From my perspective, I believe our current age of AI will ultimately transform business operations, medicine, education, engineering, and our personal lives. Though film and television like to startle their audiences with sensational displays of AI gone rogue, I like to believe the reality of AI is that it will simply translate to a better understanding, improved processes, and liberation of the human mind.

The Potential Upsides

As with any new technology, there are potential upsides and downsides to the use of artificial intelligence. Here are a few upsides.

1. Automation: AI can automate workflows and processes or work independently from a human team. For example, AI can help automate aspects of cybersecurity by continuously monitoring and analyzing network traffic and possible hacking. Similarly, a smart factory may have dozens of different kinds of AI in use, such as robots using computerized vision to navigate a factory floor or to inspect products for defects, create digital twins, or evaluate real-time analytics to measure efficiency and output.
2. Reduced Human Error: AI can eliminate many human errors in data processing, analytics, assembly in manufacturing, and other tasks through automation and algorithms that follow and adhere to the same processes and routine every single time.
3. Eliminates Repetitive Tasks: AI can be used to perform repetitive tasks, freeing humans to work on problems of a higher impact, immediacy, and importance. AI is now used to automate several mundane processes, like verifying documents or answering and transcribing routing phone calls.
4. Fast and Accurate: As much as we may want to dispute this, AI can process more information more rapidly than a human, finding patterns and discovering relationships in data that a human may not be able or have the time to do.
5. Infinite Availability: AI is not limited by the time of day or work shifts, a need for lunch or breaks, or other human needs. When running in the cloud, AI and machine learning is “always available and on,” continuously working on its assigned tasks.
6. Accelerated Research and Development: The ability to analyze vast amounts of data quickly can lead to accelerated breakthroughs in research and development. For instance, AI has been used in predictive modeling of potential new pharmaceutical treatments, or to quantify the human genome.

Some of the Downsides

Obviously, as with all new technology, there are and will be certain downsides to using AI and machine learning to complete tasks. All in all, I believe AI will be a positive outcome for the human workforce in the long run, but that’s not to say there won’t be some growing pains during the transition. This doesn’t mean we shouldn’t use AI but know its limitations so we can use it properly. Here are some examples.

1. Lacks Creativity: AI bases its decisions on what has historically occurred in the past. Thus, it’s not well suited to come up with new or innovative ways to examine old problems or situations. In many ways, the past is a good guide as to what might happen in the future, but it won’t provide perfect or infallible data. There’s always the potential for an unknown variable which resides outside the range of expected outcomes. Because of this, AI works well doing the grunt work of data collection while limiting the overall strategy decisions and ideas to the human mind.

From an investment perspective, the way we implement this is by having our financial analysts come up with an investment thesis and strategy, and then instruct AI to implement that strategy. We still need to program AI as to which datasets to examine to get the desired and best outcome for our clients. We can’t simply tell AI to “generate good returns.” We need to provide an investment universe for the AI to examine and then provide the parameters on which data points make a good investment within the given strategy.

2. Potentially Reduces Employment: It’s probably fair to include this consideration because it’s a common argument now being forwarded against the use of AI. But the truth is some uses of AI are unlikely to significantly impact human jobs.

For example, the image processing AI now uses in many new cars which allows for automatic braking in the event of a potential crash is not replacing a job. But on the other hand, an AI-powered robot assembling those cars in the factory is definitely taking the place of a human worker.

The important point to keep in mind is that AI in its current form is developed to replace humans in dangerous and repetitive work. That frees up human workers to do work which offers more ability for creative thinking, which is also likely to be more fulfilling. AI technology will also allow for increased inventions, which will help workers be more efficient in the work that they do.

3. Raises Possible Ethical Dilemmas: AI is purely logical: it makes decisions based on preset parameters that leave little room for human nuance and emotion. In many cases this is a positive as these fixed rules are part of what allows AI to effortlessly and dispassionately analyze and predict huge amounts of data. In the reverse, it makes it difficult to incorporate human-based traits such as ethics, compassion, and morality into the algorithm.

The output of the algorithm is only as good as the parameters its creators set, meaning there is obvious room for potential bias within the AI itself.

For Star Trek fans like me, think back to the 1968 episode titled “The Ultimate Computer” in season two when the M-5 computer’s inventor inserted his personal “engrams” onto the computer, rendering it incapable of making rational, informed, and moral decisions except for those its inventor would make. Certainly, if this is what’s in store for the future, it’s easy to see how this can become a challenging problem to address.

4. Can Be Abused If Not Managed Properly: As with virtually any technology, artificial intelligence has the potential of abuse and maliciousness by cybercriminals and associated parties. Although it cannot directly create or cause these conditions, human manipulation has and can result in abuse. While AI can support businesses, critical infrastructures, and industries as well as help solve some of society’s greatest challenges, these technologies can also enable a wide range of digital, physical, and political threats.

For enterprises and individual users alike to remain protected from malicious actors who are out to misuse and abuse AI, the risks and potential malicious exploitations of AI systems need to be identified, fully understood, and controlled.

Robotics

Robotics is a branch of bio and mechanical engineering combined with computer science that involves the conception, design, manufacture, and operation of specialized machines, that is, robots.

The objective of the robotics field is to create intelligent machines that can assist humans in a variety of ways, physically and mentally. A robot might resemble a human or be in the form of a robotic application, which simulates how humans engage with software to perform repetitive, procedural, or rules-based tasks. As such, robots are often used to perform dull, dirty, or dangerous tasks in the place of a human.

