Version 1.0
EDF Text Generative AI
Introduction
The syllabus outlines the knowledge the candidates need to master in order to pass the EDF Text Generative AI certification exam. It provides suggestions for preparation and highlights the benefits of taking this exam
The Effective Data Foundation (EDF)
The Effective Data Foundation (EDF) is a private collaboration that aims to promote effective use of data.
Data and data-related subjects and specializations are the focus of the Effective Data Foundation (EDF), a consortium of professionals and institutes. EDF does not limit itself to IT experts, but aims to empower all business professionals to make use of data. EDF believes that data should guide, not dictate, decisions.
Therefore, EDF values the human factor as essential, unlike organizations that claim to be data-driven.
An EDF Text Generative AI Foundation Certified professional will be able to leverage the full power of Text Generative AI. And increase their content productivity, effectiveness, and performance a thousandfold.
Certifying Organisation
Van Haren Certify
Certification levels en scope
The program exists manly out of a Foundation level for Text Generative AI. The Effective Data Foundation works closely with Van Haren Certify to ensure further development of professionals within the industry.
By becoming certified in the foundation of Text Generative AI, business professionals can master the fundamentals of Text generative AI. This will enable them to create effective and highly functional prompts that align with their business objectives while also understanding the ethical and legislative boundaries they should look out for. With this certification, professionals can gain a competitive edge in their field and stay ahead of the curve.
| Text Generative AI Foundation | The Text Generative AI Foundation Certification, which can be obtained after passing the exam. After participating a Text Generative AI, Promt Engeneer Foundation or ChatGPT for Business course. Professionals can go for the Text Generative AI Foundation Certification if they have mastered the fundamental content covered by the learning objective of the Text Generative AI Foundation Certification. This course teaches non-technical professionals about the language model and its natural language processing abilities. It covers prompt engineering, business use cases, limitations, risks, and hands-on experience using ChatGPT. Students will learn to evaluate and implement ChatGPT for real-world business problems. |
EDF Text Generative AI Foundation exam
You first need to have successfully completed the Text Generative AI Foundation exam to obtain the Text Generative AI Foundation Certificate. The exam procedure is explained in this section.
Practical information
You must pass a multiple-choice exam in which your knowledge of Agile will be tested to obtain an Text Generative AI Foundation certificate.
All exam candidates will get access to the online exam environment and will need to answer 60 multiple-choice questions within 60 minutes.
You must answer 70% of the questions correctly (or at least 42 of the 60 questions) to pass. Each question has precisely four possible answers where only one is the best answer.
You will receive the result immediately after the exam. (Digital) Access to your certificate will be given once you have passed.
Registration for the exam can be done by purchasing a participation certificate at www.vanharen.store.
| Number of questions: | 40 |
| Time (minutes) for the exam: | 60 minutes |
| % minimal passing grade | 65% |
| Open/closed book: | Closed |
| Language: | Dutch and English. |
| Exam format: | Online |
| Type of questions: | Yes. Candidates are advised to read the questions carefully. |
| Are there also negative questions included in the exam? (e.g. “which is NOT a principle off **”) | Yes. Candidates are advised to read the questions carefully. |
Levels
The Text Generative AI Foundation Certification tests candidates at levels 1 and 2, according to the Bloom Revised Taxonomy.
- Bloom Level 1: Recall & Retention
We test candidates on their ability to memorize factual information, to retain information by collecting, remembering, and recognizing specific knowledge. Knowledge includes facts, terms, answers, or terminology.
- Bloom Level 2: Understanding
We test candidates on their ability to construct meaning from oral, written, or graphical pieces of information. This is done by interpreting, summarizing, distracting, comparing, classifying, predicting, or explaining the message.
Learning objectives
In this section, you can read about how the Text Generative AI Foundation Exam is structured and which subjects you will be tested on as a candidate. It is also a tool that you can use to prepare yourself for the test.
After successfully certifying in Text Generative AI Foundation, a professionals has demonstrated that:
- He or she will be able to use a generative AI model (like ChatGPT), to create various types of content and solve different business problems.
- He or she will be able to apply prompt engineering techniques and strategies to communicate effectively with generative AI tools and optimize their output.
- And will be able to explore the features and benefits of the paid version of Generative AI applications, as well as other AI technologies and their future perspective.
The exam specifications describe the topics in the subject matter of the Text Generative AI Foundation exam, and their relative importance. Questions can be asked during the exam about the following subjects.
| Exam Requirements | Sub-modules | Weight | Bloom |
| 1.Introduction to AI |
| 10% | 1 + 2 |
| 2. Introduction to Text Generative |
| 15% | 1 + 2 |
| 3. Prompt Engineering |
| 20% | 1 + 2 |
| 4. Application of Text Generative AI in theory |
| 15% | 1 + 2 + 3 |
| 5. Application of Text Generative AI in practice |
| 5% | 1 + 2 + 3 |
| 6. Premium versions, API’s & Plugins |
| 15% | 1 + 2 |
| 7. The future of Generative AI |
| 5% | 1 + 2 |
| 8. Ethics |
| 10% | 1 + 2 |
Exam specifications
1.Introduction to AI
- What is AI?
