Experts on an NJBIZ panel last week took a deep dive into all things artificial intelligence, a topic at the top of mind for many individuals and businesses as the rapidly evolving technology becomes more widely adopted. Its use – and possible misuse – raises questions and concerns about how to best use the technology safely.

The discussion was hosted by NJBIZ Editor Jeffrey Kanige, who was joined a group of panelists with a variety of expertise and perspectives:

• Brigette Eagan – partner, Genova Burns
• Michael Edmondson – associate provost for continued learning, New Jersey Institute of Technology; vice president, The Learning and Development Initiative, New Jersey Innovation Institute
• Jason Gulya – professor, English and Applied Media, Berkeley College
• Sameer Sethi – senior vice president and chief data & analytics officer, Hackensack Meridian Health

The discussion touched on a number of areas, such as the differences between generative AI, machine learning and other subfields; keeping the human element involved as technology advances; specific business functions that can be carried out more efficiently with AI; how businesses using it are gaining an advantage; potential dangers and pitfalls; and much more.

The panelists ran through some of the ways that each of their organizations and clients are using the technology. A major theme centered on trying to determine the best application while not just forcing its application to “keep up with the Joneses.”

“As we said, everyone’s jumping on the bandwagon – so let’s jump on the background,” said Edmondson.  “Well, why are you doing this? There’s single loop learning and double loop learning. Single loop learning is – ‘OK, we set a goal; we didn’t achieve it.’ Double loop is – ‘Wait, what? Who said that goal was even right in the first place? Let’s go back and revisit not only why we didn’t get to the goal – but why that goal was established in the very first place. Why are we using AI – because the Joneses are doing it; we got to keep up with the Joneses.’”

Edmondson stressed that he is a big proponent and uses AI in a number of ways – including the launch of a special AI division at NJII. But, he said that using the technology just to keep pace with others is the dumbest thing he has ever heard.

Also in this issue of NJBIZ:

Asbury Park Brewery has joined the ranks of breweries that are tapping into AI. Click here to read what’s brewing.

Sethi jumped in and noted that simply throwing AI into the mix does not automatically equate to good results. “Has that AI been evaluated? Similar to all things in life – whether it be food or software or applications that we buy and phones that we purchase – there’s good software and there’s bad software,” said Sethi. “AI is a software, and it needs to be evaluated – and the right software has to be used.”

He added that it is the responsibility of society and us as consumers to have the right gates in place – and make the right selection process and use AI at the right place and right time. “So, I think it’s important to realize and appreciate that not all AI is built equal,” Sethi explained. “There’s going to be bad AI out there and you need to have the proper governance in place to select the right way.”

Eagan noted there are also gaps in legislation because there are no federal laws pertaining to AI. Instead it’s instead left to states. “It is a patchwork of different standards in the states that even have AI laws,” said Eagan. “And most don’t.”

Are students ready?

Kanige then turned to Gulya, who had noted earlier on in the discussion that his students were heavy users of AI, and asked whether the students were prepared for how this will affect them as they enter the world of work. He cited the example of the evolution of how resumes were evaluated over the years as technology advanced.

“From your experience, are they ready for what’s going to be facing them out there?” Kanige asked.

“In general, I would say that they are aware of a lot of what is happening,” said Gulya. “They’re actually astoundingly aware, especially for students who are close to graduation. They’re not so sure they’re prepared for it. There’s a lot of worry.”

Replay: Artificial Intelligence Panel Discussion

Click through to register to watch the full panel discussion!

He said that he has had some students who are excited about AI and use it regularly; some who were terrified by it; some worried about it being a replacement for humans; and some in the middle.

“So, there’s a huge range with my students,” said Gulya. “And I would say that now that this technology has been around a little bit longer – and students have played with it, and built stuff out of it, that has started to shift. I do think my students are becoming more and more optimistic. And maybe these are the students that I’m just interacting with – but they do have models now for learning how to use AI as a part of their work and try to figure out how to be more productive; and how to think through things.”

“Jason and I work in an industry that moves about 10 miles an hour in terms of change,” said Edmondson. “The law is always three years behind – Brigette, it’s no offense to you. It’s just law – it’s got to catch up. And Samir works at 3,000 miles an hour because he’s in health care. So, he lives in a world where we have yet to get to. Jason and I – we’re an oil tanker trying to make a U-turn in the middle of the ocean, leaking oil. And Brigette’s still on the shore. We have really different perspectives here. And that’s why this is such an interesting panel.”

