top of page
Search

Humans to Be Abolished in Podcast and Video Production?

  • Writer: Neil McCafferty
    Neil McCafferty
  • Jul 17
  • 3 min read

Updated: Jul 18

Not just yet.

It’s a line that occasionally comes up when clients ask, “Can’t we just use that AI thingy to make the podcast or video instead of doing all the recording, editing, filming, and mixing?”

Sorry to disappoint, but there’s a lot of hype around “AI tools” especially from those keen to sell them. As a branded podcast and video producer, I’ve been hands-on with many of these tools, and while some are genuinely useful, others still need close human supervision - or a full rescue.

Let’s take a look at where AI delivers real value, where it still fumbles, and why humans aren’t out of the job just yet.


🎛️ Where AI Really Helps in Podcast and Video Production


Audio clean-up: Say you’ve recorded a great interview, but there's a persistent background noise you didn’t catch at the time. Previously, you’d spend hours isolating frequencies and cleaning the track. Now, with the right tool, the fix takes minutes. A big time-saver, though ideally, you’d have recorded a perfect file in the first place (we’ve all been there).


Clip generation: Want to pull multiple short clips from a longer podcast or video for social media? Several AI tools can do this automatically and mostly get it right. Another couple of hours saved.


Transcription and subtitles: This is where things get a little dicey. AI-generated transcripts are roughly 90% accurate. But that 10% can cause real problems - especially when errors affect meaning.


🤖 When AI Gets It Wrong


I recently completed a social media clip project with animated, TikTok-style captions. A leading AI platform did the transcription. While it looked flashy, it completely changed the meaning of two key phrases:

  • “Old handwriting” became “old hundred”

  • “Can” was transcribed as “can’t”

Tiny mistakes, big consequences. We caught them, of course, but what if we hadn’t? Imagine explaining to a judge in a libel case that your AI “sort of friend” got it wrong. Vendors keep saying “it’ll get better.” I’m still waiting.


Privacy matters, too: Some AI platforms use your content to train their models. If you don’t read the fine print, you might be handing over copyright or uploading sensitive material that ends up outside your control. At Digi Media, we process everything locally on Apple Silicon Macs, not in the cloud. No surprise data leaks. No loss of rights.


🧠 Editing: Still a Human Job


Some AI podcast editors offer a quick first pass. They remove filler words and tighten up gaps. Not a bad start. But they can also butcher the natural rhythm of conversation. You’re left with a robotic-sounding audio or a video that cuts so fast between speakers it feels more like a table tennis match than a conversation.

So yes, we use AI. Selectively. It helps automate the boring bits and increase output: more clips for more platforms, less time spent on repetitive tasks.

But when it comes to creative storytelling, narrative sense, and production quality — the human still wins.


🧵 TL;DR: What Works, What Doesn’t

✅ AI is great for:

  • Audio clean-up

  • Social clip extraction

  • First-pass edits

  • Transcripts (with a human check)

❌ AI struggles with:

  • Meaning and nuance in transcription

  • Creative decisions

  • Privacy protection

  • Final polish

Verdict: The tools are useful, but the humans aren’t going anywhere.


Top Tip: If you want to produce a podcast or make some video content using (mostly) humans, do get in touch: neil@digimedia.tv


AI Tools in podcast and video production. A mixed bag,
AI explosion. It's a mixed bag.

 
 
 

Comments


©2025 BY DIGI MEDIA.
A TRADING STYLE OF TALK RUSSIAN LIMITED. REGISTERED IN ENGLAND AND WALES COMPANY NUMBER 07960287

bottom of page