Copyright Office Clarifies Cinematic AI Copyright Rules For over two years, AI-generated content has been in a legal gray zone, leaving creators uncertain whether their work—especially in film, animation, and social media—was copyrightable. That uncertainty has hampered investment in more ambitious projects. Now, the U.S. Copyright Office has drawn a clear line: AI-generated material is protected if shaped by human creativity. “Where creativity is expressed through the use of AI systems, it continues to enjoy protection,” said Shira Perlmutter, Register of Copyrights. “Extending protection to material whose expressive elements are determined by a machine, however, would undermine copyright.” Not sure who or how this would be determined.
“In due time, Generative AI in the hands of talented ‘human authors’ will transform independent filmmaking and the entertainment industry at large, leveling the playing field for creators who’ve traditionally lacked access to major studio resources or who’ve been constrained by risk averse gatekeepers.” said John Gaeta, founder of Escape.ai, a social distribution platform for AI curated entertainment, which they call Neocinema. “We see this as a coming revolution—empowering newcomers and notable visionaries of all kinds to break through, to bring ambitious and original projects to life while maintaining the peace of mind that comes from protective copyrights.”
Pika 2.1 was launched last week, offering crystal-clear 1080p resolution, razor-sharp details, seamless motion, lifelike human characters, and more. If you see it, you’ll believe it. Director and AI Master Dave Clark used Pika to make this exciting short below.
Sebasian introduced the T2V-01-Director model from Hailuo AI which enables creators to control precise camera movements with this convincing short. “Say goodbye to complex setups and hello to intuitive direction,” he said on X. Currently Available for Early Access Testing only.
Bytedance announced INFP, an AI that can make any single image talk and sing from any audio file expressively. There are a dozen examples in the X post above, starting with making the Mona-Lisa talk.
A new video upscaling tool is coming soon. It’s called DiffVSR. There’s no open source code yet. This is a big problem in Cinematic AI today. The highest res generations are 1080 but even then, the larger you make the image, the worse it looks.
“Sure, AI can do sci-fi. But you can’t just film two people talking in a restaurant,” says Albert Bozsan, Munich-based Creative Director of Storybook Studios, which makes AI films and TV shows. This short, directed by Tobias M. Huber, is a cute date scene with a dark twist. You can see it in the running at Project Odyssey by Civitai, the largest AI film competition. “I’m proud how far we were able to get with a mix of AI platforms and our own local workflows.” Wrote Bozesan on Linkedin.
I saved the best for last. Davide Bianca is one of the most interesting filmmakers working in Cinematic AI today.