Understanding the tech

Edmondson then echoed Gulya on another common theme that was discussed during the panel – the need for digital and information literacy as this technology rapidly becomes a part of our lives.

“You now have to understand what a deepfake is. If you have a 2-year-old, they’re going to have to learn what a deepfake is,” said Edmondson. “That’s just part of the talk that you are going to have to have. There are just too many examples to go into right now about deepfakes and what’s been going on across the world. But there is a severe need for digital and information literacy. And I’ll leave it at that.”

If we really want to learn about this technology, a lot of it comes down to just recognizing as much as we can about what we don’t know.
– Jason Gulya, Berkeley College

Sethi offered a note of optimism, using the explosion of the internet as an example of a recent technological boom that did not end the world. “I’m going to date myself here. But, in 1995, we all got our modems. We were able to connect to AOL and we’re chatting. And then WebMD came out and we were able to search the web using our slow 56K modems,” said Sethi. “There was a lot of fear then as well about how we were going to use the internet. But eventually we figured this out.”

He said that the points on education and exposure and other concerns during the panel were valid, but he believes that we will get there. “This is not the first time this is happening – this is a different way,” said Sethi. “AI is a much more powerful and profound technology – for sure. But I think this is just a means to growth. What’s happening is – I wouldn’t call it normal, but I will say over time, people will figure out how to use AI; where to use AI; what to believe in it; and what to not believe in it – similar to what we did with the internet.”

“I think we’re sailing, but it’s a bumpy ride,” said Eagan, reiterating the contrast in AI employment laws in different states. “I’m on the rocky ship – I guess.”

Preventing lawsuits

Near the end of the discussion, Eagan said that from the legal perspective, the goal is to avoid lawsuits.

“What we like to do is prevent litigation. So do a little fact finding,” said Eagan. “Are you using AI right now? Talk to your vendors – because employers, you are responsible for the programs that you are buying, renting, using from your vendors. Are they using AI? Is that AI vetted? That’s number one – know what you’re using externally that can attributed to you that you might have liability and exposure for. And then internally – same process. What are your employees doing? Are we at the point where we need to have some policy parameters? And then if we do, we need some training and monitoring?

“And hopefully, if there is a lawsuit,” Eagan continued, “Those steps will be defenses that we were a responsible employer; that we did everything that we were supposed to do.”

Edmondson emphasized her need for people to understand that the world will continue to be VUCA – a Cold War-era military acronym that stands for volatile, uncertain, complex and ambiguous.

“The world can only get more VUCA-driven as AI becomes a bigger part of our lives,” said Edmondson. “When I do a lot of workshops, I talk about – what is your relationship with uncertainty? Do you try to control everything? A lot of what we talked about today was control – directly or indirectly. So, you need to understand your relationship with volatility, uncertainty, complexity and ambiguity of a world that is just marked by constant change and disruption.”

“I kind of just want to echo Michael,” said Gulya. “I think a lot of it’s going to come down to dealing with uncertainty; dealing with ambiguity; and doing what we can to remain curious. I think that’s going to be huge. And, in my mind, the anti-curiosity is when we stay siloed. So, for anyone trying to figure out this technology and just isn’t sure how to go forward. My number one tip is always – talk to people. Just talk to them. And this goes to every level. If you are an employer, talk to your employees. Create a culture where they can talk to you and actually converse with you and be honest with you – not hide things.

“If we really want to learn about this technology, a lot of it comes down to just recognizing as much as we can about what we don’t know,” Gulya continued. “And trying to figure out who to talk to; how to converse; how to find that information. That’s going to be huge going forward.”

Sethi stressed that AI is here to stay, but the magnitude of how it affects us will change and evolve. “The concept of AI – at a very high level – is amplifying human intelligence,” said Sethi. “It’s not a bad thing. I think that’s a pretty good thing. If we can make better decisions using AI or any of the technology out there, it’s a really good thing. While AI is going to be given to us – and I just mentioned we don’t have a choice for the most part and it’s going to help us. It’s also important to recognize that the choice about how it gets deployed is ours. It is in our hands. It is not AI – it is humans that are going to deploy AI and use it at the right place. So that’s something that we have in control.

“And organizations and people – we all need to make those choices wisely,” Sethi cautioned. “That is on us.”