SORA AI’s Problems [And Solutions]

For everyone that points out all the inaccuracies and glitches, this is the worst it will look ever again.

I'm not worried about AI, I 'm worried about what humans will use it for... The same as I don't fear a knife, I fear the one holding it

Today: Crank out your own novel in one day. Tomorrow: Turn that novel into a full-length movie in one day.

As an aspiring videographer, it's tough to explain. Over my years of camping and traveling, I've had some STUNNING visual experiences- sunrises on flooded salt flats, yelling from clifftops at the end of a hike, walking along beautiful valley streams. The idea that someday the beauty and rarity of those moments will be washed away and oversaturated by ai is tragic.

The more AI advances the more im compelled to go outside and touch some grass

What I find crazy is when the AI does get it wrong, like with the chair, unlike things going wrong in VFX before - in that you can clearly see its an error - the AI breaking kinda breaks your brain as it morphs so damn well into the content that its within. There are no lines, no borders, no artifacting or anything. In blends so damn well in the scene and whilst its morphing you start to question what you're seeing. Its like your brain wondering what on earth the reality is that its seeing before you.

This news makes me feel like I'm standing on a beach at sea. All the water starts to recede and some of us realize that there will be a huge tsunami next, and some of us run to pick up the shells that the water used to cover. It scares me and I hope I'm just wrong.

9:40 your voice is just that good that people think it's AI

Problem is, once you have a reliable AI video detector, you can also use it to make the AI better in turn. It's an arms race in the end.

AI is advancing at a scary fast pace

It’s interesting how history repeats itself on some level. When photography came on the scene, many people at the time said photography was not art. After all, the only talent was pushing a button on a machine. Anyone could do that, and the true artists were painters, sculpters, etc. Now fast forward to simply typing a prompt into a computer and it creates a video. Anyone can do that. Ansel Adams was not just pushing a button on a camera and getting lucky. I suspect these tools will be much the same. The learning curve will be steep to get the output that you want. It may be easy to generate beautiful videos. Creating something that I meaningful is another matter.

I’ve watched 10 video commentaries about SORA on YouTube in the past two days and found your feature here is one of the very best with comprehensive knowledge about the subject as well as offering your personal opinion.

Regardless of anyone's opinion, this technology is undeniably impressive. Three years ago, I could never have imagined that something like this would even exist.

Interesting how Shutterstock partnered with OpenAI, considering that stock footage is going to be on life support soon.

I’m a graphic designer and yes, today I used AI to assist with the generation of a creative assets I needed to make as part of my work. However, the final work is not just the output of AI, it just served to play a role in the formation of the final visual asset. My ‘human’ component brought things together in a very intentional and creative way, to meet a design brief, as in most instances AI is just not ready to produce a fully realized output yet, with all the ‘real world’ factors considered. However, once that line is crossed where AI can produce, curate and collate a valid result better than a skilled and experienced human… we have a problem! What’s important is that channels like Coldfusion are bringing awareness these problems BEFORE they reach saturation, so we have time to plan, consider, and hopefully prepare for a better future where we are not all victims of the ‘Because we could’, and fall a little more into the ‘Because we should’.

One of my favorite channels asking all of the right questions... Thank you!

The transition at 4:13 from the combination of both videos is absolutely incredible!

The thing that blows my mind is that the system literally starts with a bunch of unrecognizable noise and just gradually removes the noise to reveal the final image(s). I never in a million years would have thought to do it this way. Pure genius.

Great meeting you in Sydney Dagogo and great video as usual, I was the guy who mentioned I just started a new YouTube channel 😊

Absolutely nuts how this is all progressing. Great vid