Jingle Bots: Real Tales of AI Wins, Fails, and Festive Lessons in Portfolio Management
About this webinar
What happens when people put AI to work in and around the PMO? In this candid December 2025 round-up, Stuart Easton hosts PMO Squad founder Joe Pusz, NHS Wales PMO leader James Evans, portfolio-governance practitioner Anara Asylbekova and David Dunning, who works on business governance and AI solutions.
The failures are revealing. Joe recounts hiring managers spotting candidates reading AI-generated answers during interviews, and applicants facing AI interviewers that cannot yet sustain a natural conversation. Anara recalls an attempt to rescue an ageing, badly designed system by adding AI. Stuart describes a pharmaceutical company that had set aside substantial funding for AI projects while its IT governance team imposed a moratorium on new AI work.
There are practical successes too:
- James Evans: Using Copilot to summarise messages, organise meetings and track actions in a stretched NHS Wales PMO, while retaining human quality checks.
- Joe Pusz: Applying AI to project lessons learned that organisations routinely collect but rarely revisit, looking for recurring patterns that could inform how they work.
- Anara Asylbekova: Working with a PMO team experimenting with different AI tools and agents, scoring their usefulness across tasks and deciding which are worth keeping.
- Stuart Easton: Contrasting a failed AI requirements-gathering hackathon with specialised agents that dramatically shortened newsletter first-draft preparation. The hackathon taught him to redesign a process for AI rather than simply automate the human version.
The panellists approach adoption from different angles. James looks for problems consuming scarce PMO time; Anara explores and tests tools; and David argues that technical capability alone will not persuade an organisation without a credible business case. His AI bid-writing example raises a practical information-quality challenge: where are the previous answers, and who can say they are reliable? Anara describes employees putting company information into AI tools without understanding the risks, while Stuart explains how a single vetted SharePoint folder helps his team train agents using controlled source material.
Research in context
Why does AI help with some professional tasks but make others worse? In a study published in Organization Science in 2026, researchers randomly assigned 758 Boston Consulting Group consultants to work with or without GPT-4 assistance. Across 18 tasks within the technology's tested capabilities, those using AI completed 12.2% more tasks and worked 25.1% faster on average. But on one complex managerial task outside that range, AI users were less likely to reach the correct answer.
The researchers call this the "jagged technological frontier": tasks that look similarly difficult to people can produce very different results with AI assistance. This gives useful context to Stuart Easton's contrast between an unsuccessful AI requirements-gathering hackathon and newsletter agents that accelerated drafting. Success with one application does not mean the same approach will work on another task.
How this connects to TransparentChoice
David Dunning argues that a technically promising AI solution still needs a credible business case before an organisation will support it. But what happens when several AI proposals have credible business cases and compete with existing commitments for the same funding and people? Which deserve support, and what might need to give way?
TransparentChoice helps leadership compare proposed AI investments alongside existing work, using agreed priorities, funding and capacity constraints to identify feasible portfolios. Leaders can examine which initiatives receive support, what is displaced and how the choice changes when assumptions shift. See how TransparentChoice supports portfolio decisions.
Learn more
Join the next live webinar
If AI automates routine PMO work, what should the PMO focus on next?
Laura Barnard, Rebecca Reynolds, Dr Al Zeitoun and Mickey Nakamura explore how AI disruption is changing expectations of PMOs, from reporting and administration toward decision support, advisory work and governance. Watch The New Role of the PMO in a Disrupted World.
When AI improves productivity, does that justify cutting roles?
Productivity gains do not automatically translate into positions that can be removed. Explore the evidence needed to move from time saved to workforce decisions, including the remaining workload, service requirements and timing. Read the AI productivity and restructuring analysis.
What else must be funded to make an AI investment viable?
AI initiatives can depend on data platforms, systems and scarce technical capacity that other commitments also need. Explore an AI, ERP and cybersecurity investment scenario showing why enabling dependencies and opportunity costs matter when deciding what to fund. Read the AI budget opportunity-cost analysis.
Speakers
Joe Pusz, known as PMO Joe, is the founder of PMO Squad Staffing and a leading voice in project management. He co-founded International PMO Day, The PMO Leader, and the Veterans Project Management Mentoring Alliance (VPMMA), and hosts the Project Management Office Hours podcast. Recognized as PMO Influencer of the Americas (2022) and consistently ranked among the top global PMO influencers, Joe brings extensive experience in organizational project delivery, leadership, and mentoring.
David Dunning is a Business Integrated Governance thought leader and founder of the Business Integrated Governance CIC and Deep Team. He also serves as Chairman of CPS, helping organizations improve strategy delivery, governance, and business agility.
Stuart Easton is Founder & CEO of TransparentChoice. He works with leaders and portfolio teams on decisions about which investments and initiatives to fund, continue, change, defer or stop when priorities compete for limited funding and capacity. His work combines structured decision-making with practical portfolio choices.