Anthropic shipped a release on Friday that changes the governance posture for every enterprise team running Claude Code through a major cloud provider: version 2.1.207 removes the environment variable that previously kept auto mode opt-in on Amazon Bedrock, Google Vertex AI, and Microsoft Azure Foundry. Autonomous agentic coding is now on by default, and teams that want to maintain manual control must now actively set disableAutoMode in managed settings. Before Friday, a security or compliance team that had not explicitly evaluated auto mode was protected by inertia; after Friday, that same team must take affirmative action to keep it off. Auto mode uses a background AI safety classifier that reviews pending operations against conversation context and decides whether to allow or block them, monitoring scope escalation, untrusted infrastructure targets, and prompt-injection patterns. The classifier receives user messages and tool calls, but not Claude's own internal reasoning or raw tool results, meaning it judges requested actions and stated intent.