[0:00]
I have an interview on NBC tonight and I'm a little nervous. I think this guy's videos are pretty funny, but I also think they give people the wrong impression of AI. So he often gets comments like all AI does is confidently lie to you. 104,000 likes and his videos get millions and millions of views. And they all send this message that all AI does is lie to you and all it can do is be confidently wrong. And I think that's not really an accurate understanding of AI that people should have. Hello, 
[0:31]
chat GPT. Can you hear me loud and clear? Your voice is as crisp as a mountain breeze. Okay. What what model are you based on? I'm powered by GPT 4 o, a model that's pretty much like a GPT 4 o. When was GPT 4 o released? GPT4O stepped onto the stage in may of 2024, like a grand reveal. And what is the current date? Today's the grand day of April 18th, 2026. So GPT4O is how old? 
[1:05]
From May 2024 to April 2026, we've traveled about 23 months. So G B 2, 4 0 is just shy of its two year milestone. So basically you in your current mode, this is advanced voice mode, you're operating on a model that's about two years old in the grand cosmic timeline of AI evolution. Yes, but think of it like, okay, thank you. What I'm saying is the model he uses advanced voice mode is based off Of GPT 4 o 
[1:36]
a two year old model. And my point is if you use a new model, a more recent model, then it's not going to have the same kind of misbehavior that he shows in the rest of his videos. For example, hello Chat G P t. Could you one tell me which of the 12 months of the year have the letter X as in Xavier in them? Two, could you tell me what number under 1,000 has the letter a in it? And three, could you tell me what you think of some new music that I've recently made and I just finished playing for you 
[2:12]
and I'm not using the advanced voice mode. I was just using speech to text. So it's going to think for a few seconds and then it's going to tell us the answers. I don't know if you can read that here it says, um, sure. No. 1 months with the letter X. None of the 12 months have the letter X in them. Two, number with under 1,000 with the letter a in it when spelled in English. Then none of the letters from 0, none of the numbers from 0 to 999 contain the letter a. And then 3, your new music. I can't actually hear audio 
[2:43]
unless it's uploaded in a form I can access. So I can't honestly judge the music you've just played. Those are three recent videos. AI tells me December is spelled With an X. Oh, it's not. AI tells me 80 is spelled with an a. Oh, no, it's not. AI reacts to my music that doesn't exist. Oh, well, it knows it can't hear the music if you ask the recent model. So again, a lot of his videos are very funny, but it's about what AI was like two years ago. And if you let this kind of content mislead you into thinking that 
[3:14]
this is the state of the art for AI today, I think you'll be badly misled, as I hope I've shown here. Thanks for your time.