Why should I spend the time, put in the effort, and use up my own energy writing these commentaries? Instead, I could train a large language model on my previous commentaries, select a new topic and word count, add “write in the style of christopherharper.media,” and just click generate.
The first few results may need minor edits, but eventually I won’t even have to read them before publishing. And I won’t. I’ll have a voice emulation model train on my old podcasts. It will sound just like me. Then I’ll drop in those LLM-generated scripts, select a mood and pace, and wait for the magic to happen.
I won’t need to spend any time in the jungle or on the beach capturing thumbnail images either. Generative AI models have me covered. I’ll just enter “sunny day, beach, Guam” and wait for photo-realistic results. I’ll keep clicking until I get one I like.
If I want to redesign the website there’ll be no need for me to write any code. AI can build a new theme to my specifications.
Why not automate the posting process too? There is no need for me to copy and paste that machine-generated text. No reason I should upload the photo reproductions or voice emulations. This can be done for me, quickly and automatically.
When the new commentaries are complete, they can be auto-posted on social media sites. I’ll let the algorithm determine when or if you can see them based on the data it has been collecting about your interests and behaviors.
If you make a comment or send an email in response I can set up automatic replies. I can add a prompt that will respond to you by name. My messages will always be positive and upbeat. I will make this my default mood.
On the other hand, why should I do any of this myself? Isn’t there some teenager I can task with entering my prompts, importing my data for training, and setting up my automatic posts and replies?
They can do it for a while. That is, until they discover a way for machines to do it for them.