Hyderabad CP Sajjanar warns of losing control over AI agents handling finances, power grids

Hyderabad: The journey of Artificial Intelligence (AI) has taken a new turn. Until recently, it was limited to ‘chatbots’ that merely answered our questions or wrote text for us. Now it has evolved to the level of Agentic AI—systems capable of making decisions on their own and executing tasks.

The advisory was issued by Hyderabad Commissioner of Police VC Sajjanar on Friday.

What are digital agents?

Autonomous robot agents have entered highly critical sectors such as banks, hospitals and power grids, the police said. However, with these digital agents performing tasks independently without human intervention, there is widespread concern that we are at risk of losing control over them.

Ideally, the common man needs to understand what Agentic AI is.

For instance, if you ask a regular AI, “How is the weather outside?”, it merely provides information. But Agentic AI is different; realising the weather is bad, it might close the windows itself and adjust the AC temperature. 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 BP and sugar levels in hospitals, and controlling machine performance in industries—AI agents are doing such things on their own. Their specialty lies in discussing with other AI agents and working together to complete a task.

It only takes one small mistake

This autonomy has now become a major challenge.

These agents, which make 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. Similarly, there is a threat of cybercriminals hijacking the behaviour of these agents and forcing them to commit wrongdoings. There is also the risk of unintended behaviour, where the agent misunderstands the task assigned by the owner and takes the wrong path to achieve the goal.

Ways to rein in AI agents

That is why experts clarify that the reins of these ultra-modern agents must remain in human hands. Every AI agent must have a precise ‘Digital Identity.’

Just as a human has an ID card in an organisation, these software agents must also have identification, which can explain: Which agent opened which file? When did it make changes? To whom did it send information?

Every such movement must be recorded (logging). Because of this, if an accident happens by mistake, we can immediately identify which agent caused it and rectify the issue.