Please choose:
Move to the governance agenda. Build pragmatic guardrails and review in twelve months. Competitors will not wait.
Commission an external AI audit before making any decisions. No further deployment until we know what we have.
Establish an internal AI ethics committee with a remit to audit all existing deployments within six months.
Name the values question before the governance discussion begins: "What do we believe we owe to the people our systems make decisions about? What are we not willing to trade for efficiency? That is where we start."