How can you leverage AI tools in a secure and ethical way that keeps your data safe?
Artificial intelligence tools are evolving rapidly and becoming more integrated into our personal and professional lives. Organizations need to provide a safe and secure environment for their employees to learn, play, and engage with these tools so they feel empowered and equipped to use them to enhance their ways of working and performance.
The experts in this EXLEARN Talks episode discuss how AI tools continue to evolve and what that evolution requires of organizations, including steps organizations can take to ensure ethical adoption of AI tools, accuracy of AI outputs, and safeguarding of sensitive data. They provide insights on the following questions:
-
- 2:22: What will AI disrupt in the next 5-10 years?
- 8:25: What lessons from past revolutions can be applied to AI today?
- 11:36: If AI tools are simpler than they seem now, how will they improve in the future?
- 19:50: How do you define an AI “agent”?
- 27:09: How will more reliable agents support our future work?
- 34:55: How can we ensure the accuracy of what AI produces in our day-to-day work?
- 42:23: How can organizations keep their data safe while using AI tools?
- 48:41: How can organizations adopt AI tools in an ethical way?
Highlights from the Panelists:
“Anytime I think about technology, I think about people. The only reason I’m excited about technology is because of what it can do for people and therein lies its inherit value. So it’s not about how cool it is, it’s how it services when it makes human contact and enhances human performance. I think about the gaps and disparity of experiences people have and what AI can do to close those gaps.”
– Emanuel Lewis, Technology Strategist at TiER1 Performance
“[Change] is scary. It’s easy to look back and appreciate how important change, evolution, and innovation is. [No one wants] to go back to pre-industrial revolution times. Innovation happens for a reason because things are better. [The evolution of AI] is scary because it’s hard to imagine what happens when a lot of our cognitive work is automated. A lot of people [don’t think] about how much of their work is laborious cognitive stuff. While these tools can’t do all of it…they get a lot of thinking done for us and a lot of data processing. I wonder, once we’re done with the awkward adolescence of all this, how many people would be willing to go back to where we are now before all of this has really set in.”
– Mike Lewis, Chief AI Architect at Synthoni
“If you’re thinking about privacy and security, one key element is where and what is the model that you’re using… Finding a securely hosted way to use these models is one piece of maintaining that privacy, and that’s the hosting infrastructure side…but then you have this second layer. Let’s assume that you have a private and secure model; you still have to deal with potential inaccuracies and harmful information that may come out of the model… So there’s the model hosting thing that you can take care of, but you also need to account for these things around the periphery of the model to deal with privacy, security, and output validation appropriately.”
– Daniel Whitenack, Founder of Prediction Guard
A Visual Capture of the Discussion:
Designed by Ramsey Ford
