Your Personal AI Policy

**Bridget Bergen** (0:10)
It's The Brian Lehrer Show on WNYC. I'm Bridget Bergen, filling in for Brian.
Do you have a personal AI policy? If you use AI chatbots in your professional and or personal life, maybe you should. That's one of the big lessons you would take away from a recent mini controversy involving one of the most widely known and well-regarded science communicators online. Hank Green has more than three million YouTube subscribers. If that name sounds familiar, it might be because Hank Green and his brother, John Green, hosted a number of podcasts produced by WNYC studios between 2018 and 2020
Hank's made videos about space, climate change, and his own experience with cancer. But recently, many of Green's fans were disappointed to discover that he had used AI for what he described as research purposes. My guest, Marina Bolotnikova, who reports for Vox's Future Perfect section, has a story about this. It's called Everybody Needs a Personal AI Policy, Just Ask Hank Green. Marina, welcome to WNYC.

**Marina Bolotnikova** (1:16)
Hey, Bridget. Delighted to talk to you. Thank you.

**Bridget Bergen** (1:19)
Well, it's great to have you here. And listeners, how do you use AI? And more importantly, what is your personal AI policy? Do you have any self-imposed ground rules for your use of chatbots like chat, GPT, Claw, Gemini, or any of the other ones? Give us a call at 212-433-WNYC. That's 212-433-9692.
You can help our guest, Marina Bolotnikova from Vox's Future Perfect section, Report Her Story. Everybody Needs a Personal AI Policy, Just Ask Hank Green. So Marina, I mentioned who Hank Green is, but why was his use of AI such a big controversy?

**Marina Bolotnikova** (2:02)
Yeah. So Hank Green, as you mentioned, he is huge. He's one of the OG YouTube stars. He's been on the platform for 20 years. And I think he's generally known as someone who's voraciously creative and intellectual, and is known as being a person of integrity. I would say. And so at the end of last month, some of his followers suspected that a line from one of his videos was AI written. And the line was simply, I appreciate the pushback.
I think I'm pretty good at recognizing AI generated writing to the degree that anyone can these days. And I don't think that line sounded necessarily AI generated. It didn't sound like a smoking gun to me. But nevertheless, this allegation carried on the winds of the internet. People started talking about it. Hank then responded to the allegation on Twitter and Reddit and said that the line was not AI written, but rather than leave it at that and say, I did nothing wrong. This is totally ridiculous. He did something else, which was to, you know, he made a pretty protracted statement acknowledging that he does use AI as a research aid, that he thinks he's been relying on it too heavily for that purpose. And he wrote that although the line in question was not AI written, he thinks that his over-reliance on AI for research gave the whole video a quote AI feel. And he apologized for that. And he said that he wrote that AI gives him access to tons of papers really quickly, but that that actually has been to the detriment of his work because it has not given him the freedom to find all of his own ways into and around a topic. And he said that his relationship with AI had become, quote, not healthy for me or good for the world.
So immediately after he made that statement, there was this vocal reaction on social media. There was a camp of people who were totally outraged that he would use AI at all for research for any purposes. And then there was a backlash to that backlash. There were people saying, I can't believe Hank Green is congratulating himself and prostrating to this anti-AI mob. When he did nothing wrong, he merely used AI for research. That's a totally legitimate use of the technology. It's like apologizing for using the printing press.

**Bridget Bergen** (4:48)
Well, let's listen to a short clip of Hank Green talking about this. Here is YouTuber and author Hank Green in a recent video acknowledging what he believes went wrong.

**Hank Green** (4:58)
I've become too reliant on them for advice, for editing and for sourcing. So I'm creating an AI policy for myself.

**Bridget Bergen** (5:05)
So just briefly, talk a little bit about what his AI policy covers.

**Marina Bolotnikova** (5:10)
Yeah.
So he, yeah, he made that, he released that video addressing the situation almost a week ago today and shared a new AI policy for himself that I think is a pretty good one. So he's not going to, you know, he said no portion of any script for a video will be written, edited or outlined by AI, that a thesis for a video will always originate with a human. No music or image in a video will be AI generated, and that AI outputs aren't trusted as an accurate source, that you have to, you know, check what sources the AI is using and go to the original source to verify them. And I think these are, I think these are good ideas for anyone trying to set their own guardrails. And I think that, you know, both of the camps that we're reacting to, to Hank's statement, you know, the people who hate him for using AI and the people who are angry that he would apologize for AI use, I think both of them are kind of wrong and misguided and we're misunderstanding this complex and nuanced issue, which is that AI is a really useful tool, using it to aid, but, you know, using it for your work, even if it doesn't rise to the level of full on, like having your doing your writing for you, it can still militate against having, you know, against thinking for yourself, it can start to take over your creative process and just kind of feel like it's cooking your brain. And I really recognize myself in Hank's statement and some of the, you know, I was glad that he sort of, he put this out there and is thinking through it in public. Some of the things that I've been feeling, just using AI as a research aid in my work too, really resonated with me.