What these companies didn’t take the time to understand is, A.I. is a tool to make employees more efficient, not to replace them. Sadly the vast majority of these companies will also fail to learn this lesson now and will get rid of A.I. systems altogether rather than use them properly.
When I write a document for my employer I use A.I. as a research and planning assistant, not as the writer. I still put in the work writing the document, I just use A.I. to simplify the tedious data gathering and organizing.
I just use A.I. to simplify the tedious data gathering and organizing.
If you’re conscientious, you check AI’s output the same way a conscientious licensed professional checks the work of an assistant before signing their name to it.
If you’re more typical… you’re at even greater risk trusting AI than you are when trusting an assistant who is trying to convince your bosses that they can do your job better than you.
yes, 100%, do not use an LLM for anything you’re not prepared to vet and verify all of. The longer an LLM’s response the higher the odds it loses context and starts repeating or stating total gibberish or makes up data to keep going. If that’s what you want (like a list of fake addresses and phone numbers to prototype an app), great, but that’s about all it’s going to really do.
My daughter has used AI a lot to write grant proposals, which she cleans up and rewords before submitting. In her prompts she tells it to ask her questions and incorporate her answers into the result, which she says works very well, produces high quality writing, and saves her a ton of time. She’s actually a very competent writer herself, so when she compliments the quality I know it means something.
That’s a good way to use the tool. I generally use the OpenAI option to set up a custom gpt and tell it to become an expert on the subject I’m writing about, then set the parameters. Then once I’ve tested it on a piece of the subject matter I already understand and confirm it’s working properly, I begin asking it questions. When I’m out of questions or just need a break, I go back and check the citations for each answer just to make sure I’m not getting bad data.
Once I’ve run out of questions and all the data is verified, I have it create an outline with a brief summary of each section. Then I take that outline and use that to guide me as I write. Also it seems like the A.I. always puts at least one section in the wrong place so that’s just another reason I like to write it myself and just use an A.I. summary outline.
What these companies didn’t take the time to understand is, A.I. is a tool to make employees more efficient, not to replace them. Sadly the vast majority of these companies will also fail to learn this lesson now and will get rid of A.I. systems altogether rather than use them properly.
When I write a document for my employer I use A.I. as a research and planning assistant, not as the writer. I still put in the work writing the document, I just use A.I. to simplify the tedious data gathering and organizing.
If you’re conscientious, you check AI’s output the same way a conscientious licensed professional checks the work of an assistant before signing their name to it.
If you’re more typical… you’re at even greater risk trusting AI than you are when trusting an assistant who is trying to convince your bosses that they can do your job better than you.
yes, 100%, do not use an LLM for anything you’re not prepared to vet and verify all of. The longer an LLM’s response the higher the odds it loses context and starts repeating or stating total gibberish or makes up data to keep going. If that’s what you want (like a list of fake addresses and phone numbers to prototype an app), great, but that’s about all it’s going to really do.
Oh I check the citations. I’m fully aware of A.I. hallucinations.
My daughter has used AI a lot to write grant proposals, which she cleans up and rewords before submitting. In her prompts she tells it to ask her questions and incorporate her answers into the result, which she says works very well, produces high quality writing, and saves her a ton of time. She’s actually a very competent writer herself, so when she compliments the quality I know it means something.
That’s a good way to use the tool. I generally use the OpenAI option to set up a custom gpt and tell it to become an expert on the subject I’m writing about, then set the parameters. Then once I’ve tested it on a piece of the subject matter I already understand and confirm it’s working properly, I begin asking it questions. When I’m out of questions or just need a break, I go back and check the citations for each answer just to make sure I’m not getting bad data.
Once I’ve run out of questions and all the data is verified, I have it create an outline with a brief summary of each section. Then I take that outline and use that to guide me as I write. Also it seems like the A.I. always puts at least one section in the wrong place so that’s just another reason I like to write it myself and just use an A.I. summary outline.