You're presumably thinking of the IARC monograph. You can read how the classifications work in their own preamble:
https://monographs.iarc.who.int/wp-content/uploads/2019/07/P...Specifically,
> A cancer hazard is an agent that is capable of causing cancer, whereas a cancer risk is an estimate of the probability that cancer will occur given some level of exposure to a cancer hazard. The Monographs assess the strength of evidence that an agent is a cancer hazard. The distinction between hazard and risk is fundamental.
> The identification of a cancer hazard should trigger some action to protect
public health, either directly as a result of the hazard identification or through the conduct of a risk assessment.
The key thing to recognize here is they are making preliminary assessments of existing lines of research, looking for studies into whether an agent has been associated with cancer in humans, animals, or mechanistically, and characterizing the strength of such evidence, as to whether or not something can cause cancer.
The purpose of doing this is so that other standards bodies can assess the actual dose/response relationship and your probability of getting cancer from levels of exposure. Glyphosate in in group 2A, along with things like being a barber, night shift work, hot beverages, red meat. Lots of things the EPA does not ban.
Why? Because as the organization itself says, hazard does not mean risk. The fact that studies show a robust association between one thing and some increase in cancer incidence, at any dose, does not mean the level of exposure any regular person experiences raises their probability of getting cancer by any meaningful amount we should care about or regulate.
The problem here is you're almost certainly not actually reading the publications of either IARC or the EPA, which are not in conflict with each other. You're reading science journalism, which almost universally presents this in the most misleading, fear-mongering fashion possible to attract eyeballs.