Agentic artificial intelligence (‘AI‘) moves workplace AI from recommendation to action. Agentic AI systems are designed to do more than draft, score, and recommend. They can break a goal into steps, call on other software, and complete tasks with limited human direction, such as scheduling interviews, answering employee questions, routing requests, summarising performance information, or drafting and distributing documents. An agent’s ability to carry a process across multiple steps may reduce handoffs and repetitive work, with the potential to gain significant efficiencies.
The adoption curve for agentic AI is aggressive. Gartner’s 2026 CIO and Technology Executive Survey found that AI agent deployment is expected to increase from 17% in 2026 to more than 60% within the next two years, with at least 15% of day-to-day work decisions being made autonomously through agentic AI.
However, plans for agentic AI may be moving faster than the frameworks for overseeing it. Seventy-four percent of organisations expect to use AI moderately in the next two years, but only 13% think they have the right governance in place. It is then not surprising that over 40% of agentic AI projects are predicted to be cancelled by the end of 2027, citing cost, unclear business value, and inadequate risk controls.
This Report focuses on a path for employers to capture the value of agentic AI by treating autonomy as a deliberate design decision, reimagining functions based on new capabilities, adapting governance to reflect emerging laws and developing legal doctrine, and preserving the record.
This Report offers an analytical framework for overcoming uncertainty and making strategic decisions with a legal foundation as follows:
- What agentic AI is, how it differs from earlier tools, and where employers are using it;
- How federal, state, local, and global rules treat workplace AI, and where they assume there is a human decision-maker;
- The mismatch between the current regulatory environment and advances in autonomous AI systems;
- How agentic autonomy changes risk in discrimination, agency and authority, data access and privacy, wage and hour, labour relations, restructuring, leave and accommodation, trade secrets, and with respect to vendors; and
- How litigation and discovery are developing, and how plaintiffs may adapt existing claims.
The Report concludes with a practical framework for AI-enabled workforce transformation based on defensible governance: Decide, Design, Defend.
