Prompt (Computing)

  • Command prompt, characters indicating the computer is ready to accept input
  • Command Prompt, also known as cmd.exe or cmd, the command-line interpreter in some operating systems
  • Prompt (natural language), instructions issued to a computer system (such as a text-to-image artificial intelligence) in the form of written or spoken language.
  • Prompt engineering, a concept in artificial intelligence in which the description of the task is embedded in the input, e.g., as a question, instead of it being implicitly given.

Source: Wikipedia

Maurício Pinheiro

Introduction

The evolution of Artificial Intelligence is changing the way several professions are carried out, and some of them may be at risk of extinction. For example, in areas such as accounting and finance, routine processes such as preparing spreadsheets and financial reports can be automated through AI systems, which could replace professionals in these areas. In addition, medicine is also undergoing significant changes with the development of AI-based diagnostic systems, which can be more accurate than diagnoses performed by human doctors. Thus, radiologists will be replaced by intelligent imaging software. Other professions that may be threatened include analysts, librarians, actuarial science professionals, telemarketers, administrative assistants, lawyers, attorneys, judges and many other professionals who perform routine and standardized activities, or rely on a system of rules, liturgies and standard procedures for carrying out their functions (e.g. the constitution of a country or a civil or penal code, as long as they are well written.). A skilled prompt engineer could be the key to ensuring these professions adapt and evolve along with AI rather than being wiped out entirely, or perhaps responsible for burying them entirely.

Prompt

Prompt is an interaction technique with artificial intelligence (AI) systems that allows the user to provide input in the form of text and receive a response in natural language, based on analysis of the input provided by the AI. While most people associate prompts with simple questions and answers, prompts can be much more complex and sophisticated. In fact, prompts can be designed to perform more complex tasks such as generating descriptive text, completing sentences, making predictions, analyzing sentiment, creating code, and much more. As the use of AI is expanding and replacing more and more traditional (encyclopedic) professions, there is a need for a prompt engineer, a professional able to communicate with the machine and design custom prompts for the specific needs of the client. It is important to note that, with the evolution of AI, even programmers can be replaced by a prompt engineer, since AI can create its own codes when well guided.

The prompt technique is widely used in areas such as virtual assistants, chatbots, natural language processing and machine learning. The goal is to create interactions with AI more naturally and efficiently, allowing users to obtain accurate and useful information more easily and quickly. It can be seen as a kind of input to AI, where the response generated by the machine is based on its training on large amounts of data and its ability to understand natural language. Prompts are often used in AI to provide information, perform actions, or interact with users in a natural and intuitive way. Creating effective prompts is an essential skill for prompt engineers working to develop and improve AI systems for use in a wide variety of industries and applications.

Here, Prompt Engineering is a hypothetical profession (which has become increasingly real) and which involves interaction with artificial intelligence (AI) models through prompts, not necessarily textual, but which can be of different natures, including a BCI (brain-computer interface). This profession would be responsible for creating and optimizing prompts for AI models to generate useful responses in various areas such as customer service, content generation, and more. As AI and language models become more advanced and widely used, Prompt Engineering could become more and more relevant.

A prompt engineer interacts with an AI through a process called “prompting”. This interaction involves creating prompts specific to the AI model, which will be used to generate responses to the requested questions or tasks.

Taking the GPT-3 language model as an example, the prompt engineer must first understand the model’s capabilities and limitations, as well as the specific purpose for which the model will be used. Then he needs to create prompts that are effective for the requested task.

For example, imagine that a prompt engineer works at a customer service company that uses ChatGPT to answer customer questions. ChatGPT is an example of LLM (Language Model), which uses deep learning to generate natural language responses from a text input.

For ChatGPT to be effective in customer service, the prompt engineer must create specific prompts that help the model understand the customer’s intent and provide a useful response. For example, a prompt could be something like “Hello! How can I help you today?”, which ChatGPT will use as input to generate a personalized and helpful response.

The prompt engineer can also tweak existing prompts based on data and customer feedback to further improve the model’s effectiveness. Additionally, it can test different prompts to determine which one is most effective for a given task or scenario.

In short, the prompt engineer interacts with an AI by creating and optimizing specific prompts for the requested task or scenario. In the case of ChatGPT, prompts are used to generate natural language responses from a text input.

I asked Chat-GPT to create a draft of a Pedagogical Project for a higher education course based on these skills.

Pedagogical Project: Prompt Engineering

1. Introduction

The Prompt Engineering course aims to train professionals capable of creating, developing and optimizing prompts in Artificial Intelligence (AI) systems, which can generate useful and accurate responses for users. For this, the course covers disciplines that include knowledge of AI, natural language, data analysis, mathematics and other complementary areas. It is a five-year degree program that aims to prepare students to work with Artificial Intelligence (AI) and language models to develop innovative solutions that improve communication between humans and machines.

