Should DeFi emergency powers be explicit instead of pretending they do not exist?

The recent debate around Arbitrum freezing attacker-linked assets raises a practical question: is the problem that emergency powers exist, or that users often do not know where they exist until something breaks?

A system can market itself as decentralized, but if a council, multisig, issuer, or validator set can intervene under pressure, that is part of the actual security model. It should be documented like any other assumption.

This matters more as execution moves cross-network. If a user expresses an intent and infrastructure handles routing, liquidity, and settlement behind the scenes, the user also needs clarity on who can intervene at each layer and under what conditions. SODAX calls its model a cross-network execution system, which is the right category in my view: execution systems should be judged on transparency, not just speed.

Would you rather use DeFi infrastructure with no intervention path, or one with narrowly scoped, pre-defined emergency rules?

Source: https://cointelegraph.com/features/defi-freeze-stolen-funds-everyone-agrees-it-should