Would this work?
If the number of primes of a given form is finite, then the number $$M = (4k_1 + 1)(4k_2 + 1) \ldots (4k_n + 1) + 4$$ should be composite. But the product of numbers of the form $4k + 1$ also gives $4k + 1$, so $M = 4K + 1 + 4 = 4(K + 1) + 1$, and $M$ is prime because it leaves a remainder of $4$ after dividing by any prime of the given form; also $M$ is not divisible by $4$. So that should be contradiction.
All right, the argument doesn't work as it stands. But then, can we fix it? Do we have a way to construct numbers that definitely aren't divisible by any primes congruent to $3$ mod $4$?
The trick:
Also, as noted in the comment, you're skipping steps in the logic. What exactly are those $4k_i+1$? Why should the number you construct be composite? An experienced mathematician will likely know how these gaps should be filled, but you can't always assume that your audience will, or that the audience will give you the benefit of the doubt.