Every agency in the industry right now is saying some version of the same thing: “We use AI.”
What that actually means varies enormously — from genuinely integrated workflows that save hours per project, to a Canva subscription and a chatbot, dressed up in the language of innovation to justify a higher invoice.
We use AI tools. We’ve built them into our production process deliberately over the past two years. But we’re also clear-eyed about where they add real value and where the hype outpaces the reality. This post is an honest account of both.
Why This Conversation Matters for Your Business
If you’re commissioning video content in 2026, AI is going to come up in conversations with agencies. You’ll hear it used as a selling point, a justification for speed, and sometimes as a reason why pricing has changed. You deserve to understand what’s actually being described.
The honest version: AI has meaningfully changed certain parts of the production workflow. It has not replaced the creative thinking, strategic judgment, or interpersonal skills that determine whether a video actually works for your business.
Where AI Genuinely Helps Us
Here’s what we actually use AI tools for, and the real impact on your project:
Transcription and Rough Cut Assembly When we film interviews or talking-head content, AI transcription tools convert the audio to text almost instantly. We can then build a paper edit — selecting the best lines and sequencing the story — before opening the editing timeline. What used to take two to three hours of manual logging now takes twenty minutes. The quality of the edit isn’t determined by this step, but it gets there faster.
Subtitle and Caption Generation Accurate captions are essential for social video — most people watch without sound. AI tools generate first-draft subtitles that are roughly 90–95% accurate, which we then review and correct manually. A task that used to take an hour per video now takes ten minutes. More captions, faster, at no quality cost.
Colour Matching Across Deliverables When a single shoot produces eight or ten different cuts for different platforms, AI-assisted colour matching tools help maintain visual consistency across formats without manually grading each clip from scratch. This is a genuine workflow improvement — invisible to the viewer, but meaningful for production efficiency.
Music and Audio Recommendation AI-driven music licensing platforms help us find appropriate tracks faster than manually searching libraries. Given the brief — tone, pace, platform — the filtering narrows the search significantly. Final selection is always human. But the shortlisting is much faster.
Script Drafting and Structural Suggestions For certain content types — explainer videos, product demonstrations, structured tutorials — AI tools help generate first-draft script frameworks. These are never used verbatim. But having a structural skeleton to react to, refine, and rewrite speeds up the scripting process for clients who find a blank page daunting.
AI has meaningfully changed certain parts of the workflow. It has not replaced the creative thinking that determines whether a video actually works.
Where AI Falls Short — And Why It Matters
This is the part most agencies skip in their marketing materials.
Creative direction and strategic judgment.
AI tools cannot answer the question: what does this video actually need to accomplish, and what is the most effective way to accomplish it for this specific audience? That requires experience, context, and the kind of nuanced understanding of human motivation that no current AI tool comes close to replicating. Every project we take on starts with this question, and the answer shapes everything that follows. No automation involved.
On-camera direction and talent management.
Getting a real person to perform naturally on camera — to tell their story without freezing, to seem authentic rather than rehearsed — requires a human being in the room. The ability to create a relaxed environment, read a subject’s body language, adjust questions in real time, and pull out a genuine moment is entirely a people skill. A camera can be operated remotely. Directing someone cannot.
Client relationship and brief interpretation.
Understanding what a client actually needs — as distinct from what they’ve asked for — requires conversation, listening, and judgment built over years of working with businesses. When a client says “I want something cinematic,” they usually mean something specific that has nothing to do with the technical definition of cinematic. Translating the brief into the right execution is a craft skill. It doesn’t compress.
Quality control and brand judgment.
AI tools can generate content. They cannot reliably judge whether that content is good — whether it reflects the brand accurately, whether the pacing feels right, whether a particular line lands or falls flat. Every AI output we use goes through human review before it touches a deliverable. The review is not optional.
The Honest Position on AI-Generated Video
There are tools now that can generate entire video sequences from text prompts. We’ve tested them extensively. For certain narrow applications — abstract visualisations, motion backgrounds, conceptual sequences — they have genuine utility.
For business video? The results are currently recognisable as AI-generated by most viewers. That recognition undermines the authenticity that makes video marketing effective in the first place. We don’t use fully AI-generated footage in client work, and we’ll tell you clearly if that ever changes.
Our position: AI as a workflow tool, yes. AI as a substitute for real production, not yet — and probably not for the content types that most businesses actually need.
AI as a workflow tool: yes. AI as a substitute for real production: not yet — and not for the content that actually drives results.
What This Means for You as a Client
When you work with us, AI tools make your project faster and certain elements more cost-efficient. They don’t change what you’re fundamentally buying: experienced creative judgment, strategic thinking, and professional execution from people who care whether the content actually works for your business.
If you’re talking to other agencies about AI, the right question isn’t “do you use AI?” Everyone does, in some form. The right question is: “Where specifically does AI touch my project, and what does the human review process look like?” The answer to that tells you a lot about how seriously an agency takes quality control.
Leave a Reply
You must be logged in to post a comment.