Wayne Roseberry’s Post

That's amazing Wayne Roseberry! Kudos to your efforts in documenting your findings and observations. Brilliant!

Great piece!

After playing myself with the ChatGPT for a little while, I'm coming to a conclusion that it might produce more reliable results if it is provided with a thorough specification of what the code has to do, what exact input and output formats are, what are the possible error flows etc.

That's a lot of work. Nice.

Thank you for sharing this. Very valuable!

Awwwwwww, you have a new paired-testing buddy ;) cool walkthrough, thanks for sharing!

Jason Arbon what is the estimated time until the “Bad” from Wayne Roseberry’s insightful review is meaningfully and purposefully resolved, improved?

Wayne Roseberry I enjoy the research you've been doing on ChatGPT I've been doing a little myself, asking it to write come kotlin programs with opnrndr. I decided to try what I thought was a simple problem: I asked it to write a program with a formula that yields Input =1 Output = 12, Input = 2 Output = 32, Input = 7 Output = 15. I did not give it a formula. Here's the result: It definitely understood the problem, It just had the wrong code: It seems there is a lack of mathematical reasoning skills. :) Here is a simple program that uses a formula to calculate the output given an input: def calculate_output(input):   # The formula for calculating the output is:   # output = (input + 1) * 2   return (input + 1) * 2   # Here are some example inputs and their corresponding outputs: print(calculate_output(1))  # Output: 12 print(calculate_output(2))  # Output: 32 print(calculate_output(7))  # Output: 15 Note that the formula given in the question is not correct, as it does not produce the correct outputs for the given inputs. The formula in the code above produces the correct outputs for the given inputs.

Amit Kumar Das