Cannot figure this out to save my life : s lol

Okay, brief background and description:
I own a forum that reviews products. We have a certain retailer that is having serious issues with defective products and are claiming a specific defective rate that we KNOW is way off, but I want to prove this mathematically. You can't argue with the math, right?! I have the data needed, but the math to calculate it escapes me completely. If anyone here can help our forum with this, you will be our hero!

Variables:

The company is claiming a 3% to 5% defective rate
Units: Assume 1,000,000 units sold for this problem
Needed: I would like to know the chances of receiving 1, 2, 3 and even 4 defective units in a row based on the stated defective rate and the amount of units sold.

Multiply! If the probability of any one item being defective is "between 3% and 5%", then the probability of one item being defective is between 0.03 and 0.05. The probability of two in succession being defective is between \(\displaystyle (0.03)^2= 0.0009\) and \(\displaystyle (0.05)^2= 0.0025\). The probability of three in succession being defective is between \(\displaystyle (0.03)^3= 0.000027\) and \(\displaystyle (0.05)^3= 0.000125\), etc.

I knew how to calculate it for the first defective item, but not beyond that. Thank you SO much for helping with this. For some reason, I got all the smarts for science, but not math. Go figure.
You just gave us the ammunition we needed to confront the issue, so it is very much appreciated!