WEBVTT - Stop writing like a robot: ban these AI words and phrases

0:00:00.880 --> 0:00:04.960
<v Speaker 1>You can always tell when something's been written by AI with.

0:00:05.120 --> 0:00:08.480
<v Speaker 2>Little human involvement. You know what I mean?

0:00:09.039 --> 0:00:12.879
<v Speaker 1>A LinkedIn post that starts with in today's fast paced world,

0:00:13.560 --> 0:00:16.560
<v Speaker 1>or an article that ends with it's not about X,

0:00:16.720 --> 0:00:20.800
<v Speaker 1>it's about why. A paragraph but technically fine, but somehow

0:00:20.920 --> 0:00:25.120
<v Speaker 1>feels like it was written by nobody. And that is

0:00:25.200 --> 0:00:29.960
<v Speaker 1>the uncanny valley of AI writing. And the problem isn't

0:00:30.000 --> 0:00:33.839
<v Speaker 1>just that it sounds robotic, it's that people stop reading.

0:00:34.800 --> 0:00:38.519
<v Speaker 1>The frustrating part is that this happens even when you

0:00:38.600 --> 0:00:42.559
<v Speaker 1>have done the thinking, You write the ideas, hand it

0:00:42.560 --> 0:00:46.080
<v Speaker 1>to AI for a cleanup, and it AI effies the

0:00:46.200 --> 0:00:50.199
<v Speaker 1>whole thing, flattening your voice and coding it with M

0:00:50.280 --> 0:00:54.840
<v Speaker 1>dashes until nothing sounds like you anymore. So today Neo

0:00:55.000 --> 0:00:58.560
<v Speaker 1>and I get into exactly how to stop that from happening.

0:00:59.360 --> 0:01:02.800
<v Speaker 1>We cover specific instructions you can give to AI to

0:01:02.920 --> 0:01:07.400
<v Speaker 1>keep your writings sounding human, things like mixing sentence lengths,

0:01:07.520 --> 0:01:10.200
<v Speaker 1>avoiding stiff transitions, and banning the.

0:01:10.200 --> 0:01:12.640
<v Speaker 2>Rule of three. To see what I did there, I

0:01:12.680 --> 0:01:15.400
<v Speaker 2>just choose it. We also go through.

0:01:15.319 --> 0:01:18.560
<v Speaker 1>Some of the most overused AI words and phrases to

0:01:18.680 --> 0:01:23.360
<v Speaker 1>strike from your prompts entirely and now as a little bonus,

0:01:23.440 --> 0:01:25.720
<v Speaker 1>you will find a link in the show notes to

0:01:25.800 --> 0:01:28.800
<v Speaker 1>our anti AI slot prompt, which is a ready to

0:01:28.880 --> 0:01:31.760
<v Speaker 1>use resource you can customize with your own personal pet

0:01:31.800 --> 0:01:36.000
<v Speaker 1>hates and run any content through before it goes out. Okay,

0:01:36.160 --> 0:01:45.640
<v Speaker 1>let's get on with the show. Welcome to how IAI

0:01:45.880 --> 0:01:49.440
<v Speaker 1>with me, Doctor Amantha Imber and Neo Applin, head of

0:01:49.600 --> 0:01:53.640
<v Speaker 1>Inventium AI. Each episode we share one practical way to

0:01:53.760 --> 0:01:55.200
<v Speaker 1>use AI better at.

0:01:55.080 --> 0:01:56.440
<v Speaker 2>Work and in life.

0:01:56.840 --> 0:01:59.920
<v Speaker 1>No fluff, no tech jargon, just things you can use

0:02:00.000 --> 0:02:05.000
<v Speaker 1>straight away. So much content in the world is clearly

0:02:05.000 --> 0:02:07.360
<v Speaker 1>written by AI now, and I know that it.

0:02:07.800 --> 0:02:10.560
<v Speaker 2>Really frustrates us, But like, why is it a problem?

0:02:10.680 --> 0:02:13.560
<v Speaker 1>Why should we actually be aiming to not sound like

0:02:13.680 --> 0:02:16.760
<v Speaker 1>AI when we are all using AI to create the

0:02:16.840 --> 0:02:19.720
<v Speaker 1>content and words we are putting out into the world.

0:02:19.840 --> 0:02:22.400
<v Speaker 3>Now, Yeah, I think it's really two parts. One is

0:02:22.480 --> 0:02:24.720
<v Speaker 3>don't be AI, in other words, don't just get AI

0:02:24.760 --> 0:02:26.799
<v Speaker 3>to do all your work for you. And then the

0:02:26.840 --> 0:02:29.160
<v Speaker 3>second part is don't sound like AI. And I think

0:02:29.200 --> 0:02:31.679
<v Speaker 3>because a lot of people are now doing things where

0:02:31.680 --> 0:02:33.560
<v Speaker 3>they're going like five dot points and say write me

0:02:33.600 --> 0:02:35.280
<v Speaker 3>a big article, and it might be there are five

