Skip to main content
Module 1: Creating with AI Images & Video

The landscape: what the tools do

Pick a tool by the job, not by 'which is best.'

AI image and video generation has exploded, and there are many tools. The key for a beginner: **pick a tool by the job you need done, not by which is 'best' — no single tool wins at everything.** (Tool names and versions change fast, so focus on the capabilities.)

AI image tools — grouped by strength:

  • Artistic / cinematic looks: Midjourney is known for beautiful, stylized, editorial and concept-art imagery.
  • Follows plain-language prompts, conversational editing: the image tools inside ChatGPT (its built-in image generation, which replaced DALL·E) and Google's Gemini/Imagen are great for "just describe it," multi-element scenes, and iterating in chat.
  • Photorealism and speed: Flux (an open model available in many apps) is fast and very realistic.
  • Legible text in images: Ideogram is best at rendering actual words, logos, and signage correctly (a classic weak spot for other tools).
  • Commercial-safety positioning + Photoshop: Adobe Firefly (more on its "commercially safe" claim in Module 2).

AI video tools — grouped by strength:

  • All-around quality with sound: Google Veo generates video with synchronized audio and strong prompt-following.
  • Value and cinematic motion: Kling is popular for social clips at lower cost.
  • Pro creative control: Runway offers timeline-style editing for ads and client work.
  • Dreamlike, accessible motion: Luma (Dream Machine).
  • OpenAI's Sora was a landmark text-to-video model that shaped the whole category (its exact availability has shifted — treat it as an influential example of the capability).

The basics of what they do:

  • Text-to-image: type a description ("prompt"), get still image(s) you can refine and regenerate.
  • Text-to-video: same idea, but the output is a short clip (usually a few seconds).
  • Image-to-video: animate a still image you provide.

Many tools do several of these.

The takeaway: the landscape is rich and shifting, so don't get hung up on "the best tool." Decide what you want to make — artistic image, photorealistic image, image with text, short video with sound — and pick a tool suited to that job. The skills you learn transfer across them.

The mindset: AI image and video tools are numerous and fast-changing, so choose by the job, not by "which is best" — there's no single winner. For images: Midjourney (artistic), ChatGPT/Gemini (plain-language, conversational), Flux (photorealism/speed), Ideogram (text in images), Firefly (commercial positioning). For video: Veo (all-around with audio), Kling (value), Runway (pro control), Luma (accessible), with Sora as the landmark that shaped the category. Text-to-image, text-to-video, and image-to-video are the basics. Focus on capabilities (which are stable) over version numbers (which churn), and the skills transfer across tools.

Try it

Match a tool to a job: think of something you'd like to create (an artistic image? a photorealistic one? an image with text like a logo? a short video?). Which category of tool suits that job best? Note that you'd pick by *what you want to make*, not by chasing 'the best' tool — and that the skills transfer across them.

Stay in the loop

Enjoying the free lessons? Get an email when we publish new courses and updates — no spam, unsubscribe anytime.

Discussion (0)

Ask a question or share what worked for you. Comments are reviewed before they appear.

Log in to join the discussion and ask questions about this lesson.

No comments yet. Be the first to start the discussion!