SECOND EDITION
By Diane Roberts
Texas, USA
Introduction – The Acceleration Trap
Artificial intelligence (AI) is fundamentally changing how organizations make and execute decisions. Previous waves of digital transformation improved efficiency and scale, while AI accelerates decision making. Organizations adopting AI are making decisions more frequently and at significantly higher speed (McKinsey & Company, 2023). However, faster does not mean better. In many organizations, decision speed is outpacing the structures required to ensure decisions are clear, accountable, and aligned.
The National Institute of Standards and Technology (NIST) defines governance as a “cross-cutting” function that shapes organizational culture, policies, and risk practices across the AI lifecycle (NIST, 2023). This reflects a shift from static control to an embedded organizational capability. Similarly, the Project Management Institute (PMI) positions governance as an integrated lifecycle discipline spanning data, models, and outcomes rather than stage gated oversight (PMI, 2023; PMI, 2024).
Despite these advances, a critical gap remains. Existing frameworks address risk and ethics but do not fully account for the tension between decision speed and execution clarity. This paper argues that governance must function as an execution discipline that deliberately links speed with clarity to ensure accountability, alignment, and coordinated execution. This perspective extends existing governance frameworks by explicitly connecting decision speed with execution alignment in AI-enabled environments.
The Diagnostic Lens: Decision Velocity vs Decision Clarity
Building on this gap, decision velocity reflects how quickly organizations generate, evaluate, and act on decisions in AI-enabled environments. That speed is real, and it is increasing. AI systems process large volumes of data, identify patterns, and produce recommendations in near real time. The expectation quickly becomes that decisions should move just as fast.
Decision clarity operates differently. It shows up in whether people actually understand what was decided, why it matters, and what happens next. Clarity requires defined ownership, a clear rationale, visibility into downstream impact, and alignment on what success looks like. Without those elements, speed alone does not translate into effective execution.
In practice, velocity and clarity have to scale together. When they do not, the gap becomes visible almost immediately. Decisions move faster, but accountability becomes diffuse. Teams act, but not always in alignment. Work accelerates, but outcomes do not improve at the same rate.
More…
To read entire article, click here
Editor’s note: Second Editions are previously published papers that have continued relevance in today’s project management world, or which were originally published in conference proceedings or in a language other than English. Original publication acknowledged; authors retain copyright. This paper was originally presented at the 18th Project Management Symposium at the University of Texas at Dallas in May 2026. It is republished here with the author’s permission.
How to cite this paper: Roberts, D. W. (2026). AI Without Chaos; Originally presented at the 18th Project Management Symposium at the University of Texas at Dallas in May, republished in the PM World Journal, Vol. XV, Issue IX, September. Available online at https://pmworldjournal.com/wp-content/uploads/2026/09/pmwj168-Sep2026-Roberts-AI-Without-Chaos-2nd-Ed.pdf
About the Author
Diane Roberts
Texas, USA
Diane Roberts is a Regional Vice President at Salesforce, based in the Dallas–Fort Worth area, where she leads the delivery of revenue-driving solutions for FinTech clients. Her current work focuses on enabling the responsible adoption of artificial intelligence, with an emphasis on governance, trust, and bias mitigation. Beyond her corporate leadership, Diane is a certified Texas Master Gardener dedicated to advancing food security across the DFW region. She holds PMP, Scrum Alliance, and Scaled Agile certifications and is a graduate student in the MS in Data Analytics program at SMU.
Diane can be contacted via LinkedIn: linkedin.com/in/dianewroberts