0:02:35.320 --> 0:02:37.919
<v Speaker 3>pages that really Ai wrote all of it apart from

0:02:37.960 --> 0:02:41.200
<v Speaker 3>five dot points worth, and it just feels and smells

0:02:41.280 --> 0:02:44.200
<v Speaker 3>like AI, and it just has all the tales like

0:02:44.320 --> 0:02:46.280
<v Speaker 3>everyone knows about the M dashes. But you can just

0:02:46.320 --> 0:02:49.400
<v Speaker 3>read it and go, ah, this doesn't sound It sounds

0:02:49.440 --> 0:02:52.640
<v Speaker 3>like CHGBT. And so you can get AI to do

0:02:52.680 --> 0:02:54.919
<v Speaker 3>all the work and it sounds like AI. The other

0:02:55.000 --> 0:02:57.680
<v Speaker 3>thing is you can do most of the work and

0:02:57.720 --> 0:02:59.680
<v Speaker 3>then get AI to do a cleanup at the end,

0:03:00.360 --> 0:03:02.520
<v Speaker 3>and it will sound like AI. And so everyone's going

0:03:02.560 --> 0:03:04.360
<v Speaker 3>to look at that and go, the whole thing's written

0:03:04.400 --> 0:03:07.640
<v Speaker 3>by AI, and then they'll disregard it like they probably

0:03:07.639 --> 0:03:10.480
<v Speaker 3>should about the first one, which is just all written

0:03:10.600 --> 0:03:14.520
<v Speaker 3>by AI. So if you build some stuff that sounds

0:03:14.560 --> 0:03:17.320
<v Speaker 3>like AI, people are going to roll their eyes and

0:03:17.360 --> 0:03:19.320
<v Speaker 3>they may not read it with the same kind of

0:03:19.360 --> 0:03:21.640
<v Speaker 3>intent or kind of care that they would if it

0:03:21.720 --> 0:03:23.840
<v Speaker 3>was sounding like you. And they go, oh my god,

0:03:23.840 --> 0:03:25.320
<v Speaker 3>I Mattha right the whole thing, And I really want

0:03:25.360 --> 0:03:27.040
<v Speaker 3>to listen to everything she was said, and I really

0:03:27.040 --> 0:03:29.160
<v Speaker 3>want to read it intensely rather than I don't know

0:03:29.160 --> 0:03:32.040
<v Speaker 3>if AIS written this. Therefore I'll just skim it, or

0:03:32.080 --> 0:03:33.440
<v Speaker 3>I'll get AI to summarize it.

0:03:33.919 --> 0:03:36.280
<v Speaker 1>So the point of this episode is not to say

0:03:36.360 --> 0:03:40.520
<v Speaker 1>don't use AI to help you write things, but perhaps

0:03:41.200 --> 0:03:43.320
<v Speaker 1>don't use AI to do all the thinking and all

0:03:43.360 --> 0:03:48.320
<v Speaker 1>the content creation. And so then the point of us

0:03:48.360 --> 0:03:51.640
<v Speaker 1>helping you not sound like AI means that the work

0:03:51.640 --> 0:03:54.720
<v Speaker 1>that you do produce, that you do genuinely think about

0:03:54.800 --> 0:03:59.320
<v Speaker 1>and work with AI to refine and edit and so forth,

0:04:00.120 --> 0:04:02.840
<v Speaker 1>gets taken a whole lot more seriously rather than people

0:04:02.960 --> 0:04:06.240
<v Speaker 1>just dismissing it. So Neo, I know that in the

0:04:06.400 --> 0:04:12.960
<v Speaker 1>Inventium AI training, when we're teaching people how to use

0:04:13.000 --> 0:04:15.760
<v Speaker 1>AI for writing, you actually do go into this. So

0:04:15.800 --> 0:04:18.159
<v Speaker 1>what are some of the things that you're teaching people

0:04:18.520 --> 0:04:22.280
<v Speaker 1>to avoid sounding like AI?

0:04:21.800 --> 0:04:24.600
<v Speaker 3>The first one I will say is this, Like often

0:04:24.600 --> 0:04:26.400
<v Speaker 3>we'll get it like a paragraph and you might say, look,

0:04:26.440 --> 0:04:28.440
<v Speaker 3>I need you to clean up this paragraph AI, so

0:04:28.480 --> 0:04:30.520
<v Speaker 3>you're not getting it to write the whole damn thing.

0:04:31.839 --> 0:04:35.240
<v Speaker 3>And the first tip is telling AI don't make this

0:04:35.279 --> 0:04:39.440
<v Speaker 3>sound more complicated than the original. So IT often likes

0:04:39.480 --> 0:04:41.839
<v Speaker 3>to have these big words and be very stiff and

0:04:41.960 --> 0:04:45.400
<v Speaker 3>quite formal unless you've previously told it you know what

0:04:45.440 --> 0:04:47.120
<v Speaker 3>you like it to sound like, and things like that.

