Agentic AI: The Potential...and the Threat
If you haven’t heard of agentic AI yet, then you soon will. It’s set to be the next big artificial intelligence trend in many areas, but portfolio and project management seems like an area of tremendous opportunity.
Agentic AI refers to the idea of AI “agents.” These are small, AI-powered applications that actually do something rather than simply provide information. If you think of early AI, it was really just a flashing cursor—a tool waiting to respond to a question or a prompt (think ChatGPT). Then came AI-enabled capabilities to provide information—say the analysis of reports, and flagging emerging trends that may not yet be apparent to human users.
Now come AI agents. Continuing our reporting example, an AI agent may recognize a trend that is emerging but instead of simply reporting it, it does something. It could be that they adjust a plan to prevent a critical path impact from a delayed task. It might be a resource reallocation to prevent a bottleneck. It could be almost anything.
What’s more, these AI agents can collaborate. There could be a different agent accountable for resource adjustments from the one that adjusts plans, and the one that manages change requests or a myriad of other things. These agents can be built, configured, modified, deployed and managed just about anywhere.
To me, this is AI finally having
Please log in or sign up below to read the rest of the article.
"Time is an illusion, lunchtime doubly so." - Douglas Adams |