Spring 2023

The Chatter About AI

The growing presence of generative AI chatbots energizes discussions about emerging tech and engineering practice.

BY DANIELLE BOYKIN

OpenAI’s ChatGPT came on the scene in November 2022. This generative artificial intelligence technology processes written or spoken human conversation and can generate information in response to various queries and requests. This type of generative AI has made a noticeable impact in a brief period of time and other versions of chatbots (Bard, Bing, ChatSpot, Drift, etc.) have been released. And just in case you’re wondering, this article was not written with the assistance of a chatbot.

A discussion about artificial intelligence and ChatGPT recently took place on the NSPE Communities’ Open Forum (Will AI Take Your Engineering Job?) with varying perspectives on AI and what effect it could have on the engineering workforce. Can this tech be useful with some aspects of engineering design? Will this technology be a "job killer" or replace engineers? The conversations also reinvigorated a discussion about what role licensed professional engineers should take on in the development and deployment of artificial intelligence and other emerging technologies.

According to a Deloitte AI Institute report, "generative AI has the potential to add contextual awareness and human-like decision-making to enterprise workflows, and could radically change how we do business." This led Deloitte to launch a new practice to advise clients on the transformative impacts of this technology and help them plan, build, implement, and operationalize solutions built on the leading foundation models. The report also warns that this AI technology is in its infancy and could pose risks related to privacy and security, managing bias, transparency and traceability of results, IP ownership, and equal access, especially for those at greater risk of job displacement.

Ben Amaba, Ph.D., P.E., LEED AP, CPIM, is no stranger to how artificial intelligence and other emerging technologies can have an impact on various industries and organizations. During a conference session at NSPECon22 in Philadelphia, he discussed the use and misuse of artificial intelligence in critical infrastructure systems and the importance of ensuring safety.

New technology and process improvements are essential for growing economies and improving the quality of life, says Amaba. He doesn’t believe that this type of AI will put engineers, particularly licensed professional engineers, out of their jobs. "Innovation and advancements will continue to drive change in the workforce, just as the agricultural and industrial revolutions did in the past. However, it’s crucial to put the appropriate guardrails in place to protect the public’s welfare," says Amaba, who serves on the NSPE Board of Directors and is a Fellow with the Institute of Industrial and Systems Engineers. "We must ensure that these changes are implemented responsibly and with consideration for the potential impacts on society."

NSPE supports emerging technologies in a manner that protects the health, safety, and welfare of the public through rigorous development, testing, and deployment of emerging technologies involving licensed professional engineers at each step of development (Professional Policy No. 03). NSPE recognizes the benefits of continued research and development that safely advances emerging technology while minimizing risk. Society volunteers and business entities are directed to implement this policy in relevant position statements, operating procedures, and business practices.

Public Safety and Ethical Practice

The use of chatbots and other artificial intelligence systems also raises ethical concerns.

While AI can be a valuable tool, its vulnerabilities for errors, human bias, and cyberattacks must be minimized through careful design and implementation to protect the public. The National Institute of Standards and Technology released an artificial intelligence risk management framework in January. The report emphasizes that AI risk management can drive responsible uses and practices by prompting organizations and their internal teams who design, develop, and deploy AI to think more critically about potential or unexpected negative and positive impacts. This is critical to cultivating public trust.

"To protect against these risks, it’s crucial to implement robust security protocols and to ensure that AI systems are designed and trained with ethical considerations in mind. This includes using diverse data sets, testing for bias, and incorporating human oversight and intervention when necessary," says Amaba.

He adds, "As the use of AI continues to grow, it’s important to remain vigilant and take a proactive approach to mitigate risks. By doing so, we can ensure that these powerful technologies are used responsibly and for the benefit of society."

For decades, NSPE has played an active role in ensuring that responsible engineering practice is at the forefront of emerging technologies. In 2007, the Society was a part of the Software Engineering Licensure Consortium to develop a software engineering PE exam. In 2020, the Emerging Technologies Task Force provided guidelines to adopt standards to protect public safety and measure the readiness of technologies before they are deployed. NSPE also formed the Software Professional Certification Task Force to evaluate the feasibility of developing a new certification to assist with providing greater public protection (This action was deemed necessary when NCEES discontinued offering a PE exam for software engineering in 2018). The Board of Directors recently determined that emerging technologies should be included in NSPE’s 2023 legislative and public policy focus.

Rather than focusing on licensing status, Amaba recommends that the profession prioritize the processes and standards that ensure that all engineers perform their work with integrity, professionalism, and a commitment to the public welfare. This includes following ethical principles and best practices, staying up to date with the latest technologies and methodologies, and collaborating with other professionals to achieve common goals. "As AI continues to evolve so should we. Continuously improving on the ‘minimum standard of care’ is foundational for our society to progress," he says. "By working together and focusing on the positive aspects of engineering, licensed and unlicensed engineers can contribute to the betterment of society and help address some of the most pressing challenges we face today."

Share Your Perspective

What are your thoughts on the possible impact of generative ai and other emerging technologies on engineering practice and/or the role of licensed engineers in the deployment of this technology. Is your company/organization using AI technology and if so, how? Share your thoughts and experiences with PE at [email protected] or strike up a discussion with other NSPE members in the NSPE Communities’ Open Forum.

