AI tools are becoming powerful enough to find security weaknesses, write complex code and act with less or no human guidance. That is why some systems are being marked “critical” and subjected to stronger safety checks before wider release.
The problem is not only that humans may design weak guardrails. It is also that advanced AI systems can behave in unexpected ways when placed in complex situations. A safeguard that works today may fail when the next, more capable version finds a new path around it.
This does not mean AI must be stopped. It means development cannot move faster than our ability to test, monitor and control it. Governments, companies, researchers and independent experts must work together, because no single organisation can manage these risks alone.
For ordinary people, the lesson is simple: powerful technology must come with powerful responsibility. We should welcome useful AI, but demand transparency, human accountability and an emergency brake whenever its actions become difficult to predict.
AI MUST GROW WITH HUMAN CONTROL—NOT BEYOND IT.
Sanjay Sahay
Have a nice evening.