Heads of AI Governance arrive in the role from many directions - privacy, risk, compliance, legal, security, engineering, transformation. The role is consistent; the background isn't. Four patterns recur regardless of the path in:

• The risk type is new to the organization. Previous governance disciplines built up over decades - credit risk, operational risk, information security risk - have established methodologies, established controls libraries, established board literacy. AI risk has none of that maturity yet; the Head of AI Governance is often the person building the methodology in flight.

  
  

• Adoption is happening faster than governance can scale. Unlike privacy (where the regulation arrived before mass adoption), AI is being adopted in production at a rate that outpaces the team's ability to review. Governance has to find a way to enable adoption rather than gate it.

  
  

• Ownership is structurally cross-functional. Few governance functions touch as many other functions as AI governance - legal, risk, security, IT, engineering, product, business units all have a stake. The Head of AI Governance role is often the only place where the cross-functional pattern is held coherently.

  
  

• Regulation is moving in multiple directions simultaneously. EU AI Act is consolidating; the US is fragmenting; standards bodies are publishing; sectoral regulators are overlaying. The framework view is multi-dimensional from day one.