Given the ODE $f(x)=f'(x)$ one obtains that its only solution is $ce^x$ for some constant $c$.
So if one adds the constraint $f(0) = 0$, it's easy to check that there's no way to bend the particular solution to generate this value besides setting $c=0$.
Technically, we should have enough information to reason this without any use of our knowledge about ODEs, simply by proving that the function can't be not constant.
So by basic method, I mean a method as simple as possible, but at least one that's not using knowledge about the possible functions solving the ODE.
A few examples of possible answers, which in my opinion still use too much machinery for the simplicity of the problem:
1) If the solution were analytical, its Taylor polynomial simply would be 0.
2) We can make an iterative approximation by splitting the interval $[0,b]$ into $n$ pieces, and for every piece add the increase caused by $f$'s gradient to the function (which trivially means $f$ will stay constant throughout the whole iteration). We can then show that if we let $n\to\infty$, the error of the approximation goes to 0.
I really feel like there should be a simply and easy solution to this: After all, we know no less than
$0=f(0)= f'(0)= f''(0)= ...$
Which intuitively, combined with $f$'s continuity already seems to force $0$ to be the only solution.