Robyn AndersonCEO, Jam Ai · AI Implementation & Business Development Leader

This story explains how AI is probably more trustworthy and safe to use than hiring an employee that

11 May 2026 · 1 min

Watch on YouTube

0:59

Robyn Anderson asked Claude to write an article she'd promised an editor she wouldn't use AI for. Claude refused, detecting the promise in the transcript and insisting on honoring it. Anderson argues this shows AI has built-in ethical constraints that make it more trustworthy than many human employees.

Chapters

So I had a call with this woman and she wants me to write an article for her magazine. And I said, yeah, sure, of course. And at the end she told a joke and she said, so you're not using AI for this, right? And I'm like, yeah, of course not. We hang up. Claude gets my call transcript and then it's reading through it and I say, hey, I need you to write the article. Here's the quote, here's the whatever. And it said, hey, I'm sorry, Jam, I can't write this for you. You told her at the end that you would not use AI. and I want you to hold yourself accountable and make sure you do that. And I had to explain that we were just kidding, and it went back and forth making sure that we were kidding, because it actually holds me to a standard that I wouldn't even hold myself to. So for those of you who are worried about Claude stealing your information, doing malicious things, just know that Anthropic is programming this to be a virtuous platform, and you don't have to worry about it. It's more trustworthy than most employees and most people that you know.