1
00:00:02,360 --> 00:00:02,880
Hi guys.

2
00:00:02,880 --> 00:00:03,840
Welcome to this session.

3
00:00:03,840 --> 00:00:09,000
So in this session we wanted to see another style of prompting which is reverse prompting reverse prompting

4
00:00:09,000 --> 00:00:10,880
or reverse engineering the prompt.

5
00:00:11,080 --> 00:00:16,840
So what we are basically talking about here is how you can go ahead and reverse engineer any piece of

6
00:00:16,840 --> 00:00:21,440
content to go back to the prompt which generated that content.

7
00:00:21,720 --> 00:00:26,800
So the intent over here is understanding the content which you have received, which you see right now.

8
00:00:27,200 --> 00:00:33,440
What prompt can generate that content particularly that is what we are trying to reverse engineer over

9
00:00:33,440 --> 00:00:33,800
here.

10
00:00:34,000 --> 00:00:39,280
So we have come up with two prompt formulas which you can use out here for this particular purpose,

11
00:00:39,320 --> 00:00:40,720
wherein you can give the prompt.

12
00:00:40,720 --> 00:00:47,880
And this will help to reverse engineer the content to go back to the original prompt, which was given

13
00:00:47,880 --> 00:00:49,320
to get that content out.

14
00:00:49,640 --> 00:00:55,520
So if you see the first one is where we ask ChatGPT to act like a prompt engineer expert that is able

15
00:00:55,520 --> 00:00:58,720
to reverse engineer prompts based on the text that is provided to you.

16
00:00:58,880 --> 00:01:06,290
So we give this particular prompt first, and set up the whole space stage for ChatGPT that it works

17
00:01:06,290 --> 00:01:09,650
like a reverse engineer prompt prompting expert.

18
00:01:09,650 --> 00:01:16,490
And then we once we once ChatGPT acknowledges it, then we can give the particular text to it, and

19
00:01:16,490 --> 00:01:21,410
it will reverse engineer the prompt and tell us the original prompt which was given for that content.

20
00:01:21,650 --> 00:01:22,650
This is one option.

21
00:01:22,690 --> 00:01:29,010
The second option is prompt can be wherein we are giving multiple different prompts to ChatGPT to set

22
00:01:29,050 --> 00:01:33,450
up the conversation clearly wherein we first initially say that.

23
00:01:33,450 --> 00:01:35,250
Let's talk about reverse prompt engineering.

24
00:01:35,250 --> 00:01:39,690
By reverse prompt engineering, I mean creating a prompt from a given text.

25
00:01:39,730 --> 00:01:39,890
Okay.

26
00:01:39,930 --> 00:01:43,050
Can you give me some simple examples of reverse prompt engineering.

27
00:01:43,050 --> 00:01:45,250
So ChatGPT will give us some examples.

28
00:01:45,690 --> 00:01:50,250
Then we will say can you create a very technical reverse prompt engineering template.

29
00:01:50,250 --> 00:01:53,090
So what are we doing is we are priming the tool.

30
00:01:53,290 --> 00:02:00,210
We are priming the tool specifically to have previous historical conversation data so that it understands

31
00:02:01,130 --> 00:02:02,770
reverse prompt engineering better.

32
00:02:04,130 --> 00:02:07,650
And then finally we give the prompt which is now reverse prompt.

33
00:02:07,650 --> 00:02:09,370
Engineer the following text.

34
00:02:09,410 --> 00:02:13,490
Be sure to capture the tone, syntax, language, and writing style of the text.

35
00:02:13,610 --> 00:02:19,850
So with this two different approaches, possibly you will be able to go ahead and reverse engineer the

36
00:02:19,850 --> 00:02:26,010
prompt and go back to the original prompt which generated the content which you have now.

37
00:02:26,370 --> 00:02:31,410
The intent of doing this is once you get the original prompt, you can use it on other products.

38
00:02:31,410 --> 00:02:39,810
So if you come across a really good content across anywhere, you can use ChatGPT to reverse engineer

39
00:02:39,810 --> 00:02:43,930
and take you back to the original prompt which can generate it.

40
00:02:43,930 --> 00:02:51,330
And now that you have the original prompt with you, you can imply you can apply that on other products,

41
00:02:51,330 --> 00:02:54,010
your own products, in your own business as well.

42
00:02:54,330 --> 00:02:57,890
So let's see this in action how really this is going to happen.

43
00:02:57,890 --> 00:03:00,530
So what we're going to do first is look at the first option.

44
00:03:03,090 --> 00:03:08,730
So we are going to go ahead and take the first prompt and give it to ChatGPT.

45
00:03:08,770 --> 00:03:13,090
We will say the type of content is let's say a tech company.

46
00:03:15,420 --> 00:03:17,900
Product description.

47
00:03:24,300 --> 00:03:24,940
Understood.

