I'm a bit more optimistic: the CI/CD pipeline can include adversarial model(s) which probes for the worst behavior. We already have codebases of sufficient comp

I'm a bit more optimistic: the CI/CD pipeline can include adversarial model(s) which probes for the worst behavior. We already have codebases of sufficient complexity that attacks can be inserted into the software supply chain. Obviously, this becomes an arms race. The usual strategy is to add random checks, friction to contribution and severe-enough punishments for offenders (lifetime RL bans, etc.) Expert opinions welcome - I'm a tourist.