2. Skills

A prompt engineer needs to have both technical and creative skills to perform his duties efficiently. Some of the required skills include:

  1. Knowledge in artificial intelligence and language models: A prompt engineer needs to understand how language models work and how AI can be used to solve problems.
  2. Skill in creating and optimizing prompts: the professional needs to have skills in creating and tuning prompts that can generate useful and accurate answers.
  3. Natural Language Knowledge: The prompt engineer must have a deep understanding of natural language to be able to create effective prompts.
  4. Data analysis skills: To create and optimize prompts, the prompt engineer must have data analysis skills to understand how the prompts are working and how they can be improved, as well as to structure the outputs and make them understandable.
  5. Communication skills: the professional must have communication skills to communicate clearly with other professionals involved in the project.
  6. Creative thinking: The prompt engineer must have creative thinking skills to create prompts that generate innovative and useful responses.
  7. Comprehensive culture and knowledge.

In summary, a prompt engineer needs to have technical skills in artificial intelligence, language models, natural language, data analysis and creative thinking, as well as communication skills to ensure project success.

2. Goals

  • Train prompt engineers capable of working in multidisciplinary AI systems development teams;
  • Develop technical and creative skills for creating and optimizing prompts;
  • Provide knowledge in natural language and data analysis to improve the efficiency of prompts;
  • Enable students to understand the mathematical logic behind language models and AI algorithms;
  • Encourage general culture and knowledge in complementary areas for the formation of more complete professionals.

3. Curriculum

Prompt’s Engineering course would be structured around three main themes: Fundamentals, Techniques and Applications.

The Fundamentals axis encompasses disciplines that provide a solid foundation for understanding the concepts of artificial intelligence, natural language and language models, such as Introduction to Artificial Intelligence, Mathematical Logic, Discrete Mathematics and Linear Algebra.

The Techniques axis covers disciplines that focus on the techniques and tools necessary for the development of efficient prompts, including Programming, Natural Language Processing, Machine Learning, Optimization and Algorithms, Database and Graph Theory.

The Applications axis addresses disciplines that focus on practical cases and applications of the Prompt Engineer in different sectors and industries, such as Computer Networks, Communication and Marketing, Law and Intellectual Property, and Entrepreneurship.

1st year:

Introduction to Artificial Intelligence: In this course, students will learn the fundamental concepts of artificial intelligence, including machine learning, natural language processing, and computer vision.

Mathematical Logic: This course will teach students mathematical logic and its applications in artificial intelligence.

Introduction to Natural Language: Students will learn the fundamentals of natural language and how it is processed by AI.

Discrete Mathematics: In this course, students will study discrete mathematical concepts such as set theory, Boolean algebra and graphs.

Calculus I: The Calculus I course will cover functions, limits, derivatives and applications of derivatives.

Natural Language Processing: presents natural language processing techniques and the main tools used for text analysis and automatic response generation.

Programming Fundamentals: Students will learn the Python programming language and its basic fundamentals.

Communication and Expression: This course will help students develop their communication and expression skills in public and writing.

2nd Year:

Language Models: In this course, students will study language models and how they can be applied to natural language processing problems.

Linear Algebra: Students will study linear algebra concepts and their applications in AI.

Statistics and Data Analysis: In this course, students will learn basic statistics and data analysis techniques for working with data in AI projects.

Calculus II: The Calculus II course will cover integrals and applications of integrals.

Operating Systems: Students will learn about operating systems and their applications in AI projects.

Culture and Literature: This course will help students expand their knowledge of culture and literature to develop a more comprehensive view of the world.

3rd Year:

Natural Language Processing: Students will study advanced natural language processing techniques such as semantic analysis and topic modeling.

Advanced Probability and Statistics: In this course, students will study advanced concepts of probability and statistics for working with data in AI projects.

Calculus III: The Calculus III course will cover series and power series.

Computer Networks: Students will study computer networks and their application in AI projects.

Philosophy and Ethics: This course will help students develop an understanding of philosophy and ethics to apply to AI projects.

Social Sciences: In this course, students will study the social sciences and their role in AI projects.

4th year:

Machine Learning: this course aims to introduce students to the fundamental concepts of machine learning, including supervised and unsupervised learning algorithms, model validation and data pre-processing techniques.

Optimization and Algorithms: In this course, students will learn about advanced optimization techniques, including gradient descent methods and linear and non-linear programming algorithms.

Graph Theory: Graph theory is an important area of discrete mathematics that focuses on the study of networks and connections between objects. Students will learn to apply graph concepts to prompt engineering problems such as analyzing social networks and optimizing conversation scripts.

Database: In this course, students will learn to design and implement relational and non-relational databases to store and manage large amounts of data. Students will also learn about SQL and other database query languages.

History and Arts: this course aims to broaden students’ vision and deepen their knowledge of culture and history, focusing on aspects related to literature, arts and cultural movements.

Communication and Marketing: this course aims to teach students how to communicate and promote their prompt engineering projects effectively. Students will learn digital marketing techniques, advertising and communication strategies to engage and attract users.

5th year:

Final Project: Creating Prompts: In this course, students will work in groups to develop and implement effective prompts to solve a specific problem in the area of prompt engineering. The final project will be presented to a panel of evaluators.

