Agentic AI systems are not autonomous “thinking” entities. They operate within clear limitations and require critical evaluation rather than blind trust.

Dr. George Giannakopoulos, Scientific Director of ahedd Digital Innovation Hub S.A., Senior Researcher at NCSR “Demokritos,” and co-founder of SciFY – Science For You, recently spoke on national television about the realistic implementation of Artificial Intelligence. His key message was clear: AI projects rarely fail because of technology. They fail because of strategy.

Without a defined purpose, internal ownership, and operational alignment, even the most advanced tools cannot deliver sustainable impact. He also challenged common misconceptions — that AI systems are neutral, that they fully “understand” what they produce, or that they reflect the entirety of human knowledge.

Responsible adoption requires governance, clarity, and informed decision-making.

Watch the full discussion here: