Module 3 Final Lesson
You have covered AI fundamentals, prompting, research and content creation. This lesson completes your toolkit with the full landscape of AI visual and video tools available to you as a VA right now.
1What Is It?
The AI Visual Landscape: Four Categories You Need to Know
Open with the big picture: there are now four distinct categories
When people say "AI visuals," they usually mean one thing: generating images. But in 2025, that is only one quarter of the story. There are now four distinct categories of AI visual creation tools, and each serves a different purpose for your clients.
Category 1 is image generation: tools that turn a text description into an original still image. Category 2 is AI-assisted design: platforms that help you build branded graphics, layouts and marketing materials using AI features. Category 3 is video generation: tools that create short video clips from text prompts or images, a capability that arrived at scale in 2024 and 2025. Category 4 is AI avatar and presenter tools: platforms that let you or your client appear on screen without filming yourself, or clone a voice and likeness for consistent branded video content.
By the end of this lesson, you will know which tool sits in which category, what it is best for, and exactly how to prompt it.
Category 1 is image generation: tools that turn a text description into an original still image. Category 2 is AI-assisted design: platforms that help you build branded graphics, layouts and marketing materials using AI features. Category 3 is video generation: tools that create short video clips from text prompts or images, a capability that arrived at scale in 2024 and 2025. Category 4 is AI avatar and presenter tools: platforms that let you or your client appear on screen without filming yourself, or clone a voice and likeness for consistent branded video content.
By the end of this lesson, you will know which tool sits in which category, what it is best for, and exactly how to prompt it.
Why this matters for your VA career
Clients need visual content every single day. A VA who can navigate all four categories can offer a complete visual content service, from still graphics to branded video to AI presenter clips, without outsourcing to a separate designer or videographer.
2Why Do I Need It?
The Tools: A Complete Map
Walk through all four categories with the tool grid visible
Let us go through each category. I will name the tools, tell you what they do, and flag which ones are most practical for you as a VA in Nigeria right now.
🖼 Category 1: Image Generation
Midjourney
Highest quality artistic images
- Photorealistic and fine-art styles
- Runs via Discord or web app
- Unmatched aesthetic quality
- Best for premium brand visuals
$10/moBest quality
DALL-E 3
In ChatGPT, easy for beginners
- Generate images inside ChatGPT
- Excellent text within images
- Understands long, complex prompts
- Included in ChatGPT Plus
Most accessible
Gemini (Google)
Imagen 3 inside Google ecosystem
- Generate images directly in Gemini
- Strong at diverse, realistic people
- Integrates with Google Workspace
- Free tier available
Free tierGoogle native
Grok AI (Aurora)
xAI's image generator via X/Grok
- Aurora model for image generation
- Fewer content restrictions than rivals
- Access via Grok app or X Premium
- Fast generation speeds
X PremiumFlexible
Adobe Firefly
Commercially licensed, client-safe
- Legally safe for commercial client use
- Generative fill inside Photoshop
- Background removal and extension
- Free credits available
Client-safe
Canva AI
Design and image generation together
- Text to Image inside Canva editor
- AI backgrounds and photo editing
- Magic Expand and Magic Write
- Best all-in-one tool for VAs
Best for VAs
🎬 Category 2: AI Video Generation
Google Veo 3 / Flow
State-of-the-art video generation
- Generate cinematic video from text
- Flow is Google's AI filmmaking tool
- Veo 3 produces native audio in video
- Available via Google AI Ultra plan
Google AI2025 leader
Higgsfield AI
Social-first short video from images
- Animate photos into video clips
- Cinematic motion effects and styles
- Portrait-first for social media
- Best for transforming client photos
Social video
Grok (Video)
Text-to-video via X platform
- Short video generation from prompts
- Accessible via Grok AI app
- Growing video generation features
- Integrates with X/Twitter posts
X Premium
Kling AI
High-quality text and image to video
- Realistic motion and physics
- Animate product images for ads
- Longer video durations (up to 2 min)
- Free tier with watermark
Product video
RunwayML
Professional AI video editing
- Gen-3 video generation model
- Video-to-video style transfer
- Background removal from video
- Used by professional filmmakers
Pro grade
🧑💻 Category 3: AI Avatar and Presenter Tools
HeyGen
AI avatars that speak your script
- Type a script, get a speaking video
- Clone your own voice and likeness
- 100+ pre-built avatar choices
- Used for client explainers and ads
Top pickVA use case
Synthesia
Corporate training and explainer videos
- Professional presenter avatars
- Branded templates and layouts
- Multi-language voice and lip sync
- Preferred by corporate clients
Corporate
D-ID
Animate any photo into a speaking avatar
- Upload any photo, it speaks
- Great for brand spokesperson content
- API available for automation
- Free tier for short videos
Free tier
🧠 Category 4: Multimodal AI Assistants (Create AND Analyse)
Gemini (Google)
Generate, analyse and describe images
- Upload an image, ask questions about it
- Analyse client briefs from photos
- Generate images via Imagen 3
- Works inside Google Docs and Gmail
Free tierGoogle native
ChatGPT (GPT-4o)
Image generation and visual analysis
- Describe an image, get feedback
- Generate with DALL-E 3 inline
- Read text from uploaded images
- Analyse charts and screenshots
Most versatile
Grok AI (xAI)
Real-time data plus image analysis
- Image generation via Aurora model
- Upload and analyse images in chat
- Live web and X post awareness
- Strong for trend-aware content
X/Grok appReal-time
Commercial licensing: know before you deliver
Not every AI-generated image or video is safe for paid client use. Adobe Firefly is specifically built for commercial use. Midjourney requires a paid plan for commercial rights. DALL-E, Canva AI, and Gemini allow commercial use under their standard terms. Always check the tool's terms of service before delivering AI-created content to a paying client.
3Show Me
How to Prompt Each Category of Tool
Teach prompting as distinct skill for each category
Each category of visual tool has its own prompting logic. Let me walk through the approach for each one.
Image generation prompts: 4-element structure
Video generation prompts: add motion layer
Video prompts need a motion description added to the image formula
Start with the same 4 elements as image prompts, then add a fifth: motion description. Specify camera movement (slow zoom in, pan left, tracking shot), subject movement (walking forward, hair moving in breeze), and timing. Example: "A Nigerian entrepreneur walking into a bright modern office, slow-motion, camera tracking smoothly from the side, golden morning light through windows, cinematic, high quality, 6 seconds."
Avatar tool prompts: write as a script, not a description
HeyGen and Synthesia take scripts, not image prompts
For avatar tools, you choose an avatar and then write a speaking script. The AI lip-syncs the avatar to your script. Focus your energy on writing a clear, natural-sounding script for the avatar to deliver. Keep sentences short. Write the way someone speaks, not the way someone writes. Avoid commas mid-sentence: use full stops so the AI knows where to pause.
Quick decision guide: which tool for which job
| If the client needs... | Use this tool | Notes |
|---|---|---|
| A hero image for their website or brand | Midjourney or DALL-E 3 | Highest quality stills |
| A social media graphic with their brand fonts | Canva AI | Design and image in one platform |
| A short animated clip for Instagram or TikTok | Higgsfield or Kling AI | Portrait-first social video |
| Cinematic video from a text description | Google Veo 3 / Flow | Best text-to-video quality in 2025 |
| A talking-head video without filming anyone | HeyGen or D-ID | Type script, avatar delivers it |
| A training or explainer video for corporate client | Synthesia | Professional avatar, branded templates |
| Analysing a competitor's ad or visual for feedback | Gemini or ChatGPT | Upload the image and ask questions |
| A trend-aware visual tied to current events | Grok AI | Real-time web awareness built in |
| An image safe for commercial client use legally | Adobe Firefly | Commercially licensed from the ground up |
4Let's Do It
Live Demo: Three Tools, One Client Brief
Demonstrate the workflow: same brief, three different tool outputs
Here is a real scenario. A client sells natural skincare products and wants a week of visual content for her Instagram page. Her brand is natural, feminine, and Afro-luxe. Colour palette: terracotta and cream. Audience: professional Nigerian women aged 25 to 40.
Output 1: Still brand image (Canva AI or DALL-E 3). Prompt: "A collection of handmade natural skincare products including a clay face mask jar and a shea butter tin, arranged on a terracotta ceramic tray with dried flowers, overhead flat lay, warm terracotta and cream tones, luxury beauty brand aesthetic, high resolution, clean negative space on the right for text overlay."
Output 2: Animated Instagram Reel clip (Higgsfield). Take the still image from Output 1 and animate it: "Cinematic slow zoom into the product arrangement, soft bokeh, warm golden light gradually brightening, 6 seconds, social media vertical format."
Output 3: Talking-head product intro video (HeyGen). Choose a natural-looking avatar. Script: "Hi! I am so excited to share these with you. This is our new Adire Shea Collection. Pure, natural, and made with you in mind. Link in bio to shop." Avatar delivers it on camera. No filming required.
