revspace7777
Newbie
- Nov 14, 2022
- 12
- 9
BACKGROUND:
I've been playing around a lot with OpenAI's new product GPT-3. For those that don't know, it's a natural language model that'll spit out human readable text based on user prompts. (If you don't know what it is, I recommend playing around with it yourself: https://beta.openai.com/playground
After getting used the controls & experimenting with prompts, I've so far been able to have it produce...
ONE ~1500 word article on the causes & dangers of microplastics in the water supply
ONE ~900 word article on how granulated activated carbon water filters work
A number of compelling product descriptions
Write product titles based on a product description
The first article took me about 30 minutes to produce because I manually asked it answer a bunch of questions one after the other, then went behind & did some editing, formatting, etc. Second article took 60 seconds. I gave it the blog post title & section outline and it wrote 900 words in 60 seconds.
APPLICATION:
The application is obvious. Use GPT-3 (and next year, GPT-4) to produce as much content as you can before OpenAI & Google figure out how to build in machine-readable artifacts to the output so Google can NERF the thing.
If you get good at the prompt engineering you can produce a lot of content between now & then.
Would love to hear everyone's thoughts, ideas, etc.
I've been playing around a lot with OpenAI's new product GPT-3. For those that don't know, it's a natural language model that'll spit out human readable text based on user prompts. (If you don't know what it is, I recommend playing around with it yourself: https://beta.openai.com/playground
After getting used the controls & experimenting with prompts, I've so far been able to have it produce...
ONE ~1500 word article on the causes & dangers of microplastics in the water supply
ONE ~900 word article on how granulated activated carbon water filters work
A number of compelling product descriptions
Write product titles based on a product description
The first article took me about 30 minutes to produce because I manually asked it answer a bunch of questions one after the other, then went behind & did some editing, formatting, etc. Second article took 60 seconds. I gave it the blog post title & section outline and it wrote 900 words in 60 seconds.
APPLICATION:
The application is obvious. Use GPT-3 (and next year, GPT-4) to produce as much content as you can before OpenAI & Google figure out how to build in machine-readable artifacts to the output so Google can NERF the thing.
If you get good at the prompt engineering you can produce a lot of content between now & then.
Would love to hear everyone's thoughts, ideas, etc.
