Talk shows applause are manipulating you, says GPT - try it yourself!
I recently watched an interesting political talk show about the gap between rich and poor and the Gini coefficient. The show had many competent guests from both sides - the left, the rich and the less fortunate. All in all, it seemed quite balanced.
After a while, however, I noticed my tendency to "shift" to the left. As I thought about it and watched the arguments go back and forth, I realised that I hadn't been consciously focusing on something important: The applause of the audience!
I started paying close attention to the applause alongside the talk and noticed one thing for sure - the audience was leaning left on wealth distribution and less on liberating the economy.
Without any judgement on whether this is good or bad, I just found it fascinating and became curious if we could automate this analysis using AI. I'm pretty sure it can find even more interesting patterns than humans can. And that is what I want to share with you here.
****Disclaimer:****
Although this post may sound political, it's not. You can see it as an application of AI to gain more visibility into the intricacies of our society and also the media. It is up to you to decide what to do with this information.
If you recall my previous post, the three things I love about AI---the last one being transparency. This is exactly what AI can help with here.
But why talk so much, try it for yourself. Here is how.
- Choose a YouTube talk show that has an applauding audience.
- Transcribe the show.
- Provide the transcript to ChatGPT and ask it to identify patterns using a specific prompt.
- Observe the results.
Step 1: Select any YouTube talk show that features an applauding audience---it doesn't matter if it's satire, comedy, etc. You can also simply ask Perplexity.
Transcribe it using any tool you like, for example, Tactiq.
Copy the text somewhere.
Step 2: Open GPT and paste this prompt, adding the transcript at the end of the prompt.