I recently finished PMI's new The Standard for Artificial Intelligence in Portfolio, Program, and Project Management and found it to be a thoughtful framework for governance, risk management, ethics, stakeholder engagement, and responsible AI adoption.
One theme that stood out to me throughout the document was the emphasis on data quality, human oversight, accountability, and decision-making.
As I reflected on the standard, I found myself thinking about a related question.
Most organizations don't make decisions directly from raw data.
They make decisions from interpretations of data:
- Dashboards
- Reports
- Metrics
- Summaries
- Recommendations
- Executive briefings
Historically, those interpretations were created primarily by people.
Increasingly, AI is participating in that process.
AI can now generate meeting summaries, portfolio analyses, risk assessments, prioritization recommendations, executive updates, and decision-support artifacts.
Which raises an interesting question.
As AI becomes more involved in generating the information leaders consume, should organizations be paying as much attention to the quality of interpretations as they do to the quality of data?
In other words:
If data quality has traditionally been the foundation of good decision-making, how should organizations think about validating AI-generated interpretations before those interpretations influence decisions?
I'm curious how others are approaching this.
Do you see the next challenge as primarily a data quality problem, an AI governance problem, or something else entirely?
