1. What is generative AI?

Artificial Intelligence (AI) is a broad term for software that can perform tasks that normally need human thinking. Most traditional AI is discriminative — it looks at existing content and classifies or predicts. For example, a spam filter reads your email and decides "spam" or "not spam". It does not create anything new.

Generative AI is different. Instead of just classifying things, it creates brand-new content — text, images, audio, video, and more — that did not exist before. Give it a short instruction (called a prompt) and it produces something original.

Think of it this way: traditional AI is like a judge in a music competition who scores existing songs. Generative AI is like the musician who composes a completely new song on demand.

Generative AI can create an impressive range of content types. Here are the main ones you will practise in this course:

• Text — marketing copy, social-media captions, product descriptions, emails, scripts, reports, translations.

• Images — product photos, logos, social-media graphics, illustrations, and even photorealistic scenes from scratch.

• Audio — voice-overs for ads, background music, podcast intros, and spoken narration.

• Video — short promotional clips, animated explainers, and AI-generated talking-head videos.

In Ethiopia, businesses are already using these capabilities. A small clothing shop in Mercato can use generative AI to write a punchy TeleBirr payment-reminder message in Amharic. A startup in Addis Ababa can generate a professional product banner without hiring a graphic designer. A radio producer in Gondar can create a jingle in minutes rather than days.

How does generative AI produce such convincing content? At a high level, the process has three stages:

1. Training — Engineers feed the model enormous amounts of data: billions of web pages, images, audio recordings, and more. The model learns the statistical patterns of how words, pixels, and sounds fit together.

2. The prompt — You give the model an instruction. This is your starting point. The quality of your prompt directly affects the quality of the output. You will practise writing strong prompts throughout this course.

3. Generation — The model uses the patterns it learned during training to produce new content that is statistically consistent with your prompt. It does not copy existing content; it creates something new by predicting what words, pixels, or sounds should come next.

The most important generative AI models today include large language models (LLMs) like ChatGPT and Claude for text, diffusion models like DALL·E and Midjourney for images, and audio/video models like ElevenLabs and Sora.

With great creative power comes real responsibility. Using generative AI ethically and safely is not optional — it is a professional necessity.

Here are the key ethical principles you will apply throughout this course:

1. Transparency — If you used AI to create content for a client or the public, disclose it where required. Passing AI work off entirely as your own can mislead clients.

2. Copyright and originality — Generative AI is trained on existing content. The output may resemble real artists, photographers, or writers. Before publishing, verify you have the legal rights to use the output commercially.

3. Accuracy — AI can confidently produce false facts (called hallucinations). Always fact-check text output before publishing, especially for health, legal, or financial topics.

4. Bias and representation — AI image models may produce stereotyped or Eurocentric visuals by default. For Ethiopian brands, add explicit cultural context in your prompts to get representative output.

5. Brand safety — AI can generate content that is off-brand, offensive, or inappropriate. Always review output before publishing and establish a human approval step in your workflow.

Generative AI is most powerful when it is part of a thoughtful human-led workflow — not when it replaces human creativity entirely.

A practical content-creation workflow looks like this:

1. Human defines the goal — What is the message? Who is the audience? What platform is this for?

2. Human writes or refines the prompt — The better your prompt, the better the AI output.

3. AI generates a draft — Fast, cheap, and often surprisingly good.

4. Human reviews and edits — Check facts, fix tone, ensure brand safety, and add the human touch.

5. Human approves and publishes — Final quality control stays with the human.

In this course, you will move through all five steps for text, image, audio, and video content. By the end, you will be able to produce professional-grade content far faster than before — without sacrificing quality or ethics.