Kevin Szczepanski, Data Security & Technology Practice Area co-chair, participated in Buffalo Business First Table of Experts panel discussion on the growing impact of artificial intelligence across industries. The conversation explored AI’s potential to increase efficiency while also raising legal, operational, and cybersecurity considerations. The resulting article
Kevin emphasized that organizations must approach AI thoughtfully as they adopt the technology. “Some industries are full steam ahead and damn the consequences. Other industries are at the other extreme, that they can’t do this because there are too many bad things that could happen,” he said. “We need to strike the right balance. AI is the new semiconductor. It’s the iPhone. It may even be a more significant revolutionary technology.”
He noted that while AI tools are improving rapidly, they are still imperfect. “It’s not yet perfect,” Kevin said, adding that despite current limitations, the technology already represents a significant advancement compared to what existed previously.
Kevin also highlighted the growing legal and operational risks associated with AI–driven infrastructure, including the complex relationships among data center owners, operators, and technology providers. “You’re talking about a business that could be down for several days or weeks, so entrusting our hardware to a third party when it comes to AI is going to increase our risks,” he said.
In the legal profession, Kevin explained that AI can significantly improve efficiency but still requires human oversight. “Yes, there is a tremendous time savings because AI did for me in 10 minutes what might have taken me 10 days to do,” he said. “But I still added my few hours or days’ worth of human oversight. I’m still getting the right answer and still doing it cost effectively for my clients, but I’m not citing fake law.”
Kevin encouraged organizations to experiment with AI while implementing safeguards and clear strategies for responsible use. “Anyone reading this piece, if you haven’t experimented using AI, you should,” he said. “But the rush to AI shouldn’t devolve into, ‘Well, I have to do it, so let’s buy something and start rolling.’ It has to be the result of a strategic plan.”
He also stressed the importance of protecting sensitive information when using AI platforms. “When you put sensitive data into an open model you have forfeited your right to protection of that information,” Kevin said.
Looking ahead, Kevin predicted that organizations will quickly move from experimentation to more effective implementation. “We’re at that natural place where everybody was afraid to use it, now people are using it and the next step is they are using it correctly and getting results,” he said. “My guess is within 12 to 18 months it will pass through the next step when we are getting better results.”
Buffalo Business First subscribers can read the full article here.