Total time for all three outputs: approximately 35 to 45 minutes. Result: a full week of varied visual content across three formats, produced from one brief, using three different AI tools.
Output 1: Still brand image (Canva AI or DALL-E 3). Prompt: "A collection of handmade natural skincare products including a clay face mask jar and a shea butter tin, arranged on a terracotta ceramic tray with dried flowers, overhead flat lay, warm terracotta and cream tones, luxury beauty brand aesthetic, high resolution, clean negative space on the right for text overlay."
Output 2: Animated Instagram Reel clip (Higgsfield). Take the still image from Output 1 and animate it: "Cinematic slow zoom into the product arrangement, soft bokeh, warm golden light gradually brightening, 6 seconds, social media vertical format."
Output 3: Talking-head product intro video (HeyGen). Choose a natural-looking avatar. Script: "Hi! I am so excited to share these with you. This is our new Adire Shea Collection. Pure, natural, and made with you in mind. Link in bio to shop." Avatar delivers it on camera. No filming required.
Total time for all three outputs: approximately 35 to 45 minutes. Result: a full week of varied visual content across three formats, produced from one brief, using three different AI tools.
5Now You Try
The Multi-Tool Visual Challenge
Run a 30-minute exercise using at least two categories of tool
You have 30 minutes and one brief: a wellness coach in Abuja who works with professional women on stress management and mindset. Brand personality: calm, modern, and grounded. Colours: sage green and warm white.
Task A: Generate one still brand image using DALL-E (in ChatGPT) or Canva AI. Save the prompt you used.
Task B: Take any image (your generated one, or a stock photo) and try animating it using Higgsfield or Kling AI. Describe the motion you want. Save the clip.
Task C: Open HeyGen (free trial). Choose an avatar. Write a 3-sentence script for the coach's brand intro. Generate the video.
At the end: save everything into a Google Drive folder called "Lesson 12: Multi-Tool Output." This is your first multi-format visual content package.
Task A: Generate one still brand image using DALL-E (in ChatGPT) or Canva AI. Save the prompt you used.
Task B: Take any image (your generated one, or a stock photo) and try animating it using Higgsfield or Kling AI. Describe the motion you want. Save the clip.
Task C: Open HeyGen (free trial). Choose an avatar. Write a 3-sentence script for the coach's brand intro. Generate the video.
At the end: save everything into a Google Drive folder called "Lesson 12: Multi-Tool Output." This is your first multi-format visual content package.
The key principle of prompting for visuals
Specificity always wins. The more detail you give any visual AI tool, the closer the output is to what you imagined. Vague prompt, vague result. Specific prompt, specific result. The prompt is the brief. Treat it like one.
6Your Task
Lesson 12 Assignment
This lesson's assignment
Create a 4-piece multi-format visual content set for a real or imaginary client, using at least 3 different tools from at least 2 different categories.
- 1
Choose a client. Write a one-paragraph brief: business, brand personality, colour palette, audience, and the 4 pieces of visual content you will create (at least one still image, one video or animated clip, and one piece from a category of your choice).
- 2
Create each piece using the most appropriate tool from today's lesson. For each: write the prompt you used, note which tool you used, and save the output. Document everything in a Google Doc called "Visual Creation: Lesson 12."
- 3
Take your strongest still image into Canva. Add the client's brand name and a short call to action. Export at the correct dimensions for the intended platform. Save as a PNG.
- 4
In your Workbook, write 2 sentences on each tool you used: what it did well and what you would change next time.
- 5
Add the 5 prompts from this lesson's Prompt Library to your AI Prompt Library Google Doc under the heading "Lesson 12: AI Visual and Video Creation."
7What Did You Create?
Module 3 Complete: You Now Have a Full AI Toolkit
Close Module 3 with a summary of the complete AI skillset they have built
Look at where you started at the beginning of Module 3. You knew AI existed. You may have used ChatGPT once or twice. Now think about what you have today.
You understand what AI is and how it thinks. You can write a structured prompt for any task using the RCTF framework. You can research any topic and produce a client-ready brief within an hour. You can create platform-specific written content for six channels. And now you can produce professional visual content across four categories: still images, branded graphics, animated and cinematic video clips, and AI avatar presenter videos.
That is not beginner knowledge. That is the skillset of a well-rounded digital creative professional. Module 4 takes all of it into building and running an actual VA business. You are ready.