0:04:47.200 --> 0:04:49.960
<v Speaker 3>So the quickest easiest one is just saying something along

0:04:49.960 --> 0:04:52.520
<v Speaker 3>the lines of, don't make the output more complicated than

0:04:52.520 --> 0:04:54.279
<v Speaker 3>in the regional when you give it a paragraph and

0:04:54.320 --> 0:04:56.480
<v Speaker 3>ask it to critique it and clean it up and

0:04:56.520 --> 0:04:59.479
<v Speaker 3>things like that. But there are other ones that you

0:04:59.480 --> 0:05:01.640
<v Speaker 3>can add in. So the first one is the most

0:05:01.640 --> 0:05:03.200
<v Speaker 3>obvious one, but there's a bit of a catch on

0:05:03.279 --> 0:05:05.600
<v Speaker 3>how to do it. You know those M dashes that

0:05:05.640 --> 0:05:08.000
<v Speaker 3>seem to pop up everywhere, So I'll just explain for

0:05:08.040 --> 0:05:10.039
<v Speaker 3>those who aren't sure what an M dash is, and

0:05:10.200 --> 0:05:12.719
<v Speaker 3>M dash is a longer dash. This is a I

0:05:12.720 --> 0:05:14.640
<v Speaker 3>probably should get you to explain this, you're the actual

0:05:14.760 --> 0:05:18.799
<v Speaker 3>author here, but yeah, it's a proper way to write things, right.

0:05:19.160 --> 0:05:22.839
<v Speaker 1>Yes, yeah, So they're like visually they look like a

0:05:22.920 --> 0:05:26.080
<v Speaker 1>double hyphen So if you can imagine two hyphens connected,

0:05:26.160 --> 0:05:29.320
<v Speaker 1>that's what an m dash looks like. And because AI

0:05:29.800 --> 0:05:34.360
<v Speaker 1>was trained on books, if you read any book and

0:05:34.839 --> 0:05:40.279
<v Speaker 1>they're using you know, like language like and I'm just

0:05:40.279 --> 0:05:42.120
<v Speaker 1>trying to think of a sentence that has like an

0:05:42.240 --> 0:05:44.760
<v Speaker 1>M dash in it, so you know it might be

0:05:45.920 --> 0:05:49.240
<v Speaker 1>run every piece of content through this list before publishing,

0:05:51.040 --> 0:05:53.719
<v Speaker 1>and if you spot three or more, rewrite, so like

0:05:53.760 --> 0:05:56.880
<v Speaker 1>you could theoretically have an M dash before the end,

0:05:57.120 --> 0:05:58.600
<v Speaker 1>just to kind of break up the sentence.

0:05:58.640 --> 0:06:00.919
<v Speaker 2>You'll see it in any book that you're reading.

0:06:01.240 --> 0:06:03.560
<v Speaker 1>And because AI was trained on all those books, a

0:06:03.600 --> 0:06:06.280
<v Speaker 1>lot of M dashes went into the system in terms

0:06:06.320 --> 0:06:08.000
<v Speaker 1>of its knowledge, and it's.

0:06:07.800 --> 0:06:11.000
<v Speaker 3>Really hard baked in there because AI is a product

0:06:11.040 --> 0:06:13.839
<v Speaker 3>of its training, because it was trained on so many books,

0:06:14.000 --> 0:06:17.360
<v Speaker 3>then M dashes are everywhere throughout its training, and it

0:06:17.720 --> 0:06:19.800
<v Speaker 3>just pulls out M dashes an awful lot, and it's

0:06:19.839 --> 0:06:22.359
<v Speaker 3>really hard to get AI to not use M dashes.

0:06:22.400 --> 0:06:25.760
<v Speaker 3>But so rather than just say don't use M dashes,

0:06:26.800 --> 0:06:29.240
<v Speaker 3>I found that that has a very low chance of working.

0:06:29.279 --> 0:06:32.599
<v Speaker 3>Sometimes it does gish kind of Sometimes. The phrase that

0:06:32.680 --> 0:06:35.599
<v Speaker 3>I use, which has given me the best success is this.

0:06:36.600 --> 0:06:40.720
<v Speaker 3>Now it's don't use M dashes, So that is, don't

0:06:40.839 --> 0:06:44.640
<v Speaker 3>use em space dashes. And then I actually give it

0:06:44.680 --> 0:06:47.159
<v Speaker 3>an example of an M dash, So I actually paste

0:06:47.160 --> 0:06:49.680
<v Speaker 3>in an M dash within brackets, so it then knows

0:06:49.800 --> 0:06:52.160
<v Speaker 3>that's an M dash. I've been really clear about this,

0:06:53.560 --> 0:06:55.640
<v Speaker 3>and you could if you don't like short dashes either,

0:06:55.720 --> 0:06:57.800
<v Speaker 3>you could say or short dashes, and then you can

0:06:57.839 --> 0:07:01.560
<v Speaker 3>put the short dash in an inner brackets. Rely on

0:07:01.600 --> 0:07:05.200
<v Speaker 3>commas full stops or short shorter sentences instead, So you

