i simplified the upper bound
i think the point of all this isn't really the or otherwise but rather to get the upper bound which the method in the exam doesn't do
chatgpt isn't reliable for checking because it isn't as robust as other older and more trustworthy technologies such as calculator, wolframalpha, spreadsheets, etc. so i don't think they would do that
ai got off to a bad start being very bad at maths. it's been getting better but not to the point that it could be relied upon for checking.