We need to become adept at prompting AI models, as there’s a massive difference between a vague prompt and very well-
defined, specific prompt that sets a context window for the model to operate within. A very banal prompt results in what you’d
expect, a banal answer. So I think in order to be a good prompt engineer, you first have to be educated. As such, I do not think that includes an over-reliance on technology.
PROF. ALEKSANDRA PRZEGALIŃSKA-SKIERKOWSKA Vice-Rector for International Cooperation and ESR, Kozminski University in Warsaw
playing the game; which strategies work, which games are extremely popular and which are not. You have a data set that will inform your model greatly and should probably be the basis of a model that creates interesting solutions for you in the future. And then there is also an aspect that I find very interesting, which is human machine interaction. Somebody promotes these models, somebody asks the questions, somebody pushes them further.
I’ve tried this model myself, with my co-author, as we started out with the aim of writing a book about AI using AI. What we got was something I’d describe as ‘mid.’ It was okay-ish, but it was nothing special. We had to work very diligently on each and every chapter, allowing the model to generate bits and pieces, but ultimately we were responsible for bringing it all together. We were the ones who had to give it our unique style and only in this manner were we able to create a book that I consider to be quite interesting.
I'm happy that we attempted to write the book in this way, as I understand that ultimately, if you want to take a shortcut, you will probably end up having a design that’s propelled in the wrong direction. I’m sure in the same way that you could create a game design using any available AI model, but if you want something special, that approach is not feasible.
So we’ll still need to learn how to code? We can’t leave everything to Artificial Intelligence?
I think there will be a distinction between low quality education that will be very much reliant on technology instead of teaching. Tis type of learning will simulate the teaching process where students will generate their work using AI, and the teacher, instead of checking the work, will use an AI model to give them feedback. It’s a scenario that’s going to
42
create a very weird loop. Tat's one way to go, but I am a firm believer in an education that happens outside the technological realm.
I think that's the best way forward, and I made my recommendations to the Ministry of Higher Education in Poland recently, for the need to gather expert opinions on how we should move forward in the AI revolution. I believe it’s best to practice and develop skills without these models. We have to be able to prompt the models in interesting ways and not settle for a ‘mid’ solution, but actually something unique and interesting.
We need to become adept at prompting AI models, as there’s a massive difference between a vague prompt and very well-defined, specific prompt that sets a context window for the model to operate within. A very banal prompt results in what you’d expect, a banal answer. So I think in order to be a good prompt engineer, you first have to be educated. As such, I do not think that includes an over-reliance on technology. Tat’s not the way forward.
I’ll also say, that while some people think that programming is dead, I don't think it's the case. Te majority of these models today require supervision of highly-skilled people. Technical systems require supervision of people who understand exactly what these models do - line by line. Saying, ‘let's get rid of programming’ is not a good thing.
I understand that the majority of the population might not be interested in programming, but everyone deserves a chance to be introduced to the topic. And some of these people will be extremely interested in that and can develop further in this direction. And I don't think that's going to stop because there are limits to the translation between code and natural language. It's quite a messy border. So if you don't program, probably you will not be able to supervise something like ChatGPP or any other model to help build upon it.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74 |
Page 75 |
Page 76 |
Page 77 |
Page 78 |
Page 79 |
Page 80 |
Page 81 |
Page 82 |
Page 83 |
Page 84 |
Page 85 |
Page 86 |
Page 87 |
Page 88 |
Page 89 |
Page 90 |
Page 91 |
Page 92 |
Page 93 |
Page 94 |
Page 95 |
Page 96 |
Page 97 |
Page 98 |
Page 99 |
Page 100 |
Page 101 |
Page 102 |
Page 103 |
Page 104 |
Page 105 |
Page 106 |
Page 107 |
Page 108 |
Page 109 |
Page 110 |
Page 111 |
Page 112 |
Page 113 |
Page 114 |
Page 115 |
Page 116 |
Page 117 |
Page 118 |
Page 119 |
Page 120 |
Page 121 |
Page 122 |
Page 123 |
Page 124 |
Page 125 |
Page 126 |
Page 127 |
Page 128 |
Page 129 |
Page 130 |
Page 131 |
Page 132 |
Page 133 |
Page 134 |
Page 135 |
Page 136 |
Page 137 |
Page 138 |
Page 139 |
Page 140 |
Page 141 |
Page 142
