Is Sora far scarier than ChatGPT?

The Daily Mail calls it terrifying. The Washington Post calls it shockingly realistic but also super weird.? Lawyers, academics and security experts are raising alarms.??

We’ve peeked at Sora, OpenAI’s new text-to-video product. It's impressive but far from perfect,? and it’s already got plenty of people worried about the abuse of AI and job losses.

Is video just a natural evolution of GenAI? Or should we be worried about deep fakes and misinformation, especially as half the world’s population will be voting in elections this year? Is it a great opportunity, or a threat to things like IP, jobs, our safety and even democracy?

Obviously, GenAI Video isn’t ready for public use yet, but I think these are fair questions.

Propagandists and election interferers will love GenAI Video. Imagine how much more quickly they can produce and spread misinformation. I think there will be challenges around authentication, and I wonder if it will dent public trust in what we see online.?

I’m glad to see tech firms taking GenAI seriously. Meta and other tech firms are creating teams and working together to prevent misuse, but Facebook and X hardly have the best track record in content moderation.?

I think we’re getting better at keeping up with the curve, if not staying ahead. ChatGPT has only been here for 15 months or so, and it feels to me like we're learning pretty quickly how to embed AI into our everyday lives and how to spot misuse.?

Even though more and more people get their news from Facebook, I think there’s a chance GenAI Video will make traditionally trusted sources like the BBC that can verify the authenticity of their content more valued. That's where I’ll go to get information, not social media.?

Who knows, maybe as trust in video content goes down, it will reaffirm the importance of face-to-face meetings and real human contact! That would surely be a good thing.?

I also worry about where Sora gets its inspiration from. Will it use the images and videos we post online? Will it use content from creators without permission? If a prompt asks for a video of a child playing at home, which child will it use? There are safeguarding issues here as well as ethical issues.?

Even if OpenAI puts in guardrails, this is open-source technology and there are already reports of Chinese tech firms building their own ‘open Sora’.??

Controlled engineering languages like those used by Microsoft are in the minority these days, and almost everything else is open source. It’s great for innovation, but where does the responsibility fall for oversight? Is it going to be YouTube or Meta? Governments and regulatory bodies? We’re starting to see some protections and governance around AI, and I hope that happens faster.??

Worries aside, I’m excited about the use cases for text-to-video. So many of the businesses I talk to are now using tools like ChatGPT and Midjourney to save money or save time and seize opportunities. It won’t be long before GenAI Video will be in the mix too.?

Will jobs be at risk? My view has always been that for every threat to a skill or a profession, there is an opportunity.?

Yes, lower-skilled roles in creative industries, and even roles in TV and film like video editing, will be at risk. Who needs makeup artists when Sora could add 20 years to an actor in seconds? US filmmaker Tyler Perry says he has cancelled plans to invest $800 million in a studio on the basis that GenAI Video will disrupt the industry, but I suspect he’ll still invest. Just in a way that makes use of this tech.?

There will be opportunities for people who embrace GenAI Video first and learn how to use it effectively. A lot of companies I meet are bringing in AI expertise to find the right AI use cases for them. It always starts with consultants but it quickly becomes full-time staff. It’s not just about prompt engineering and ChatGPT. As this week shows, GenAI is a broad area.?

There’s a whole career there for people who truly understand the implications of Large Language Modelling. There is still a skills gap in this area, and I see it as one that young people can fill.? The problem is education lagging.?

I doubt most schools and higher education institutions are moving fast enough to reflect what’s happening in GenAI, in terms of discussing it as an opportunity, rather than a tool that should not be misused. Most GenAI courses are being taught online, and most people who develop GenAI skills are doing it because they have a personal interest in AI. I’d encourage young people to see where their interest in AI can take them. There’s so much good info out there, a young person can teach themselves prompt engineering and become valuable to an employer quite quickly.??

ChatGPT is here to stay, and although I share some of the concerns around GenAI Video,? it’s just another tool we should carefully embrace. The earlier we educate people on how it can be manipulated, and how it can be safely used for good, the better.?

What did you think of Sora, and how excited or worried are you about its impact on your business, and your life?? Tell me your view in the comments below!?