Thinking in clues

0:00 Hopefully, by the end of today I'll have shown you ways that you can you might know some of them, some of them you might know some of them you might not know. Uh, but by the end you should be able to look at a text and say, "I'm pretty sure that was written by AI." Or, "I'm not sure at all uh that that I'm pretty sure, sorry, that that was written by a human." So, here's what you should look out for.

0:24 First of all, who am I? And why does why do I matter that I'm teaching you any of this today? Um, I am James. Some of you may know me, you may have had lessons with me. Uh, I am James from Easy English with James. I have a YouTube channel, if you want to check it out, uh, over on YouTube. Uh, I've been an instructor for years and years now, uh, in English, but also at English for IT.

0:47 I'm also somebody who, like you, actually learned languages, German, Spanish, a little bit of French, a little bit of Vietnamese, as well. So, I have been the confused English student, as well. And I also, um, over years and years and years, basically self-taught myself to to to use various different coding languages to build lessons, courses, uh, and various different things. I'm from the UK, as you probably already guessed. Uh, so this accent is real, it's my real accent that I'm using.

1:10 And I've got loads to go through today, so I'm going to basically start straight away. The first thing that we should know is that nothing I tell you today is 100% provable or factual. Normally, you want facts, right? But actually, today we're thinking in clues, not proof. If you see one clue, I think this is written by AI. I think this is almost certainly written by AI. Okay? So, let's have a quick look. Where are we?

2:01 Oh, what's this? Ah, I think my microphone must have been having a bit of a problem there, having an issue. So, I hope you've you've not missed the first little bit there, but you can probably see what I've scrolled through. If anybody couldn't hear me, uh if there were any issues, this is me. Uh I'm going to get started straight away cuz I don't want I want to make sure we get through all the stuff today.

2:23 Essentially, I was saying that we think in clues. So, look at this carefully. When you see five of the things together that we're going to look at today, it probably means it was written by AI. Uh if you just see one in a text, it's probably not true. What we should remember at all times is that AI is trained to sound clear, polite, and safe unless you tell it otherwise. That means that if you don't tell it, it will always try to sound clear, polite, and safe.

2:48 When everything is trained to be clear, polite, and safe, this in itself is like an accent. If I'm always trying to be clear to you and polite to you and safe with you, that actually sounds like an accent in itself. So, when everything is smooth, this smoothment smoothness itself becomes an accent. Hopefully, by the end of today, you will read emails or Facebook posts or Twit tweets on X uh like a detective and you'll be able to detect some of that accent that I'm talking about here.

3:10 The first thing that I will tell you about and that we need to look at is, of