Entrepreneurship Project: this course aims to teach students the skills needed to start and manage a business. Students will learn about entrepreneurship, strategic planning, market analysis and business development.

Law and Intellectual Property: This course aims to provide students with an understanding of the fundamental principles of law and intellectual property, with a focus on patents, trademarks and copyright.

World Culture and History: this course aims to broaden students’ vision and deepen their knowledge of world history and culture, focusing on aspects related to cultural diversity and social movements.

Supervised Internship: in this discipline, students will have the opportunity to apply their knowledge in a real work environment, under the supervision of experienced professionals in the field of prompt engineering. Students will be able to work on challenging projects and develop practical skills for their future career.

4. Methodology

The Prompt Engineering course will be offered in face-to-face (or semi-face-to-face) mode, with theoretical and practical classes, laboratories and integrative projects. The adopted methodology will be based on practical activities, stimulating the resolution of problems, the creation and optimization of prompts, and the application of techniques and tools studied in real projects.

In addition, the course will feature extracurricular activities, such as technical visits, lectures by market professionals, seminars and workshops.

5. Evaluation

The evaluation of student performance will be carried out on an ongoing basis, through tests, assignments, presentations, projects and participation in classroom and extracurricular activities.

6. Completion of course work

The Course Completion Work (CCW) will consist of creating and implementing an original prompt in a real application, using techniques and tools learned throughout the course. The CCW will be guided by a professor and evaluated by an examining board.

7. Supervised Internship

The supervised internship will be mandatory and will have a minimum workload of 200 hours, to be completed at the end of the course. The internship must be carried out in companies or institutions that can offer the student practical experience in the field of Prompt Engineering.

8. Faculty, physical structure and investment

The teaching staff of the Prompt Engineering course must be composed of highly qualified and multidisciplinary professors, with doctorates in areas such as Artificial Intelligence, Computing, Linguistics, Mathematics, Philosophy and Ethics. This diversity of knowledge areas is essential for teaching the technical and creative skills necessary for the prompt engineer, as well as for a humanistic and culturally rich general education.

To support teaching and research in the Prompt Engineering course, a modern and well-equipped physical infrastructure will be required, including classrooms with audiovisual resources, computer labs equipped with high-speed computer networks and software necessary for the course disciplines , in addition to study rooms for students.

Considering the investment required to maintain this infrastructure, the monthly investment amount to keep the course running would be around US$150,000.00, including equipment maintenance, software updates and hiring of technical staff. For the monthly fee, a value of US$ 2,500.00 per student is suggested for two classes of 50 students, which would total a monthly income of US$ 250,000.00. This would cover costs and still invest in ongoing improvements in infrastructure and faculty.

In addition to being funded by tuition fees and investments in infrastructure, the prompt engineering course can also benefit from extension projects. These projects may involve partnerships with companies and organizations that need solutions in the area of artificial intelligence and natural language processing. In this way, the course can offer opportunities for students to apply their knowledge in real situations and develop practical skills, while meeting the demands of society. These projects can be financed through public notices, government funds and resources, as well as through partnerships with companies and organizations. Financing for extension projects is a way of adding value to the course and enriching students’ training.

Warning

Nowadays, educational institutions are increasingly adapting to the needs of the ever-changing job market, offering more flexible and customizable courses. Instead of a rigid curriculum, students have the opportunity to choose optional and elective subjects after a common basic cycle in major areas of knowledge (Exact, Humanities and Biology). This more adaptable approach allows students to develop relevant and up-to-date skills for the market, instead of learning professional techniques and tools, which are often outdated and imposed by professional corporatism. Furthermore, this approach also encourages students to explore interdisciplinary areas and discover new fields of interest that could be useful in their future careers. In this way, flexibility in course development can help promote teaching that is more in line with market needs and student expectations.

This pedagogical project of Prompt Engineering, which has a multidisciplinary faculty and a specific physical infrastructure, and even so can be considered plastered and inflexible compared to this trend. While the program was designed to produce engineers who are highly skilled in interacting with AI, it is important that it can adapt to technological changes and the needs of students and the evolving job market.

Conclusion

The Prompt Engineering course aims to train highly qualified professionals to develop and improve efficient prompts, capable of generating accurate and useful answers in different contexts. With a practical and innovative methodology, the course would be structured around three major themes, extracurricular activities and a mandatory supervised internship, providing students with a complete and enriching experience.

By creating a more flexible and adaptable curriculum, Prompt’s Engineering pedagogical project could attract even more students and increase the relevance of the course in the job market. Such an approach could also help ensure graduates are equipped with the skills and knowledge needed to keep pace with technological changes and job market demands, even as some traditional professions are being threatened by AI.

However, it is important to remember that, as much as we would be investing in a highly promising and innovative field, technological evolution can change everything from one moment to another. Well, until the AIs are aware, this profession could disappear too… in any case, constant professional improvement and a flexible and agile posture of adapting to changes will always be necessary.

Copyright © 2023 AI-Talks.org

Similar Posts