You understand what AI is and how it thinks. You can write a structured prompt for any task using the RCTF framework. You can research any topic and produce a client-ready brief within an hour. You can create platform-specific written content for six channels. And now you can produce professional visual content across four categories: still images, branded graphics, animated and cinematic video clips, and AI avatar presenter videos.
That is not beginner knowledge. That is the skillset of a well-rounded digital creative professional. Module 4 takes all of it into building and running an actual VA business. You are ready.
- At least one still image generated with a documented prompt
- At least one video or animated clip created
- At least one HeyGen or avatar video produced
- One Canva branded graphic exported at correct dimensions
- "Visual Creation: Lesson 12" Google Doc saved in Drive
- Tool evaluation notes written in Workbook
- Lesson 12 prompts added to AI Prompt Library
- Module 3 complete: AI fundamentals, prompting, research, content, and visuals
WWorkbook
Lesson 12 Student Workbook
Instructions
Complete all 5 parts. The tool-evaluation section (Part D) is the most important: recording what each tool did well and where it fell short is how you build the instincts to choose the right tool for every future brief.
📌 Part A: Tool Categories
Name the 4 categories of AI visual tools from this lesson:
Category 1:
Category 2:
Category 3:
Category 4:
Write one tool from each category that you would most likely use for a client, and why:
Cat 1 pick:
Cat 2 pick:
Cat 3 pick:
Cat 4 pick:
🖼️ Part B: Image Prompt Practice
Write a complete image prompt for a client of your choice. Label all 4 elements in square brackets:
Format: [SUBJECT] ... [STYLE] ... [MOOD] ... [TECHNICAL] ...
Now adapt it as a video prompt by adding a motion description:
Add: camera movement, subject movement, duration, format (vertical/horizontal)
🎬 Part C: HeyGen Avatar Script Practice
Write a 3-sentence HeyGen script for a client's brand intro video. Remember: write for speaking, use full stops for pauses, keep it natural:
Line 1:
Line 2:
Line 3:
🔬 Part D: Tool Evaluation (Assignment)
For each tool you used in the assignment, write one strength and one limitation you discovered:
Tool 1:
Strength:
Limitation:
Tool 2:
Strength:
Limitation:
Tool 3:
Strength:
Limitation:
🎓 Part E: Module 3 Reflection
What is the single most useful AI skill you gained from Module 3 as a whole?
Name one specific service package you could now offer to clients using the AI visual tools from this lesson. What would you include and what would you charge?
TTask
Build a 4-Piece Multi-Format Visual Content Set
Time: 60 to 90 minutes
Output: 4 pieces of visual content across at least 3 tools and 2 categories, plus a documented Google Doc and one branded Canva graphic.
Part 1: Write the Client Visual Brief
- 1
Choose a client (real or imaginary). Open a new Google Doc titled "Visual Creation: Lesson 12." At the top, write the brief: business name, what they sell, brand personality (3 adjectives), colour palette, target audience, and the 4 content pieces you will produce.
- 2
Create 4 sub-sections in the doc: "Piece 1 (Still Image)," "Piece 2 (Video / Animation)," "Piece 3 (Avatar / Presenter)," and "Piece 4 (Your Choice)." Each section will hold the tool you used, the prompt or script, and a note on the result.
Part 2: Create Piece 1, Still Image
- 1
Write your 4-element image prompt (subject, style, mood, technical). Label each element. Paste it into your Google Doc.
- 2
Run the prompt in DALL-E (ChatGPT), Canva AI (Text to Image), Gemini, or Adobe Firefly. If the first result needs improvement, adjust one element at a time. Note what you changed.
- 3
Download the best result and save it to a Drive folder called "Lesson 12 Visuals."
Part 3: Create Piece 2, Video or Animated Clip
- 1
Open Higgsfield, Kling AI, or Google Veo 3 (if available). Upload your still image from Piece 1, or write a text-to-video prompt.
- 2
Add a motion description to your prompt: camera movement, subject movement, duration, and vertical or horizontal format. Generate the clip.
- 3
Save the video clip to your "Lesson 12 Visuals" folder. Note the tool and prompt in your Google Doc.
Part 4: Create Piece 3, Avatar Presenter Video
- 1
Open HeyGen or D-ID (both have free trials). Select an avatar that fits the client's brand: professional, approachable, diverse. Note: HeyGen has a strong selection of African avatars.
- 2
Write a 3 to 5 sentence script for the avatar. Write it for speaking: short sentences, natural language, full stops for pauses. The avatar will lip-sync to this script.
- 3
Generate the video. Download and save to your "Lesson 12 Visuals" folder. Paste the script into your Google Doc under "Piece 3."