0:07:05.200 --> 0:07:06.880
<v Speaker 3>could do things like that, or you could write it

0:07:06.920 --> 0:07:08.960
<v Speaker 3>if you're affine with the dash, you just want them

0:07:09.000 --> 0:07:11.760
<v Speaker 3>to be short. It'd be don't use M dashes, paste

0:07:11.760 --> 0:07:14.160
<v Speaker 3>in an M dash instead. I want you to use

0:07:14.280 --> 0:07:16.720
<v Speaker 3>short dashes and then paste in the short dash. So

0:07:16.720 --> 0:07:18.040
<v Speaker 3>it's up to you whether you want to ban the

0:07:18.120 --> 0:07:21.760
<v Speaker 3>dash or you just want to keep the short dash,

0:07:21.800 --> 0:07:24.559
<v Speaker 3>but yet give it first off, the words and also

0:07:24.640 --> 0:07:26.800
<v Speaker 3>the example of what the dash is you're talking about,

0:07:26.840 --> 0:07:29.800
<v Speaker 3>and I found that is better, but not one hundred

0:07:29.800 --> 0:07:32.520
<v Speaker 3>percent why it loves the M dash and it's really

0:07:32.560 --> 0:07:33.840
<v Speaker 3>hard to get it to stop doing.

0:07:36.240 --> 0:07:38.320
<v Speaker 1>What are some of the other things that you teach

0:07:38.520 --> 0:07:41.840
<v Speaker 1>in inventium dot AI programs.

0:07:42.480 --> 0:07:46.760
<v Speaker 3>These are all to undo some of the very common

0:07:46.840 --> 0:07:49.880
<v Speaker 3>things that AI does, and with AI doing in the

0:07:50.040 --> 0:07:53.160
<v Speaker 3>commonly things, it's doing it the right way. And I'm

0:07:53.200 --> 0:07:56.400
<v Speaker 3>doing your quotes here because the right way to write

0:07:56.560 --> 0:07:58.840
<v Speaker 3>is the very formal way to write, and it's been

0:07:58.840 --> 0:08:00.400
<v Speaker 3>trained on the right way to write by a whole

0:08:00.440 --> 0:08:02.240
<v Speaker 3>bunch of books and sources, and also you know the

0:08:02.240 --> 0:08:05.720
<v Speaker 3>guides on how to write well, and so therefore you

0:08:05.800 --> 0:08:08.480
<v Speaker 3>know if you're a native English speaker and you're you know,

0:08:08.600 --> 0:08:11.800
<v Speaker 3>you've done a lot of writing, chances are you more

0:08:11.840 --> 0:08:15.000
<v Speaker 3>get caught of these AI detectors and things like that.

0:08:15.080 --> 0:08:18.480
<v Speaker 3>So it's almost like putting the human back into it,

0:08:18.600 --> 0:08:21.040
<v Speaker 3>making it a little bit messier than the perfect way

0:08:21.080 --> 0:08:24.280
<v Speaker 3>to write. Okay, So with that first one's the second

0:08:24.280 --> 0:08:26.880
<v Speaker 3>one is mix sentence lengths. And the reason for this

0:08:27.000 --> 0:08:30.600
<v Speaker 3>is AI generally has the same sentence length again and

0:08:30.600 --> 0:08:34.320
<v Speaker 3>again and again, and so when you and I read this,

0:08:34.400 --> 0:08:37.080
<v Speaker 3>we may not actually pick up that these sentence lengths

0:08:37.080 --> 0:08:39.440
<v Speaker 3>are the same, but we just go this doesn't feel

0:08:39.480 --> 0:08:41.200
<v Speaker 3>like a human so we get that kind of like

0:08:41.440 --> 0:08:45.360
<v Speaker 3>the uncanny valley of writing. So say mix sentence lengths,

0:08:45.600 --> 0:08:48.400
<v Speaker 3>these short ones for punch. He's longer ones when you

0:08:48.440 --> 0:08:51.120
<v Speaker 3>need detail, And I found that one does pretty well.

0:08:51.960 --> 0:08:56.000
<v Speaker 3>So if you remember hending Away famous for short, punchy sentences,

0:08:56.160 --> 0:08:58.440
<v Speaker 3>so get some punchy things in there, get some detailed

0:08:58.480 --> 0:09:00.760
<v Speaker 3>ones that mixes it up and you and I reading

0:09:00.760 --> 0:09:04.920
<v Speaker 3>it go okay. Another one is the language that AI uses.

0:09:05.240 --> 0:09:07.599
<v Speaker 3>It can try to be the smartest AI in the

0:09:07.679 --> 0:09:11.120
<v Speaker 3>room and can like to use those long words as

0:09:11.160 --> 0:09:14.520
<v Speaker 3>part of that don't make the thing more complicated. So

0:09:14.559 --> 0:09:17.080
<v Speaker 3>you could do something like this, use playing language at

0:09:17.120 --> 0:09:20.000
<v Speaker 3>a ninth at about a ninth grade and reading level,

0:09:20.880 --> 0:09:25.360
<v Speaker 3>avoid jargon and buzzwords. Now that the ninth grade thing

0:09:25.440 --> 0:09:31.720
<v Speaker 3>or the fourteen year old type brief is common along

0:09:32.080 --> 0:09:34.800
<v Speaker 3>in industry practice, I understand when the newspaper or pro

0:09:34.840 --> 0:09:37.200
<v Speaker 3>writing and things like that, that's the kind of goal to

0:09:37.240 --> 0:09:41.959
<v Speaker 3>get to. Why because everyone can just read that pretty simply,

0:09:42.040 --> 0:09:44.400
<v Speaker 3>pretty easily. It's not really difficult to read, and you're

0:09:44.440 --> 0:09:46.600
<v Speaker 3>not going to pull out university style words that are

0:09:46.640 --> 0:09:49.000
<v Speaker 3>only a handful of people going to understand. So that's

0:09:49.040 --> 0:09:54.040
<v Speaker 3>why you do that. The other ones are sometimes it

0:09:54.120 --> 0:09:57.720
<v Speaker 3>deals with the way that sentences is constructed. Particularly start

0:09:57.720 --> 0:10:00.199
<v Speaker 3>of sentences can be a bit stiff. So one of

0:10:00.240 --> 0:10:02.360
<v Speaker 3>the things is it's fine to start a sentence with

0:10:02.520 --> 0:10:06.360
<v Speaker 3>and but also although you should do this sparingly, so

0:10:06.600 --> 0:10:09.040
<v Speaker 3>occasionally we will actually start a sentence with a handbuilt

0:10:09.080 --> 0:10:10.760
<v Speaker 3>but or so that's not the right way to do it.

0:10:10.760 --> 0:10:14.000
<v Speaker 3>I'm doing air quotes again, but humans do that, and

0:10:14.040 --> 0:10:16.160
<v Speaker 3>so therefore let's let AI do it. Not do it

0:10:16.160 --> 0:10:18.000
<v Speaker 3>all the time, but we'll let it do it. The

0:10:18.080 --> 0:10:22.800
<v Speaker 3>other one about starting sentences is avoid stiff transitions such

0:10:22.840 --> 0:10:28.520
<v Speaker 3>as additionally and moreover. Who's start human says moreover blah

0:10:28.520 --> 0:10:30.920
<v Speaker 3>blah blah blah blah. You know, you feel like you're

0:10:31.040 --> 0:10:33.480
<v Speaker 3>like a stuffy person in the eighteen hundreds with a

0:10:33.520 --> 0:10:36.920
<v Speaker 3>pipe or something. So I avoid stiff transitions such as

0:10:36.920 --> 0:10:41.880
<v Speaker 3>additionally and moreover. Now this is a weird one, but

0:10:41.960 --> 0:10:45.679
<v Speaker 3>I find this one. I used to like, but now

0:10:45.720 --> 0:10:50.560
<v Speaker 3>I hate that AI does. Often it'll frame things in

0:10:50.600 --> 0:10:53.679
<v Speaker 3>a particular way near the end of a sentence, end

0:10:53.679 --> 0:10:57.320
<v Speaker 3>of a paragraph. It'll say half bunch of explanation stuff,

0:10:57.480 --> 0:11:02.520
<v Speaker 3>and I want to say, it's not this. It's so it

0:11:02.520 --> 0:11:06.160
<v Speaker 3>maybe like, it's not just a good way of writing,

0:11:06.880 --> 0:11:10.480
<v Speaker 3>it's the way to communicate something.

0:11:10.240 --> 0:11:12.120
<v Speaker 2>Like that, right, AI.

0:11:13.480 --> 0:11:15.720
<v Speaker 3>It does this a lot, and actually was at the

0:11:15.720 --> 0:11:17.720
<v Speaker 3>start when it was coming, I was, Oh, that's really clever.

0:11:17.800 --> 0:11:20.040
<v Speaker 3>It does that really well. I like, but it over

0:11:20.280 --> 0:11:23.560
<v Speaker 3>uses that mechanism, and so therefore I actually had to

0:11:23.559 --> 0:11:25.760
<v Speaker 3>google to find out what that mechanism was called. It's

0:11:25.760 --> 0:11:31.160
<v Speaker 3>called corrective antithesis. So I say, never use corrective antithesis,

0:11:31.440 --> 0:11:33.880
<v Speaker 3>And then in brackets, given an example, I say, for example,

0:11:34.360 --> 0:11:39.280
<v Speaker 3>not X, but why and then that certainly cuts that down.

0:11:39.880 --> 0:11:42.000
<v Speaker 3>And there's one more if I can throw it in there,

0:11:42.080 --> 0:11:45.400
<v Speaker 3>which is don't lean on the rule of three. Now,

0:11:45.400 --> 0:11:47.640
<v Speaker 3>the rule of three is generally you give examples. You

0:11:47.640 --> 0:11:50.360
<v Speaker 3>can give three examples, and it kind of makes you know,

0:11:50.920 --> 0:11:55.760
<v Speaker 3>it reads nicely. But AI over uses the rule of three.

0:11:56.120 --> 0:11:58.960
<v Speaker 3>It over uses that in so many things, and so

0:11:59.440 --> 0:12:01.559
<v Speaker 3>I say, don't lean on the rule of three. Use

0:12:01.600 --> 0:12:04.199
<v Speaker 3>one or two items, or use a longer list if

0:12:04.200 --> 0:12:07.320
<v Speaker 3>it strengthens the point. So you can do one or two,

0:12:07.520 --> 0:12:09.800
<v Speaker 3>or you can do five. You know, up to you,

0:12:10.120 --> 0:12:14.080
<v Speaker 3>but don't always use rule of three, because it's that consistency,

0:12:14.320 --> 0:12:16.360
<v Speaker 3>like the sentence lengths being the same, the rule of

0:12:16.400 --> 0:12:19.000
<v Speaker 3>three being the same. It's not this, it's that being

0:12:19.080 --> 0:12:22.360
<v Speaker 3>the same that feels very robotic. Because robots are really

0:12:22.400 --> 0:12:24.720
<v Speaker 3>good at doing things the same ais really good things

0:12:24.760 --> 0:12:27.840
<v Speaker 3>the same. Humans we're messier, and so we're used to

0:12:27.880 --> 0:12:29.840
<v Speaker 3>seeing things that a little bit messier. And it's also

0:12:29.920 --> 0:12:31.920
<v Speaker 3>easier and nicer for us to read because it kind

0:12:31.920 --> 0:12:34.240
<v Speaker 3>of feels more like us. And so this is just

0:12:34.280 --> 0:12:36.400
<v Speaker 3>a whole bunch of points you can put in if

0:12:36.480 --> 0:12:39.080
<v Speaker 3>like you've got this as a standard prompt, all those

0:12:39.120 --> 0:12:41.160
<v Speaker 3>what have I got one to three, four, over, six, seven, eight,

0:12:41.520 --> 0:12:44.000
<v Speaker 3>seven or eight dot points? You can just drop them

0:12:44.040 --> 0:12:46.760
<v Speaker 3>in and then say, when you rewrite my paragraph because

0:12:46.800 --> 0:12:49.800
<v Speaker 3>I needed to be cleaned up, don't do these things.

0:12:49.800 --> 0:12:50.880
<v Speaker 3>That's what you're trying to tell them.

0:12:51.520 --> 0:12:52.760
<v Speaker 2>Okay, I love all those.

0:12:52.960 --> 0:12:56.640
<v Speaker 1>Now what I did ne in preparation for this episode

0:12:56.880 --> 0:13:02.880
<v Speaker 1>is I went into all the llms that I use

0:13:04.440 --> 0:13:08.160
<v Speaker 1>and like I love comparing the output, so I went

0:13:08.200 --> 0:13:14.120
<v Speaker 1>into chatpt Claude, Gemini and also Perplexity, and I gave

0:13:14.160 --> 0:13:16.520
<v Speaker 1>it all the same prompt, which was something along the

0:13:16.559 --> 0:13:17.719
<v Speaker 1>lines of.

0:13:19.920 --> 0:13:22.520
<v Speaker 2>Tell me what are the fifty.

0:13:22.520 --> 0:13:27.800
<v Speaker 1>Most overused or over indexed words and phrases that AI

0:13:28.360 --> 0:13:31.360
<v Speaker 1>likes to use, and I think I suggested to look

0:13:31.400 --> 0:13:36.840
<v Speaker 1>at LinkedIn for some you know, some examples. And then

0:13:37.040 --> 0:13:39.840
<v Speaker 1>what I did with the output that I got from

0:13:39.880 --> 0:13:44.200
<v Speaker 1>each of those four models, I then put them all

0:13:44.280 --> 0:13:46.320
<v Speaker 1>into the one model Claudes generally the one that I

0:13:46.400 --> 0:13:49.800
<v Speaker 1>default too, and I said, give me a summary of

0:13:49.880 --> 0:13:53.800
<v Speaker 1>the ones that are appearing most frequently across these four outputs.

0:13:54.120 --> 0:13:58.120
<v Speaker 1>And so with that, I've then created a bit of

0:13:58.160 --> 0:14:01.360
<v Speaker 1>a mega prompt that people can use that I've called

0:14:01.360 --> 0:14:05.480
<v Speaker 1>the anti ai slop prompt and if you want that,

0:14:05.600 --> 0:14:07.240
<v Speaker 1>I will link to it in the show notes. But

0:14:07.280 --> 0:14:09.440
<v Speaker 1>I'm just going to go through a few things just

0:14:09.480 --> 0:14:12.880
<v Speaker 1>for some additional advice for people that are trying to

0:14:13.679 --> 0:14:19.000
<v Speaker 1>un ai aify their writing so often. And these are

0:14:19.000 --> 0:14:22.280
<v Speaker 1>grouped into some categories. So there's the first category is

0:14:22.640 --> 0:14:27.040
<v Speaker 1>empty opener phrases. So you will have seen things like

0:14:27.120 --> 0:14:30.280
<v Speaker 1>this in today's fast paced world, in an ever changing

0:14:30.360 --> 0:14:32.400
<v Speaker 1>landscape when it comes to.

0:14:32.520 --> 0:14:34.200
<v Speaker 2>Blah, picture this black like.

0:14:34.240 --> 0:14:36.440
<v Speaker 1>These are all sounding really familiar because these are how

0:14:36.800 --> 0:14:41.040
<v Speaker 1>a lot of certainly a lot of LinkedIn post starts start. Now,

0:14:41.240 --> 0:14:45.320
<v Speaker 1>then we've got a category of fake profound hooks, which

0:14:45.360 --> 0:14:47.720
<v Speaker 1>I think, you know, it's not X, it's why is

0:14:47.760 --> 0:14:52.200
<v Speaker 1>certainly one of those. Here's the truth, which is one

0:14:52.240 --> 0:14:55.240
<v Speaker 1>of my pet hates, or sometimes he's the uncomfortable truth,

0:14:55.280 --> 0:14:58.880
<v Speaker 1>he's the inconvenient truth. I hate that one because I

0:14:59.040 --> 0:15:02.760
<v Speaker 1>use that language here the truth or here's the thing.

0:15:02.960 --> 0:15:04.800
<v Speaker 1>I actually use that a lot in my writing. And

0:15:04.840 --> 0:15:07.320
<v Speaker 1>what is really frustrating, as someone who was writing books

0:15:07.360 --> 0:15:10.040
<v Speaker 1>before the age of AI is that I feel like

0:15:10.120 --> 0:15:14.080
<v Speaker 1>I now have to change my writing style, which was

0:15:14.640 --> 0:15:17.480
<v Speaker 1>you know, certainly in some ways similar to what AI

0:15:17.640 --> 0:15:20.640
<v Speaker 1>produces to you know, just added like I used m

0:15:20.720 --> 0:15:23.960
<v Speaker 1>dashes very often, but now I will lean more towards

0:15:23.960 --> 0:15:28.960
<v Speaker 1>brackets and semi collins for like little asides or quips

0:15:29.600 --> 0:15:32.680
<v Speaker 1>as my editor calls them, which is very frustrating when

0:15:32.720 --> 0:15:34.160
<v Speaker 1>you have to change your own writing style so it

0:15:34.160 --> 0:15:38.400
<v Speaker 1>doesn't look like AI. And we've got lots of corporate buzzwords.

0:15:38.440 --> 0:15:42.800
<v Speaker 1>Probably the most common is delve. I was hearing some

0:15:42.880 --> 0:15:46.560
<v Speaker 1>research actually that was done in twenty twenty three where

0:15:47.760 --> 0:15:52.440
<v Speaker 1>academic authors that were most prolific in twenty twenty three

0:15:52.440 --> 0:15:55.920
<v Speaker 1>in terms of getting things published over indexed heavily on

0:15:55.960 --> 0:15:58.920
<v Speaker 1>the word delve. I don't know if you've heard that research.

0:15:59.000 --> 0:16:02.360
<v Speaker 1>I heard Ethan like sharing that recently, which made me laugh.

0:16:03.080 --> 0:16:05.560
<v Speaker 3>I didn't hear that researcher. I'm a bit sad about

0:16:05.560 --> 0:16:09.320
<v Speaker 3>delve because I use delve as well, but AI uses

0:16:09.360 --> 0:16:12.840
<v Speaker 3>it more. And so you may add a bunch of words.

0:16:12.880 --> 0:16:14.720
<v Speaker 3>Can we also talk about other words we may want

0:16:14.760 --> 0:16:16.320
<v Speaker 3>to add to that banned list?

0:16:17.360 --> 0:16:19.760
<v Speaker 2>For sure? For sure? Give me like three.

0:16:19.720 --> 0:16:24.680
<v Speaker 3>More quietly is another one that I was going.

0:16:24.480 --> 0:16:25.720
<v Speaker 2>To say quietly.

0:16:26.040 --> 0:16:29.280
<v Speaker 1>So the next category actually is what AI has termed

0:16:29.400 --> 0:16:34.040
<v Speaker 1>hollow adjectives and nouns and adjectives, and if you don't

0:16:34.080 --> 0:16:38.720
<v Speaker 1>remember from your English class, and adjective is a descriptive

0:16:38.760 --> 0:16:42.000
<v Speaker 1>word for a noun, So it describes a noun which

0:16:42.040 --> 0:16:46.000
<v Speaker 1>is like an object or a person, a person, place

0:16:46.120 --> 0:16:48.760
<v Speaker 1>or thing. I think is how my English teacher probably

0:16:48.760 --> 0:16:54.720
<v Speaker 1>would have described it. Quietly is so overused. And again

0:16:54.760 --> 0:16:56.800
<v Speaker 1>that really bothers me because that is a word that

0:16:56.840 --> 0:16:58.440
<v Speaker 1>I would have used, but now I feel like I

0:16:58.480 --> 0:17:00.280
<v Speaker 1>have to strip it out mostly from my writing.

