Five AI tool creators should know

Currently, the most demanding application of AI video technology is text-to-video AI. The idea is that anyone would be able to type a hint into a text field, and within seconds – or, more likely, minutes – a video would be generated using the AI ​​model. Some big companies of the tech world are doing research on this. meta announced make videoGoogle has announced imagen video And fenakiAnd NVIDIA Says this is also getting involved.

Not all of those projects are available for public use, and the output of these systems is far from photorealistic: figures are distorted and transformed, and there are obvious changes between frames. But demos are public statements of research and development in the text-to-video field from large and powerful companies.

As a creative, the thought of my job being replaced by AI so soon is horrifying, but as a techie, I’m curious. I wanted to see how capable we are of typing text into a website and getting better video. So in the latest episode of my YouTube series full frameI tested five AI video creation tools – Gen-2, kyber, , synesthesiaAnd boring – To see what the limits of our AI technology currently are.