There's very little epidemiological evidence to suggest glyphosate is a human carcinogen. Millions of people are exposed to it, some in significant amounts on work sites. If there's any correlation between heavy occupational exposure to glyphosate and human cancer, it's very weak.

Similarly: there's not a lot of first-principles logic to carcinogenicity of glyphosate. It's active in a metabolic pathway that animals don't have.

Finally, there's a lot of fuzz around glyphosate that would lead us to be skeptical of its danger. In particular: glyphosate is most closely associated with genetic modification and Monsanto (part of the appeal of glyphosate is that you can genetically engineer plants that resist it). But glyphosate itself is not a product of genetic engineering!

It's not impossible that glyphosate is a human carcinogen! But it's not impossible that a lot of things are carcinogens. To be intellectually consistent about the precautionary response to glyphosate, you're going to have to get rid of a whole lot of other features of modern life as well.

Acrylamide, which is produced in basically all human dry-heat cooking, is a known human carcinogen. The science behind the risks of acrylamide is, relative to that surrounding glyphosate, very solid. But we're never going to "ban" acrylamide, or, realistically, do anything to mitigate our exposure to it.

In that light, it seems reasonable to push back on public policy interventions for a useful chemical that has, based on current evidence, much less risk than "almost all cooked food".