- Where do you find it in practise?
- Two types van AI
- Why is AI important? The impact on our society
Introduction to Text Generative
- Generative AI in general
- What can ChatGPT do?
- Limitations, risks, hallucinations
- Large language models
- (Optional) Architectuur: GPT, SFT, RLHF
Prompt Engineering
- Tips and tricks for first use
- Strategies and Techniques
- Expert role
- Prompt structure
- Business field examples
- Problem setting
- Design prompt structure
- Show ChatGPT output (applied)
Applyed Tekst Generative AI
- Content generation
- Brainstorm and idea validation
- Understanding and researching complex topics
Premium versions, API’s & Plugins
- GPT 4
- Showcase ChatGPT Plugins
- Advanced Data Analysis
- API for custom AI Chatbot
The future of Generative AI
- Other NLP models
- Text-to-image
- Text-to-music
- Text-to-speech
- Text-to-video
- Text-to-3D
- Future perspective
Ethics
- Biases in AI algorithms
- Unemployment due to automation
- Privacy and security
- Misinformation
- Copyright
Key terms and concepts
The Generative AI Fundamentals Certification uses several key terms, concepts, and definitions in the list below. You can use these definitions to support and clarify topics related to the exam. Pay attention! If you only learn these terms, then you are often not sufficiently prepared to pass the exam.
| Key term | concept |
| Generative AI: | Generative AI: A branch of AI that enables a computer to create new and realistic artifacts, such as images, video, music, speech, text, software code and product designs, based on existing data. |
| Natural Language Processing (NLP): | Natural Language Processing (NLP): A subfield of artificial intelligence that focuses on the interaction between computers and humans using natural language. |
| Artificial Intelligence (AI): | Artificial Intelligence (AI): A branch of computer science that aims to create machines that can perform tasks that typically require human intelligence, such as learning, reasoning, problem-solving, and perception. |
| Machine Learning (ML): | Machine Learning (ML): A subset of AI that focuses on developing algorithms that can learn and improve from data, without being explicitly programmed. |
| Supervised Learning: | Supervised Learning: A machine learning approach where the model is trained on labeled data, where the output is known, and the goal is to learn a general rule that maps inputs to outputs. |
| Reinforcement Learning: | Reinforcement Learning: A type of machine learning where an AI model learns from feedback given by humans in order to improve its performance. |
| ChatGPT: | ChatGPT: A language model developed by OpenAI that uses natural language processing (NLP) to generate human-like responses to text prompts. |
| Proximal Policy Optimization (PPO): | Proximal Policy Optimization (PPO): A form of reinforcement learning used in the ChatGPT model. |
| Prompt: | Prompt: A text input given to the ChatGPT model to generate a response3. |
| Prompt Engineering: | Prompt Engineering: The process of creating high-quality prompts that result in accurate and appropriate responses from ChatGPT. |
| Use Case: | Use Case: A practical application of ChatGPT in a business or real-world scenario. |
| Limitations: | Limitations: The potential drawbacks and shortcomings of ChatGPT, including accuracy and bias issues. |
| Risks: | Risks: Potential negative consequences of using ChatGPT, such as misinformation or ethical concerns. |
| Ethical Guidelines: | Ethical Guidelines: A set of rules and principles to ensure the responsible and ethical use of ChatGPT and other AI technologies4. |
| Foundation Models | Foundation Models: Large-scale AI models that are trained on a broad set of unlabeled data and can be used for different tasks with additional fine-tuning12. |
| Generative Pretrained Transformer (GPT): | Generative Pretrained Transformer (GPT): A family of foundation models that use deep neural networks to generate natural language texts. |
| DALL·E | DALL·E: A generative AI model developed by OpenAI that can create images from text descriptions using a GPT architecture. |
| Text-to-image: | Text-to-image: A generative AI technique that converts natural language texts into images . |
| Text-to-music: | Text-to-music: A generative AI technique that converts natural language texts into music or lyrics . |
Exam regulations
General rules
An Text Generative AI Foundation certification via the Agile Consortium is an honorary title, and fraud is not tolerated. Your exam will be immediately rejected if fraud is found to have been committed during or after completion of the exam. As a result, you will not be reimbursed for your examination fees.
If you fail to pass the exam, you will not receive a certificate. This also means that you must purchase and take a new exam for your certification. Every candidate only gets one attempt per exam to succeed.
Sharing of exam questions is illegal
It is not allowed to share exam questions with others or make them public. This is a violation of the copyright and IP of the Agile Consortium and Van Haren Learning Solutions. Doing so can lead to legal action by Van Haren Learning Solutions with potentially harmful consequences.