Panelists discuss AI governance at CMU, K&L Gates conference on computational ethics – The Business Journals
Listen to this article 2 min
K&L Gates previously donated $10 million to CMU to study ethical concerns surrounding AI.
STORY HIGHLIGHTS
In 2016, law firm K&L Gates made a $10 million gift to Carnegie Mellon University to study ethical issues posed by artificial intelligence. Throughout the near decade that’s followed, the technology has radically evolved, becoming not just mainstream but a revolution. Since then, the two have partnered to host conferences on the ethics and governance of the technology at an enterprise level, and this year’s includes some notable names.
Keynote speakers for the conference are DJ Patil, a computer scientist who previously served as Chief Data Scientist for the United States Office of Science and Technology Policy under Barack Obama and currently works in Venture Capital, and Natasha Crampton, a vice president at Microsoft and the company’s Chief Responsible AI Officer. Additional panelists include Abridge CEO Shiv Rao, whose company recently closed a $250 million Series D round, and CMU Machine Learning Department Director Zico Kolter, who sits on the board of ChatGPT maker Open AI. During a panel on AI governance, Kolter acknowledged that the rapid pace of innovation makes developing guidance a moving target.
“One big concern I have is that oftentimes we’re not speaking the same language, because we have very different notions of the capabilities that are either in the current AI systems or are coming soon to AI systems,” Kolter said. “(An AI that) could do anything that a person can do over the course of a year, basically that opens up kind of a mind-boggling array of possibilities in terms of the risks.”
Fellow panelist Carol Smith, lead of CMU’s AI Trust Lab supporting the Department of Defense, noted that there has to be balance between setting tangible, hard stop guidelines without making it “painful.”
“My priorities are always to make sure that we’re making a useful, balanced system that’s evidenced based and that people understand the systems they’re working with and I don’t know we can say that right now,” Smith said. “My work is with people doing high risk activities, people who are dealing with higher stakes than e-commerce and other challenges and so I have a different approach, I think, than my colleagues. But I also worry that we don’t want to make it such a big stretch for people to follow guidance that they ignore it or they don’t do it because it’s too painful. So there’s a balance.”
Beyond balancing enterprise and governmental deployment, panelists discussed something else that requires balance — the delicate balance between speculative hype promising AI capabilities more akin to a science fiction novel than real life and the possibility that that science fiction novel could become real life.
“I’ll be very straightforward about it, a lot of what we consider in the safety and security committee at Open AI is what we kind of refer to as catastrophic risks, you know, do we think that AI systems are capable of producing bio weapons that could easily wipe out the world?” Kolter said. “Do we think that AI systems are capable of producing bio weapons that could easily wipe out the world? Do we think they are capable of producing cyber attacks that could essentially produce zero day exploits for most deployed software out there? Then to me, still a bit far fetched but less and less as time goes on, do we worry about sort of true loss of control, like scenarios where we have an AI system that we not longer have the ability to control in the sense that we do right now, and these are questions that still feel a bit like science fiction to many of us.”
© 2025 American City Business Journals. All rights reserved. Use of and/or registration on any portion of this site constitutes acceptance of our User Agreement (updated August 13, 2024) and Privacy Policy (updated December 17, 2024). The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of American City Business Journals.
source
This article was autogenerated from a news feed from CDO TIMES selected high quality news and research sources. There was no editorial review conducted beyond that by CDO TIMES staff. Need help with any of the topics in our articles? Schedule your free CDO TIMES Tech Navigator call today to stay ahead of the curve and gain insider advantages to propel your business!
K&L Gates previously donated $10 million to CMU to study ethical concerns surrounding AI.
STORY HIGHLIGHTS
In 2016, law firm K&L Gates made a $10 million gift to Carnegie Mellon University to study ethical issues posed by artificial intelligence. Throughout the near decade that’s followed, the technology has radically evolved, becoming not just mainstream but a revolution. Since then, the two have partnered to host conferences on the ethics and governance of the technology at an enterprise level, and this year’s includes some notable names.
Keynote speakers for the conference are DJ Patil, a computer scientist who previously served as Chief Data Scientist for the United States Office of Science and Technology Policy under Barack Obama and currently works in Venture Capital, and Natasha Crampton, a vice president at Microsoft and the company’s Chief Responsible AI Officer. Additional panelists include Abridge CEO Shiv Rao, whose company recently closed a $250 million Series D round, and CMU Machine Learning Department Director Zico Kolter, who sits on the board of ChatGPT maker Open AI. During a panel on AI governance, Kolter acknowledged that the rapid pace of innovation makes developing guidance a moving target.
“One big concern I have is that oftentimes we’re not speaking the same language, because we have very different notions of the capabilities that are either in the current AI systems or are coming soon to AI systems,” Kolter said. “(An AI that) could do anything that a person can do over the course of a year, basically that opens up kind of a mind-boggling array of possibilities in terms of the risks.”
Fellow panelist Carol Smith, lead of CMU’s AI Trust Lab supporting the Department of Defense, noted that there has to be balance between setting tangible, hard stop guidelines without making it “painful.”
“My priorities are always to make sure that we’re making a useful, balanced system that’s evidenced based and that people understand the systems they’re working with and I don’t know we can say that right now,” Smith said. “My work is with people doing high risk activities, people who are dealing with higher stakes than e-commerce and other challenges and so I have a different approach, I think, than my colleagues. But I also worry that we don’t want to make it such a big stretch for people to follow guidance that they ignore it or they don’t do it because it’s too painful. So there’s a balance.”
Beyond balancing enterprise and governmental deployment, panelists discussed something else that requires balance — the delicate balance between speculative hype promising AI capabilities more akin to a science fiction novel than real life and the possibility that that science fiction novel could become real life.
“I’ll be very straightforward about it, a lot of what we consider in the safety and security committee at Open AI is what we kind of refer to as catastrophic risks, you know, do we think that AI systems are capable of producing bio weapons that could easily wipe out the world?” Kolter said. “Do we think that AI systems are capable of producing bio weapons that could easily wipe out the world? Do we think they are capable of producing cyber attacks that could essentially produce zero day exploits for most deployed software out there? Then to me, still a bit far fetched but less and less as time goes on, do we worry about sort of true loss of control, like scenarios where we have an AI system that we not longer have the ability to control in the sense that we do right now, and these are questions that still feel a bit like science fiction to many of us.”
© 2025 American City Business Journals. All rights reserved. Use of and/or registration on any portion of this site constitutes acceptance of our User Agreement (updated August 13, 2024) and Privacy Policy (updated December 17, 2024). The material on this site may not be reproduced, distributed, transmitted, cached or otherwise used, except with the prior written permission of American City Business Journals.
source
This article was autogenerated from a news feed from CDO TIMES selected high quality news and research sources. There was no editorial review conducted beyond that by CDO TIMES staff. Need help with any of the topics in our articles? Schedule your free CDO TIMES Tech Navigator call today to stay ahead of the curve and gain insider advantages to propel your business!
