Discover practical steps for implementing AI governance immediately. We explore mapping your AI footprint, creating digestible guides, vetting vendors and ensuring transparent disclosure for a cohesive organizational strategy.
These tips are part of the session "AI Governance and Responsible Use," where our speaker, Andrea Gils Monzón, CEO and co-founder at Shiftmakers Agency and member of the PRSA National Board of Directors, helps PR professionals learn and understand how to approach AI ethically for their organizations.
This session was part of the PRNEWS PRO Online Training Workshop: "The AI Shift: Practical Strategies for PR Leaders."
Watch the full session at this link.
Full transcript:
[ANDREA GILS MONZON]
Now, what does that mean in terms of governance, um, and, and kind of what are the things that you can do right now? And, and you wanna start practically, you wanna start lean, um, and you wanna make sure that you're aligned to y- your workflows. first of all, you ha- you need a baseline. So, you want to map your current AI footprint. So think about, where is AI already showing up in, in how your team works, even in the smallest ways, because understanding your baseline is going to help you set very informed guardrails.
You want to create guidance that is digestible, that is easy to understand, that people are not going to roll their eyes once you give it to them. So think about a one-page responsibly AI use guide, um, that your team can print out, have it, you know, hanging on the wall, um, that tells them how to prompt, when to disclose, um, what should always be re- reviewed by a human and at what point, right?
Things like that. Another thing that you want to do, and I believe there's a session about this, um, in the, in the AI Shift sessions with PRNEWS here, is how to vet your vendors. So, you want to review how they manage your data. Before AI or, or, or, or now, you, you still wanna understand how they use your data and manage it, right?
And how they use that to train their models, and if they say, "Oh, uh, we use, we use AI," ask a little more about what kind of AI. Is there a, is there a wrapper over, over a specific LLM? Things like that. Um, you want to understand how they handle sensitive and proprietary information, especially because cybersecurity issues are going to continue to be, uh, a big problem that we're going to be dealing with. And you want to, if you see the news that XYZ Company was hit with a cyberattack, you want to immediately be able to know who was impacted and how is that impacting your entire operations.
I talked about disclosure of language, so m- being very transparent, uh, in, in your ops and in your content strategy. So, whatever you agree on, you don't have to say necessarily that every post was, was, uh, written with AI, but you wanna make sure that you're consistent with it and that everyone is using the same language. You wanna think about your contracts with vendors, with, um, with your contractors, with your, your hires, um, all kinds of things that you wanna make sure that there's a con- that standard language of how they use AI or they shouldn't use AI, things like that.
And ultimately, all of these things, you want to be doing them in a way that is aligned with the entire organization, and that means, yes, partnering with HR, partnering with IT and legal, but also with other, uh, major teams in your organization, because as I said earlier, you want to have a diverse set of perspectives when you're trying to problem solve for this and build a, a strong governance across your organization.
It has to be inclusive, it has to be consistent, and it has to be enforceable. So, you need to be able to ask those different team leads how this translates into their day to work, into the- their day-to-day work.