Welcome to the forum. By chance, this question is very similar to the one immediately prior to yours; see
http://www.medhelp.org/posts/HIV-Prevention/Testing-Waiting-Period/show/1593012
1) With an antibody-only test, such as Oraquick, a 9 week result is virtually 100% reliable -- maybe truly 100%. In the preceding thread, I provided the link to yet another thread that explains why lots of agencies continue to recommend 3 months, even though it probably is unnecessary.
2) Test performance per se does not vary with risk. But overall interpretation does. If someone describes an exposure that carries only 1 chance in a thousand of catching HIV, a negative test that is "only" 99% acurate reduces the person's chance of having HIV from 1 in 1,000 to 1 in 100,000. However, if the exposure is one that carries a 10% risk of infection, then the same test result reduces the likelihood the person caught HIV to 1 in 1,000. And improvement, but not nearly as reassuring as the 1 in 100,000 figure. (In your case, you say nothing about your exposure, which would have been helpful in assessing your risk.)
3) Whether I have seen someone with such late seroconversion is a somewhat silly question. It happens I have not, but it has happened. However, it probably has happened only with the older HIV antibody tests, not in use for over 10-15 years.
I hope this helps. Best wishes-- HHH, MD