♻️Ethical, Responsible and Trustworthy

What's the difference in these concepts?

What does it mean to be Ethical, Responsible and/or Trustworthy? And aren’t those all really just the same thing anyway? Is Responsible AI also ethical and trustworthy?

Yes, and no.

Here’s an early take at a process that we’re learning how to explain and teach:

1. Ethical. State your values and your principles clearly up front. The goal is to be specific and comprehensive enough in your principles that when you later encounter the really hard-to-solve problems, your principles can serve as guides vs. roadblocks.

2. Responsible. Once you have your principles, will you stick to them? Or do you make arbitrary decisions based on expedience or fear? Responsibility is about owning up to our own choices and not passing the buck. Each mistake is a learning experience, that should make at most once.

3. Trustworthy. Even if we’re Ethical and Responsible, we still need to evaluate how it went. Being Trustworthy means taking the feedback, good and bad, and factoring that back into the process. Did we get some principles wrong? Hopefully not, but let’s be honest and fix them. Did we leave some people out of the solution? Let’s own up and address it. Trust comes from customers seeing us take responsibility for making the most ethical choices and sticking to it, even when it's hard. It’s their window into our process, in reverse.

These three steps, collectively ERT, form a repeatable cycle from early planning to development, go-to-market, and evaluation from the field, feeding back to next steps and improvements. The earlier in the cycle we get it right, the better the wheel spins. But reality always wins.