I tested OpenAI's Sora, and here are my thoughts

Say hello to OpenAI's Sora. 

Yesterday I upgraded to Sora Pro, which allows me to upload images of people. 

I trained it on a photo of me and gave it this simple prompt:

"Use my likeness to create a cinematic scene of me as a time-traveling secret agent, standing confidently on a rooftop overlooking a futuristic New York City. Dress me in a sharp, high-tech suit with glowing accents on the lapels and cuffs, paired with futuristic sunglasses. Show a full-body perspective with a slight upward angle, capturing drones and holographic projections of the city skyline around me. Add a stormy sky with lightning reflecting off the glowing buildings to enhance the dramatic effect."
 

🚫 Did it do everything perfectly? No. 
🚫 Did it follow all of my instructions? No. 
🚫 Did it convince me it will replace humans? No way. 

Don't get me wrong, what I produced in just under an hour from my couch is impressive. I even got it to generate clips of me in different positions, based simply on one photo prompt. The results would be even better if I trained it on 50 images of me from various angles and with different facial expressions. 
Meanwhile, the $200 per month price tag for Sora Pro (the free version bundled with the $20 per month you pay for ChatGPT Plus doesn't allow you to upload images of people), while not cheap, is peanuts when one considers the costs associated with video production and video effects. 

This technology is not there yet, but it's developing quickly. Very quickly. 

Harnass it now. It won't be long before AI-assisted technology is the norm for creative production and development. 

Have you dabbled with Sora yet? What are your thoughts? Let me know in the comments! 👇🏾👇🏾👇🏾