At BearJam, we’ve been working together. AI in video production The last few years. What started as a tool for small quick fixes has evolved into a complete medium of its own. Not only have we produced completely AI-generated ads and content, but we’ve also produced a number of hybrid projects.
Over the past 18 months, AI video has gotten really good. The tool is really capable. Costs are coming down. Creative possibilities are expanding faster than most people in this industry have time to process them.
So why hasn’t it become more popular yet?
It’s not a technique. Technology works. The barrier is trust, and until the industry gets serious about it, we’ll continue to have the same circular conversation about whether AI belongs in professional production.
That’s right. There is a stumbling block here.
Why AI-generated videos still have quality issues
It’s the bad jobs that go the fastest. Strange AI hands and melted faces are shared as evidence of the technology’s limitations, but quietly effective AI campaigns centered around truth, insight, or emotion are rarely shared. Confirmation bias plays a big role.
But audiences have always embraced constructed images, from heavily manipulated product photos to elaborate practical effects. We don’t usually judge creative work by whether it’s made in-camera, digitally, or through complex production processes. Decide if it works.
The same standards should apply to AI. Tools do not determine quality. That’s the idea.
AI Copyright, Creative Ownership and Consent
Who owns the AI-generated content? Where did the training data come from? Who agreed to what?
This is a trust issue with the most legitimate foundations, and one that the industry should take seriously. The question of justice for creators whose works were used to train models without their consent remains unresolved. Until that happens, a layer of anxiety will remain, and that anxiety will delay decision-making.
It is a reasonable step to document the prompt and prove its origin. However, it has limitations if the underlying model is built on a controversial foundation.
There are even bigger questions worth sitting through. When you listen to Sam Fender on streaming platforms, he receives royalties. Could there be a world where commissioning creative work in a particular artist’s style means the artist receives something in return? I think that’s the direction it’s headed. Platforms that stay ahead of the curve, rather than waiting to be regulated, will gain more trust from the creative community and the brands that rely on them.
Mistrust of AI platforms in production
At the root of this lies broader geopolitical anxiety. The major AI video platforms are mainly American or Chinese. For brands and broadcasters with data governance obligations, big questions arise about where prompts are stored, how the output is used, and what rights apply.
It’s not paranoia. That’s due diligence. And the platforms haven’t always made it easy to find clear answers. Until there is more transparency and ideally more European alternatives, this will remain a real obstacle, especially for large and sensitive organizations.
The platform needs to improve here. Trust is not given. It’s being built. And now some of them are making it more difficult than it needs to be.
Why human creativity remains important in AI video production
Creators are understandably wary. Our professional identity is closely tied to the skills we have developed over the years. Eyes, instincts, techniques. When a tool comes along that reproduces some of that, it’s not just confusing. It feels personal.
But the creators I’ve seen succeed with AI haven’t ignored it. They are the ones who used this tool to do things that were previously impossible with their budgets, schedules, and briefs. Skills change, but they don’t disappear. Quick production, creative direction, flair: these are still very important. Perhaps now more than ever, the difference between great and mediocre AI output depends almost entirely on the humans behind it.
When it comes to job security, if you want to stop being left out for lunch, learn how to cook differently. The most public roles are those built around performing repeatable tasks. A growing number of roles are built around judgment, relationships, and creative direction. The manufacturing industry has weathered this type of disruption before. From movies to digital, broadcasting to social. Each time, the industry has adapted. This moment is bigger, but the logic is the same.
Fully AI-generated video production, hybrid or traditional video production?
One of the most useful reframings I have found is to think of AI engagement as a spectrum rather than a dualism. On one end, it is completely AI-generated from concept to delivery. The other is traditional production, where AI does not play a role. In between is the hybrid where most of the interesting work currently resides. AI-assisted concept, compositing AI-generated elements with live-action, and AI-powered versioning and localization at scale.
There are no hierarchies here. The right approach depends entirely on your brief, budget, and what needs to be done with the work. Being transparent about where a project falls on that spectrum is itself an act that builds trust. With our clients, our audiences, and the industry as a whole.
Knowing which model fits which situation is now a difficult task.
How AI is expanding access to high-quality video production
I recently had a call with a prospective client, a startup company launching a new supplement. We needed video content that covered multiple use cases, audience segments, and geographic regions. Traditional shooting that met all these requirements was out of budget. They were getting the lowest prices from overseas companies, but they wanted a partner who understood their brand and would work with them repeatedly. They were willing to pay a little more for it.
That conversation stayed with me. Because what they were describing wasn’t actually a conversation about AI at all. It was a conversation about trust.
This is an opportunity overlooked in the debate over whether AI is good or bad for the industry. AI opens doors for clients who previously didn’t have access to full-scale production. This allows ambitious creatives to be achieved even on budgets that would otherwise have been impossible. And when those relationships are built well, and you become a partner who understands your brand, thinks strategically, and delivers work that delivers, your brief will grow. What started as an AI-driven content project becomes a deeper engagement. More ambitious research will follow. In some cases, that means adding AI. In some cases, that means completely live-action productions. This tool is a starting point, not a ceiling.
Customer expectations have changed. Many brands are now actively seeking to implement AI in their pipelines. They want the scale, speed and creative ambition it unleashes. Agencies that can’t provide that will send the conversation elsewhere. But the real advantage goes beyond agencies that simply deploy tools. It is passed on to those who use them in a way that earns their trust. with clients, creative partners, and audiences.
There is a huge gap in expertise between those who are serious about this technology and those who sit on the sidelines. It won’t stay open forever.
AI has a PR problem. But it’s something that can be resolved. Agencies that do their job, build trust, and keep their clients on board won’t have that problem for a long time.
About: BearJam is an award-winning video production company based in London. The company combines traditional manufacturing expertise with AI-powered production.
