Relying on human-in-the-loop for every agent anomaly is unscalable. An effective governance model uses automation and agent 'interrogation' to resolve low and medium-risk issues. Human oversight is reserved exclusively for critical incidents, preventing security teams from being overwhelmed.

Most developers admit to giving pull requests only a cursory glance rather than pulling down the code, testing it, and reviewing every line. AI agents are perfectly suited for this meticulous, time-consuming task, promising a new level of rigor in the code review process.

While crypto firms seek access to next-gen AI for security testing, the real insight is that current-generation models are already proving superior to human auditors. For example, crypto custodian Fireblocks found that an existing Anthropic model detected critical vulnerabilities that multiple professional security audit firms had missed.

An Anthropic study on user behavior found that as AI generates more polished outputs like working code, users become less evaluative and more trusting. This "verification gap" is a critical flaw in human-AI collaboration, as polished results should trigger more scrutiny, not less.

Claude Code's "AutoMode" uses one AI to check if another AI's proposed actions are safe, replacing constant user permission prompts. This is more secure than relying on users prone to "yes-fatigue" and simultaneously creates a better, more seamless user experience.

Contrary to the belief that AI introduces risk, Intercom's AI code review agent led to a lower revert rate for AI-approved code. This demonstrates that AI can enhance safety and quality, not just speed, by systematically enforcing standards on every pull request.

As AI agents generate vast amounts of output, human review becomes an impossible bottleneck. The solution emerging is multi-agent systems where a separate 'grading agent' automatically scores and requests revisions on an agent's work against a predefined rubric, as seen in Anthropic's 'Outcomes' feature, enabling scalable quality assurance.

Chris Fregley argues that manually reviewing AI-generated code is slow and ineffective. He has replaced traditional code reviews and unit tests with a focus on robust, continuous evaluation frameworks ("evals") and correctness checks that run in the background, allowing for faster and safer code deployment.

Anthropic's study reveals a paradox: expert users grant AI agents more freedom via auto-approval while also interrupting them more frequently. This suggests mastery involves active, targeted supervision to guide the agent, not a passive "set it and forget it" approach.