Governance in the age of agentic AI
The future of work is changing rapidly. It was only a few short years ago when generative AI moved from specialised use to widespread, mainstream adoption. There was great excitement over new ways of working. Fast forward to 2026, and AI technologies have rapidly evolved into autonomous decision makers that have developed capabilities to learn, reason and execute decisions. AI agents exhibit human-like agency acting on our behalf. The opportunities of this powerful technology are truly limitless, changing how we govern and manage organisations, delegate responsibilities and execute strategies.
However, there are risks. Rushed deployment, driven by fear of missing out or by moving too quickly without fully considering the technology’s limitations, the ethical dilemmas its use in high-risk use cases raises, and existing or emerging legal and regulatory obligations, is almost guaranteed to result in failed pilot programs and poorly executed business cases.
Agentic AI projects are at risk of being cancelled due to escalating costs and unclear business value propositions. This is driven by a lack of awareness into the evolving nature of risk management, governance and oversight of non-human identities in the workplace. Agent AI delegation is now critical in orchestrating a new digital workforce of agents that have the potential to drive positive changes in the workplace spilling into economic and societal value.
The technology can execute repetitive, labour-intensive activities and execute complex workflows with limited human oversight, augmenting the way we work. The potential productivity gains of AI are well-documented. However, in the case of agentic AI systems that are becoming increasingly autonomous, critical questions arise over the quality, integrity and veracity wherever decisions are made.
The orchestration of AI agents is at the heart of what distinguishes a truly effective governance and oversight framework, reducing organisation-wide risks including agentic drift, the potential for data manipulation, malicious cyber incursions and privacy breaches.
The rise of agentic commerce through agent payments and binding contractual agreements necessitates an understanding of the legal and regulatory landscape as a fundamental measure of successful deployment.
As governance professionals, we must explain why decisions were taken, how they were executed, and whether those decisions can withstand challenges after the fact.
Done properly, agentic AI systems can drive demonstrable business value, lift workplace productivity, improve ways of working and decision-making. The future of work has arrived, and governance professionals have a role to play in driving better decisions everywhere they are made.
Footnote:
We have worked comprehensively with our sponsors, Diligent, Mallesons, Seek, EthicAI, University of Melbourne’s Centre for AI and Digital Ethics and the Governance Institute’s AI Governance Expert Advisory Panel to offer an authoritative white paper on the future of governance in the age of agentic AI.