I've written before about how I used Mistral model to generate outlines for my Project Phoenix essays. Well I was bored over the weekend, heard about Google's Bard service which is their alternative to GPT, and spent a few hours trying it. I was hooked. So completely and totally hooked. To generate my 'steal this idea' projects, all I had to do was give it a template, and ask it to ask me contextual questions to be able to generate a proposal. Then it went off with its questioning. Truth be told, the questions weren't easy and they weren't the kind you could half-assedly answer. In certain ways, writing the essay by myself would probably have been faster as there wasn't the need to have the middle man to convert my language into the language of the proposal world. But there Bard was, doing the semantic translation, and it all sounds so good! I'm five proposals in, not completely written written, but with two or three versions of each that I've saved. When I do a final round, I'm going to combine those versions to come up with a final draft that I'll then post on the website. Was it quicker than just writing it out myself? No. Was it more thorough, oh so much so! And now my template aligns with everybody else's, which is the biggest win that could be obtained.
Now I'm thinking if I can use this to convert the outlines into proper essays by having it ask me more questions...
I'm writing more on this later in a separate post, but if anybody was unconvinced about the efficacy of gen AI, let this be the clarion call: this is the disruptive stuff specially for people with editor/writer skills, the dynamics is going to favor much more towards the people who are more experienced with evaluating information and organizing events etcetera. The generalist shall rule in the hyper-specialized world!
No comments:
Post a Comment
Tell me what you think. I'll read, promise.