In early 2024 OpenAI demonstrated Sora which was the first real "text to video" AI system. It was not released for public use for a long while allowing many others to develop similar systems. Now, in 2025, there are many ways for ordinary users to create AI video at reasonable cost, even for free.
[A] OpenAI Sora - video from text. openai.com/index/sora. It is worthwhile visiting the website to view the examples. Follow the link above and scroll down the page.
You can now generate still images with SORA with a free ChatGPT.com account but you need to purchase one of their plans to generate video.
This is the image generation page from Sora, with two versions of an image generated. Those with a paid plan can then generate a video based on the image.
[B] Grok Grok does not yet have a very sophisticated interface but it does produce interesting images and then turn them into 5 second videos for free. If you put Grok on your phone or tablet then it will quickly turn any photo there into a video.
Simple animation for free from barmaid image - all by Grok pic.twitter.com/wTb7K3rHwm
— The Silver AI Project - free AI training materials (@SilverAIProject) September 15, 2025
[C] Midjourney www.midjourney.com is a subscription tool that is really good at image creation, with the ability to go on to produce a video if desired. Sometimes it has weak spots. Here it wouldn't provide a good picture of a barmaid serving beer but the alternative was not too bad.
I think the previous Grok barmaid video wins. I couldn't persuade Midjourney to give me a foaming pint of beer at all π€ͺ pic.twitter.com/qnUNLIJWAI
— The Silver AI Project - free AI training materials (@SilverAIProject) September 15, 2025
[D] Google Veo 3 on YouTube Announced 16 Sept 2025 with the following clip. Appears to suggest that YouTube creators will be able to generate short video clips for free. Google Veo 3 itself is a powerful video world generation tool.
Your next viral video could start with a single prompt thanks to AI. πΉ
— Google DeepMind (@GoogleDeepMind) September 16, 2025
A custom version of our Veo 3 Fast model is now available in @YouTube Shorts, generating clips with sound. Rolling out in πΊπ²π¨π¦π¬π§π¦πΊπ³πΏ#MadeOnYouTube pic.twitter.com/LY0h8YkqT6
[E] Runway is a subscription AI tool runwayml.com. Very often videos produced by AI have to be combined together and then have soundtracks of some kind added, for example music. Runway allows this and has many other powerful video editing features.
The image below is of a typical Runway timeline with four videos being combined with a music track and some text to make a music video like those linked at the www.dumdave.com music video site.