Part 5: Finish in Canva and Deliver
- 1
Take your best still image into Canva. Create a design at the correct dimensions for the intended platform (1080x1080 for Instagram square, 1920x1080 for YouTube thumbnail, 1200x628 for LinkedIn post, etc.).
- 2
Add the client's brand name, a tagline or call to action, and any brand colour accents. Make sure all text is clearly legible against the background image.
- 3
Export as a high-resolution PNG and add it to your "Lesson 12 Visuals" folder as your fourth and final piece.
🎓
Module 3 Complete!
You have built a complete AI creative toolkit spanning images, video, avatar content, and branded design. Module 4 takes all of this into the business of being a VA.
AIPrompts
Lesson 12 Prompt Library
Add to your Prompt Library Google Doc
Open your "My AI Prompt Library" document, add the heading "Lesson 12: AI Visual and Video Creation," and paste all five prompts below.
AI Visual and Video Creation Prompts
Prompt 1: The universal image prompt template (use for Midjourney, DALL-E, Gemini, Grok Aurora, Canva AI, Firefly)
"[SUBJECT: describe who or what is in the image with full detail, ethnicity, clothing, setting, and specific objects] [STYLE: name a visual reference, e.g. 'editorial style inspired by Forbes Africa magazine', 'luxury brand campaign aesthetic', 'fine art documentary photography'] [MOOD: describe the emotional atmosphere, e.g. 'confident and aspirational', 'calm and grounded', 'celebratory and vibrant'] [TECHNICAL: specify orientation (portrait/landscape/square), lighting type (natural/golden hour/studio), composition (overhead flat lay/close-up portrait/wide environmental shot), and resolution (high resolution, photorealistic)]"
Use this for every still image you generate. Fill all four sections before running it. More detail in every section produces better, more specific results.
Prompt 2: Video / animation prompt for Higgsfield, Kling AI, or Google Veo 3
"[SUBJECT: describe the main subject in detail] [SCENE: describe the environment and setting] [MOTION: describe camera movement (slow zoom in / pan left / tracking shot), subject movement (walking forward / hair blowing in breeze / slow head turn), and any other animation] [MOOD: describe the atmosphere and emotional tone] [TECHNICAL: duration (6 seconds / 10 seconds), orientation (vertical 9:16 for social media / horizontal 16:9 for YouTube), quality (cinematic, high quality, smooth motion, 4K)]"
The motion description is what separates a video prompt from an image prompt. Always specify both camera and subject movement. Start with 6-second clips and extend once you are happy with the style.
Prompt 3: HeyGen avatar script template for a client brand intro video
"Hi! I am [client name or brand spokesperson name], and I am so glad you are here. At [business name], we believe that [core brand value or mission in one sentence]. Whether you are looking for [product/service benefit 1] or [product/service benefit 2], we have got something made just for you. Come and explore. I promise you will love what you find."
This is a script template for the HeyGen text-to-speech field, not a prompt for image generation. Fill in the bracketed sections for each client. Keep each sentence under 15 words for natural delivery. Paste it directly into HeyGen's script box.
Prompt 4: Gemini or ChatGPT prompt for analysing a competitor's visual or ad
"I am uploading an image of a competitor's social media post / advertisement / website screenshot for [my client's industry]. Please analyse it and tell me: (1) What visual style and design choices are being used? (2) What emotions or values does this image communicate? (3) Who appears to be the target audience based on the visual choices? (4) What is working well that my client could learn from? (5) What could my client do differently to stand out from this in the same space?"
Upload the competitor image and paste this prompt into Gemini or ChatGPT GPT-4o. This gives your client a free competitive visual audit that would cost thousands of naira from a marketing consultant.
Prompt 5: Canva AI Magic Write prompt for social media caption to accompany a generated image
"Write a social media caption for [platform: Instagram / LinkedIn / Facebook] to accompany a [describe the image, e.g. 'product flat lay of natural skincare items on a terracotta tray']. The brand is [brand name]. Their audience is [target audience description]. The tone should be [e.g. warm and aspirational / professional and authoritative / fun and relatable]. The post goal is [e.g. to drive traffic to the website / to build brand awareness / to announce a new product]. Include a strong opening line that stops the scroll, 2 to 3 body sentences, a clear call to action, and 5 relevant hashtags."
Use this in Canva's Magic Write feature or in ChatGPT alongside your visual creation workflow. Pairing a strong visual with a purpose-written caption turns a single AI output into a complete ready-to-publish social media post.