Hyderabad: Hyderabad Commissioner of Police, VC Sajjanar on Friday, February 6, cautioned the public against “Agentic Artificial Intelligence”.
Sajjanar said that AI has evolved drastically from assisting humans earlier to now, taking decisions on its own. The Commissioner expressed concerns over the use of AI for assistance in banking and healthcare sectors.
“With these digital agents performing tasks independently without human intervention, there is widespread concern that we are at risk of losing control over them,” Sajjanar said in a post on X.
CP explains agentic AI
Urging the people to understand what agentic AI is , the commissioner said that if someone asks a regular AI, “How is the weather outside?”, it merely provides information. But Agentic AI is different; realizing the weather is bad, it might close the windows itself and adjust the AC temperature.
He explained that this means it doesn’t just think; it puts thoughts into action. Freezing accounts if suspicious transactions occur in the financial sector, altering medication dosages based on a patient’s blood pressure and sugar levels in hospitals, and controlling machine performance in industries, these agents are doing such things on their own. Their speciality lies in discussing with other AI agents and working together to complete a task.
He said that the autonomy gained by AI has become a major challenge. “These agents, which take thousands of decisions in minutes without human supervision, can lead to massive danger if they make even a small mistake anywhere. For example, if an AI agent makes a wrong decision in the stock market, losses worth crores could occur in mere seconds,” the commissioner said.
He said that there is also a threat of cybercriminals hijacking the behavior of these agents and forcing them to commit wrongdoings. He also raised concerns over ‘Unintended Behaviour,’ of AI, where the agent misunderstands the task assigned by the owner and takes a wrong path to achieve the goal.
Control of AI should remain with humans
Sajjanar opined that the control of AI should remain with humans and that all AI agents must have a digital identity which could help track their activities.
“Ultimately, there must be strict rules (Governance) regarding how much freedom these agents should be given and what data they are allowed to access. ‘Trust Frameworks’ must be designed to ensure human permission is sought before making crucial decisions,” Sajjanar said.
He concluded by saying that while speeding up work through technology is essential, security is even more critical.