A New Technological Challenge

If you speak to almost any university professor or administrator, they will share that change happens slowly within academia. Generative AI is presenting a challenge to this narrative as colleges and universities are figuring out how to respond to the growing student use of chatbots.

David James, Ph.D., P.E., F.NSPE, an engineering educator, recently spoke with an English professor about the impacts of students using generative AI for essays and reports. She shared that this led some colleagues at her northwestern university to drop writing requirements. "I was horrified to hear that because the one thing we expect an English major to know how to do is write properly," says James, the director of the Solar and Renewable Energy Program and associate professor in the Department of Civil and Environmental Engineering and Construction at the University of Nevada, Las Vegas.

Since students are using this technology, educators must engage with them through the learning process, says James, "We can’t hide our heads in the sand and completely reject [this technology]. It’s too late for that. We can’t shut Pandora’s box."

Melissa Bowles-Terry, the director of the UNLV Faculty Center, says the university understands that while AI is changing rapidly, it’s important to acknowledge that AI is another tool that can be used to create new information and knowledge.

The UNLV Provost formed a task force at the beginning of this spring semester to provide assistance for faculty concerned about the use of ChatGPT and who have interests in teaching with AI. The task force created a guiding document with ideas about authentic assessment and ideas for syllabus language regarding use of AI in assignments and also held related workshops.

"As long as we’re keeping in mind the human element and how our uniquely human skills are adding to the conversation, I’m excited about using AI in education," says Bowles-Terry. "It could reshape the whole architecture of the Internet. Google search shaped the way we use the current web, and has done so for almost 20 years. This could give a new shape to information seeking."

The response to this guidance has been positive and UNLV educators are diving right in with using generative AI. "One instructor who teaches business communication had her students generate specific types of messages using ChatGPT, then analyze the problems and weaknesses with those AI messages. This seems like a fantastic ‘real life’ use of the technology," she says.

Testing It Out

Like his university colleagues, David James has tested out ChatGPT and Google’s Bard with a few different queries (see sidebar example). The queries produced what James describes as encyclopedia type summaries (without source citations) that some students may be inclined to use for term papers and reports. His assessment is that it’s not quite ready for primetime. "The problem is how do you know what sources are being used and if they are reliable. Are these chatbots using [legitimate] sources or disreputable sources from all of the nasty conspiratorial areas? That could present a huge danger."

James also concluded that this tech will not replace an engineer’s judgement at the present time, but there could be a different landscape within the next decade. "The most likely possibility is similar to when we started to use computers to replace drudgery calculations so that we could concentrate on the more difficult and creative parts of our jobs."

He adds, "As a civil engineer, every design you do is tied to a piece of ground or a piece of water. You need site specific information to design and build to meet the needs of a client or community. I don’t see that either ChaptGPT or Bard anywhere near able to do that."

James plans to address the use of generative AI at the start of the fall semester with his students in a capstone senior design course and a graduate course on introduction to research methods. The first assignment will be to evaluate and discuss the effectiveness of the chatbots in providing information for class projects. "We are going to have the conversation and see what this can do. Can you use this in a senior design report or research proposal? Nope. Can you design a bridge based on this information? Nope. Can you trust it? Don’t know," he says. "I may be surprised at the ingenuity of my students who may be far ahead of me in use of this technology. I’m going to put it right on the table at the beginning.

DAVID JAMES’ BARD (GOOGLE) PROMPT:
"REVIEW CITY OF LAS VEGAS ENGINEERING DRAWINGS"

Bard Response

"Sure, I can help you review city of Las Vegas engineering drawings. Here are some things to look for:

Accuracy: The drawings should be accurate and up-to-date. This means that they should show the current state of the infrastructure, including any changes that have been made since the drawings were last updated.

Clarity: The drawings should be clear and easy to understand. This means that the symbols and labels should be clear and unambiguous, and the drawings should be well-organized and easy to follow.

Completeness: The drawings should be complete and include all of the information that is necessary to construct or maintain the infrastructure. This includes information on materials, dimensions, and construction methods.

Consistency: The drawings should be consistent with each other. This means that the same symbols and labels should be used throughout the drawings, and the drawings should be drawn to the same scale.

Legible: The drawings should be legible and easy to read. This means that the lines should be dark and clear, and the text should be large enough to be easily read.

If you find any problems with the drawings, you should bring them to the attention of the engineer who created them. The engineer will then be able to correct the problems and ensure that the drawings are accurate, clear, complete, consistent, and legible.

Here are some additional tips for reviewing engineering drawings:

Ask questions. If you don’t understand something, don’t be afraid to ask the engineer for clarification.

Use a magnifying glass. If the drawings are difficult to read, use a magnifying glass to help you see the details.

Take notes. As you review the drawings, take notes of any problems that you find. This will help you to remember the problems and to discuss them with the engineer.

Be patient. Reviewing engineering drawings can be time-consuming. Don’t rush through the process. Take your time and make sure that you are satisfied with the drawings before you approve them.

David James’ Evaluation

This is not a specific evaluation of drawings for the City of Las Vegas. It provides a generic ‘how to’ for review of drawings that might come from a best practices manual or text book. It fails to notice that many engineering drawing reviews are now conducted electronically using review and mark up software. The response is still stuck in the paper review era.