48
00:03:25,100 --> 00:03:25,620
Okay.

49
00:03:25,700 --> 00:03:26,540
And then.

50
00:03:31,860 --> 00:03:33,220
We will give the second prompt.

51
00:03:33,620 --> 00:03:33,860
Great.

52
00:03:33,860 --> 00:03:40,300
The text I would like to reverse engineer is and we'll give the example from here.

53
00:03:41,020 --> 00:03:42,540
Let's say the example is this.

54
00:03:44,060 --> 00:03:46,220
This is the content which we have got hold of.

55
00:03:47,860 --> 00:03:54,380
And what we expect out of ChatGPT is give us the original prompt for this which will generate this kind

56
00:03:54,380 --> 00:03:55,100
of content.

57
00:04:02,300 --> 00:04:09,420
So you can see it has generated the prompt as well which will help us to generate this kind of content

58
00:04:09,460 --> 00:04:10,340
ideally speaking.

59
00:04:11,180 --> 00:04:14,820
So this is one approach which you can easily use out here.

60
00:04:15,140 --> 00:04:17,780
The second approach let's have a look at that as well.

61
00:04:19,710 --> 00:04:23,710
In the second approach, we start the conversation with this where we say.

62
00:04:28,030 --> 00:04:30,870
It understands reverse engineering what it is.

63
00:04:31,670 --> 00:04:36,070
Then we ask ChatGPT to give us an example of prompt engineering.

64
00:04:39,870 --> 00:04:42,150
So it will give us some example of prompt engineering.

65
00:04:42,150 --> 00:04:43,430
Reverse prompt engineering.

66
00:04:54,550 --> 00:04:58,910
Right now it is still giving us the result for the first prompt.

67
00:04:59,070 --> 00:05:02,510
Now we are asking the second one asking for an example of a reverse prompt engineering.

68
00:05:07,710 --> 00:05:11,830
Now we are going to ask ChatGPT to create a template for reverse prompt engineering.

69
00:05:17,310 --> 00:05:18,950
So we are priming the tool.

70
00:05:19,150 --> 00:05:25,120
We are giving a lot of data to ChatGPT to understand from reverse prompt engineering, because our intention

71
00:05:25,120 --> 00:05:30,840
is to ask it to create a particular prompt for the original content at the end.

72
00:05:33,120 --> 00:05:36,360
And now this is the final prompt which we want to give.

73
00:05:41,880 --> 00:05:45,640
You can see it is giving us the response for the prompt right now.

74
00:05:51,560 --> 00:05:53,040
And now we can give our.

75
00:05:53,080 --> 00:05:57,240
We'll ask ChatGPT to prompt engineer the following text.

76
00:05:57,440 --> 00:05:57,720
Okay.

77
00:05:57,760 --> 00:06:03,280
So let's say this is a product for which which has a very high reviews.

78
00:06:03,280 --> 00:06:04,040
Number of reviews.

79
00:06:04,040 --> 00:06:05,400
Good rating already.

80
00:06:05,440 --> 00:06:07,720
We want to reverse engineer the prompt.

81
00:06:07,760 --> 00:06:11,680
We want to know the original prompt which can generate this kind of a headline.

82
00:06:11,680 --> 00:06:13,440
We can reverse engineer for this.

83
00:06:13,480 --> 00:06:17,320
We can reverse engineer for the description of the product right here.

84
00:06:17,800 --> 00:06:22,640
Multiple things, whichever things which is needed for you for your own product listing, you can ask

85
00:06:22,640 --> 00:06:26,120
it to reverse engineer and take you back to the original prompt.

86
00:06:26,400 --> 00:06:28,320
So I'm taking the headline for the time being.

87
00:06:28,320 --> 00:06:28,440
mean.

88
00:06:31,760 --> 00:06:32,920
I've given the headline.

89
00:06:40,280 --> 00:06:45,360
And now we are asking it to reverse engineer that original text it is taking.

90
00:06:58,080 --> 00:07:01,480
Now you can see it is generating the reverse engineered prompt for us.

91
00:07:02,960 --> 00:07:06,400
This we can use to generate this kind of a headline going forward.

92
00:07:06,440 --> 00:07:11,880
So now once you have the the original prompt with you, you can use it on any product.

93
00:07:12,240 --> 00:07:12,480
Okay.

94
00:07:12,480 --> 00:07:15,040
You can just change the product name over here.

95
00:07:15,040 --> 00:07:21,280
And the style tone syntax remains the same, but you can use it on any other product of your own okay.

96
00:07:21,320 --> 00:07:25,680
For your product descriptions and it will write in that particular style.

97
00:07:26,000 --> 00:07:26,880
I hope this makes sense.

98
00:07:26,880 --> 00:07:29,840
You understand the concept of reverse prompting now.

99
00:07:29,880 --> 00:07:34,200
Thank you so much guys for listening to this and I will see you in the next video.