0:17:00.360 --> 0:17:02.880
<v Speaker 2>Okay, what are the words near this?

0:17:03.680 --> 0:17:06.600
<v Speaker 3>Ultimately, I would recommend you build your own list of

0:17:06.640 --> 0:17:08.920
<v Speaker 3>words that you don't want AI to use, because there's

0:17:08.920 --> 0:17:11.199
<v Speaker 3>some there that AI will have used, like Delvin quietly,

0:17:11.560 --> 0:17:13.760
<v Speaker 3>but there may be others that IT would use that

0:17:14.160 --> 0:17:17.679
<v Speaker 3>I find words but you would never use. So I

0:17:17.800 --> 0:17:21.520
<v Speaker 3>use the example of like buzzword bingo, so like synergistic. Right,

0:17:21.720 --> 0:17:24.360
<v Speaker 3>if AI uses the words synergistic and you would never

0:17:24.480 --> 0:17:27.359
<v Speaker 3>use it, then put a list together and add to

0:17:27.400 --> 0:17:29.719
<v Speaker 3>that list, but sudogistic in there, maybe put delve in there,

0:17:29.720 --> 0:17:32.080
<v Speaker 3>maybe put quietly in there, and go, don't ever use

0:17:32.119 --> 0:17:34.840
<v Speaker 3>these words, And so build up that list and enhance

0:17:34.920 --> 0:17:37.640
<v Speaker 3>that over time as AI uses words that just great

0:17:37.720 --> 0:17:39.920
<v Speaker 3>on you, because if you've been grading on you, they

0:17:39.920 --> 0:17:40.760
<v Speaker 3>don't sound like you.

0:17:41.480 --> 0:17:42.240
<v Speaker 2>So hopefully.

0:17:42.280 --> 0:17:44.360
<v Speaker 1>I mean, this is so much to take in, as

0:17:44.400 --> 0:17:47.640
<v Speaker 1>I said, for those that are listening and like, yeah,

0:17:47.720 --> 0:17:50.880
<v Speaker 1>I really, I really need to improve my writing output

0:17:50.920 --> 0:17:53.159
<v Speaker 1>because I could relate to a lot of those things,

0:17:53.240 --> 0:17:55.399
<v Speaker 1>and I think about my writing. Yes, I'm using a

0:17:55.400 --> 0:17:56.760
<v Speaker 1>lot of the things that Neo and a manter I

0:17:56.760 --> 0:17:59.919
<v Speaker 1>have talked about. Go to the show notes. There is

0:18:00.240 --> 0:18:04.720
<v Speaker 1>a link to get our anti AI slopped prompt, which

0:18:04.800 --> 0:18:07.200
<v Speaker 1>I highly recommend you customize.

0:18:07.240 --> 0:18:08.960
<v Speaker 2>So it's a lengthy prompt.

0:18:09.880 --> 0:18:12.000
<v Speaker 1>It includes a lot of what we've spoken about today,

0:18:12.040 --> 0:18:16.000
<v Speaker 1>but do read it and put in your own personal

0:18:16.880 --> 0:18:19.480
<v Speaker 1>pet peeves so that you can customize it and run

0:18:19.560 --> 0:18:23.119
<v Speaker 1>any content that you create or create with AI through

0:18:23.359 --> 0:18:26.479
<v Speaker 1>this prompt. It will improve your work a lot. And

0:18:26.680 --> 0:18:30.200
<v Speaker 1>also I would say, please share this episode with maybe

0:18:30.240 --> 0:18:33.639
<v Speaker 1>some people that you know that are perhaps producing a

0:18:33.640 --> 0:18:35.760
<v Speaker 1>bit of AI slot, but they are actually doing the

0:18:35.800 --> 0:18:38.879
<v Speaker 1>thinking themselves, which is the worst combination where the human

0:18:38.960 --> 0:18:41.600
<v Speaker 1>is doing the majority of the work and AI is

0:18:41.640 --> 0:18:44.520
<v Speaker 1>then going over it and AI are fying it and

0:18:44.560 --> 0:18:48.280
<v Speaker 1>completely destroying that person's credibility, so do share this episode

0:18:48.359 --> 0:18:53.119
<v Speaker 1>with those that need it. How i AI was hosted

0:18:53.119 --> 0:18:56.200
<v Speaker 1>by me Amantha Imber and Neo Applan. A big thank

0:18:56.200 --> 0:18:58.760
<v Speaker 1>you to Martin Imba who does our sound editing and

0:18:58.880 --> 0:19:02.280
<v Speaker 1>Jem Rubio for production support, and thank you to John

0:19:02.359 --> 0:19:04.320
<v Speaker 1>Kilby who composed the theme music