Facebook parent company Meta just released an AI system that automatically generates video clips of anything you can imagine—and all it needs from you is a simple text prompt.
It’s called Make-A-Video, and it’s part of a category of tools called “generative AI” that businesses need to be paying attention to.
What it is
Meta’s Make-A-Video is an AI system that automatically generates video clips from text prompts. That means you type in a prompt (say, “a horse drinking water”), and the AI system automatically generates a video of that very thing. It can even generate the video in different styles (things like “surreal” or “stylized”).
- Once the system makes you a video, you can even generate variations of the ones you like for near-unlimited creative possibilities and variations.
- Today, Make-A-Video generates short snippets without audio, though we expect that to change as the technology progresses.
- Make-A-Video is not yet publicly available, but we expect that to change in due time, in a similar fashion to other generative AI tools like DALL-E 2, which went through a public testing period before being commercially released.
Why it matters
In Episode 20 of the Marketing AI Show, Marketing AI Institute founder/CEO Paul Roetzer and myself discussed why Make-A-Video matters for every person in marketing and business today.
- You will soon be able to create almost anything you want using AI. “What’s happening is we’re moving towards this ‘text-to-everything’ environment, where we’re going to be able to prompt a machine to create anything we can imagine,” says Roetzer. That means expect a world where anyone, regardless of skill level, can create videos, images, text, code, and much more simply by typing into a prompt window using the words they already know.
- The technology to do so is already here. “It’s really important for people to realize we’re not talking about 3-5 years out,” says Roetzer. “ We are talking about stuff that is here right now.” One example: Today, any marketer of any skill level can go use a machine to write a blog post paragraph-by-paragraph on any topic, adding AI-generated images as they go.
- Marketers and business leaders need to pay attention to this now. It’s not just about video. Or content. Or images. Recent rapid advances in AI language models already enable futuristic-sounding use cases that most marketers and leaders are not paying attention to. Things like: Chatbots that perfectly mimic celebrities (or your customers). Synthetic versions of people (such as your CEO) that engage with humans in a completely realistic way. “There are zero technical obstacles to doing these things today,” says Roetzer.
What to do about it
You can’t yet use Meta’s Make-A-Video, but you need to start understanding the category of AI tools it belongs to: generative AI. Here are a few ways to do that.
Check out the full episode:
If you want to learn more about this and other important AI topics in the news, check out the full episode of the Marketing AI Show Episode 20: The AI Bill of Rights, Building AI-Driven Companies, and Meta’s Make-a-Video.