There's a little part of me that wants to say, "But Doc, you wrote this article perfectly well"... but honestly, I've had the same experience as the author. For academic and linguistic tasks, it's...
There's a little part of me that wants to say, "But Doc, you wrote this article perfectly well"... but honestly, I've had the same experience as the author. For academic and linguistic tasks, it's just remarkable. In the past month I've used GPT-4 and the API to:
Flesh out a few trouble spots in Real Analysis proofs
Translate passages from the Pali Canon with word-by-word explanation
Explain ffmpeg so I can mix audio down to 2 stereo channels
Troubleshoot a new Arch installation
Learn about leaders in civil rights movements outside America
Brainstorm worldbuilding ideas for a new RP campaign
Create sample sentences for me to practice fresh Japanese and Chinese vocab words
Extract data from my journal entries and write Python code to plot the stats it extracted
... and like probably fifty other things. If any of you have only tried the free version, I cannot recommend GPT-4 highly enough. It's the best $20/mo I think I've ever spent. Not to mention Copilot, which makes everyday coding like 2x as fast. Honestly, I don't know if I'd have had the confidence to attack half of the intellectual tasks in the list above without knowing I had its support. I'm kind of ashamed to admit it.
This is how I've felt about chatGPT in the context of coding as well. It's like an extremely fast jr developer with a seemingly infinite breadth (but not always depth) of knowledge, but needs to...
In doing so, I’ve come to realize that dealing with ChatGPT is like working with an incredibly brilliant, hard-working — and occasionally hungover — intern. That’s become my mental model for considering the usefulness of ChatGPT.
This is how I've felt about chatGPT in the context of coding as well. It's like an extremely fast jr developer with a seemingly infinite breadth (but not always depth) of knowledge, but needs to be explicitly told what to do and have their work proofread afterwards.
Someone on the net (you can have me arrested and/or sued, I do not remember where) shared how she used ChatGPT to protect her mental health when dealing with clients. Working at a car dealership,...
Someone on the net (you can have me arrested and/or sued, I do not remember where) shared how she used ChatGPT to protect her mental health when dealing with clients.
Working at a car dealership, and tasked with handling repairs on just sold, new cars. Obviously, those clients are not happy.
She recruited the machine to answer the requests in a friendly tone, shielding her from the psychological stress.
Of course she still has to review the mails, but the main stressor is reduced tenfold. (yea, I made that number up)
Another passersby chimed in on how she uses it to set boundaries with her clients in web development, again shielding her from the stress.
There's a little part of me that wants to say, "But Doc, you wrote this article perfectly well"... but honestly, I've had the same experience as the author. For academic and linguistic tasks, it's just remarkable. In the past month I've used GPT-4 and the API to:
ffmpegso I can mix audio down to 2 stereo channels... and like probably fifty other things. If any of you have only tried the free version, I cannot recommend GPT-4 highly enough. It's the best $20/mo I think I've ever spent. Not to mention Copilot, which makes everyday coding like 2x as fast. Honestly, I don't know if I'd have had the confidence to attack half of the intellectual tasks in the list above without knowing I had its support. I'm kind of ashamed to admit it.
This is how I've felt about chatGPT in the context of coding as well. It's like an extremely fast jr developer with a seemingly infinite breadth (but not always depth) of knowledge, but needs to be explicitly told what to do and have their work proofread afterwards.
Someone on the net (you can have me arrested and/or sued, I do not remember where) shared how she used ChatGPT to protect her mental health when dealing with clients.
Working at a car dealership, and tasked with handling repairs on just sold, new cars. Obviously, those clients are not happy.
She recruited the machine to answer the requests in a friendly tone, shielding her from the psychological stress.
Of course she still has to review the mails, but the main stressor is reduced tenfold. (yea, I made that number up)
Another passersby chimed in on how she uses it to set boundaries with her clients in web development, again shielding her from the stress.
Personally, I think this is brilliant!