Earlier today my friend sent me a clip of Alan Watts, and said it was one of his favorite by him. If you don’t know who Aan Watts is, he’s one of the main figures in the exportation of Asian philosophies to the rest of the world (Buddhism, Taoism, Hinduism), and simplifying some of the ideas for unfamiliar audiences.
He and J. D. Krishnamurti were a big influence on me in my late teens and early twenties and helped me find my own faith and undersranding of the world outside of organized religion.
Both of them speak about life in a way that doesn’t conflict with religion but doesn’t require it either, its more about self reflection and ones relationship with their life and mind than anything else but its too abstract to fit into academic fields other than philosophy, so they are called both philosohers and spiritual teachers, depending who you ask.
I haven’t listened to Watts in years, and I was a little hesitants because rhe video was 20 minutes and I’ve heard so much Watts, its impossible that I would hear somethinf completely new or life changing by now.
As I listened I was glad i did because so much of it was relevant to things I’ve been thinking about recently. It was a great lecture but I felt there were a few things that didn’t resonate about it but nothing that sounded completely un-Watts.
As I got to the end I hear one line and one line alone that stuck out like a sore thumb. It sounded oike a youtube influencer more than Alan Watts:
“So there you have it, five things you should never share with anyone”.
That’s influencer speak.
I asked something my friend “Are you sure that’s really Alan Watts”? The idea hadn’t even crossed his mind that it could be AI generated. We both know that is possible these days but we fet strongly that we were listening to Alan Watts because it sounded so much like Alan Watts.
“It’s him.”
“How do you know?”
“I have a good ear”
WHAT!?!
This is where I tend to be more skeptical than the average person.
Admittedly , even if it was fake it was a pretty decent lecture, I don’t feel like I completey wasted my time.
I started to think of how easy it would be to create a nearly true- to-life Alan Watts lecture.
1 Feed GPT or Grok a Watts lecture and ask it to “rewrite as a list with a catchy title”.
2 Get some other AI (not sure which is popular) to read it in Alan Watts voice
3 upload it with a catchy thumbnail and tags
Literally anyone with the will under the age of 65 could do that.
We’ve watched hundreds of videos like this over the years, so apparently neither of us bothered to read the title, only the thumbnail, but the title had it right there :
“Alan Watts Influenced”
It was his voice, perfect as far as I could tell (I have a good ear too 😝). It had his face in the thumbnail and his name in the title. It sounded like what he might say. So we didn’t even bother to read the long wordy title to the end.
In the end I think the reason I was fooled was the socal proof. He sent it to me as if it was really Alan Watts so I didn’t question. When I search things myself I am always more careful, but if a friend acts as if something is a bird, Im going to assume its a bird and not a fake bird, until I have a reason to doubt it.
In a 20 minute video there was only one line that had me doubting.
I’m still a little shocked at how easily fooled I was, and I am the skeptical type. I don’t rely on “I have a good ear”, i mean i think I do, but that’s not enough for my skepticism to disappear when ai start questioning.
This is such a strange world we are living in and I don’t think any of us have an answer for how to navigate it.
I don’t think things like this should he illegal, its kind of cool to hear AI rewrite Alan Watts but remain faithful about it, but I don’t think people should be profiting off others work or passing AI off as not AI.
For now they can try and label AI content but its wuickly reaching the point where there will be no way to detect the information.
For now the best things you can do are
1 Read the title carefully
2 listen for uncharacteristic statements
3 Do a google search
4 Look at the date.
The number one most reliable thing is the date. If it’s from 15 years ago, there is no chance it’s AI generated. But what happens when the majority of content comes from after 2020? That metric will no longer matter. What happens when the AI is indistinguishable in every aspect? What happens when theres so much unreliable information that GPT and Google can’t tell you if it’s something the person actually said.
In the end, we can only trust our own senses and our heart. Unlike my friend I don’t rely on my heart to trll me if it’s really Alan Watts or not, I use it to tell me if the information is worrh my attention and useful. Wherher or not its Alan Watts will become more and more of a mystery, I guess similar to the words of famous figures before video footage.
Here is the video btw, check the comments, no one mentions AI…
Edit: soon after found myself watching another video that was even more convincing but this time i was paying closer attention. Completely different topic, but if you listen carefully you realize this video uses a lot of sentence patterns which are not natural in every day speech and sound AI generated, though its a little subtle:
The only real troubling thing to me is how no one in the comments seems to suspect its AI, those comments may be deleted by the uploader or maybe everyone is already THAT fooled by AI