While the field of robotics and exploration of the potential uses and functionality of robots has grown substantially in the 21st century, the idea certainly isn’t new. Many science fiction movies of the 1950s will attest to that. Today, industrial robots, as well as many other types of robots, are generally used to perform mostly repetitive tasks. They can take the form of a robotic arm, a collaborative robot (i.e., cobot), a robotic exoskeleton, or a traditional humanoid-appearing robot.

Industrial robots and robot arms are now widely used by manufacturers and warehouses throughout the United States and world. Computer programming and algorithms, remote-controlled manipulators, actuators, control systems, and real-time sensors to provide physical dexterity, along with automation, control what a robot or robotic system does.

Water Well and Supply Professions

Artificial intelligence and robotics have already made inroads in the water well industry. While the final impact of artificial intelligence on the water well and water systems industries remains to be seen, there is no doubt the use of AI combined with improved smart devices will continue to play a greater role in how wells and water systems are built and function.

AI will continue to develop and expand in mathematically dependent areas such as hydraulic modeling, pump selection based on flow rate and head, and system frictional C values and losses. When dealing with such critical and time-sensitive variable functions of flow and pressure, as most water system designers and operators regularly deal with, the need for reliable and accurate data becomes paramount.

But this data is only a tool which requires someone with the accumulated knowledge and experience that only human beings can provide to be available to interpret, analyze, and most importantly, verify the data.

Although temporarily shutting down a production line may cost a few dollars in lost revenue in a factory setting, this same type of event if allowed to occur with treatment of a potable water system can be potentially disastrous to water potability and human health in just a few minutes.

The conclusion from this diatribe of mine simply means that in my opinion the use of smart devices, artificial intelligence, and similar computer-based systems should be reserved for the repeatable and mundane functions.

By that I mean functions such as individual facility status and reporting, overall system monitoring and observation, alarm monitoring and notifications, pre-programmed setpoint adjustments, programmed pump selections based on known parameters, and process oversight. I think those functions are okay, rather than having AI making the actual decisions related to chemical ratios and feed rates, source selection, responding to atypical water level or pressure changes, and many other parameters which must be reserved for human operators.

In fact, if properly configured and programmed, AI-based computers and smart systems can enhance the abilities of present human intelligence and streamline decision making instead of actually making and implementing the decisions themselves.

AI Applications in Water Well Industry

One of the promising applications in AI is a technology called electrical signature analysis (ESA). As an example, ESA makes reliable and remote monitoring of submersible pumps and motors possible. ESA measures the current and voltage being supplied to the motor driving the pump. Because it captures the motor’s vital electrical signals, ESA sensors are current and voltage probes that are installed in the motor control cabinet.

This bypasses the need to install sensors directly on or near pump motors. When powered by AI and machine-learning algorithms, gigabytes of raw data can be translated into comprehensive performance insights to inform and support a condition-based monitoring (CBM) approach, allowing users to schedule maintenance at a convenient time, thus avoiding unplanned downtime.

Another potential benefit to the use of AI is the expanded use of variable frequency drives. With the proven benefits of ESA technology, increasing use of VFDs presents a powerful opportunity to provide enhanced condition monitoring data to offer complete insight into the health of the pumping unit’s full drive train.

VFDs already generate electrical data relating to the control and operation of the motor and the pump it is driving, such as input and output current, voltage, and supply frequency. Since ESA measures high-quality current and voltage data generated at the motor’s supply, there is an opportunity to integrate the technology into the VFD sufficient to enable remote condition monitoring from the drive, motor, and pump.

Eventually most municipal, commercial, and industrial water purveyors will likely reach similar conclusions, viewing AI and machine learning as a process that will ultimately assist and empower human beings and teams to be able to make timely repair decisions rather than being a magic bullet itself.

Although robotics will undoubtedly continue to play a major role in how we manufacture, transport, and move around items in our homes, businesses, and industries, the ultimate application and use of this technology to the water well and supply industries is largely unknown at this time.

More than likely, when the technology of AI and robotics expand in the water well industry, they will be limited to assisting with the more mundane and routine procedures involved with drilling and pump installation work, such as drill rod joint handling, assembly, and disassembly (illustrated in Figures 1a-1c), cementing sanitary seals and placing annular filter packs in wells, downhole well inspection, rehabilitation and repair, and heavy equipment loading and unloading drill rods, pump bowls, and column and shaft.

The use of robotics in the water supply industry will most likely involve routine maintenance and service of conveyances and storage facilities, such as inspecting and cleaning water conveyance channels and pipelines and underwater inspection and cleaning, as well as robotic welding of pipelines and water storage vessels.

______________________________________________

Well, this concludes this first Engineering Your Business installment for 2025. During the rest of the year, we will examine improvements in many aspects of the water well industry including well rehabilitation, water pump types, materials and construction, drilling rigs and pump hoists, and the influence of computers and solid-state electronics on the water well industry to name just a few. We will continue next month with an overview of new downhole video camera systems.

Until then, work safe and smart.

---

Ed Butts, PE, CPI, is the chief engineer at 4B Engineering & Consulting, Salem, Oregon. He has more than 40 years of experience in the water well business, specializing in engineering and business management. He can be reached at epbpe@juno.com.