IIT Madras' Ravindran: agentic AI changes the risk equation, guardrails must keep pace
ravi_iitm · x · 2026-09-29
In an interview with DT Next, Prof Balaraman Ravindran, head of IIT Madras' Wadhwani School of Data Science and AI, argues AI is shifting from chatbots to agents that plan tasks, talk to other AI systems and act on users' behalf — expanding capability and the risk of uncontrollable unintended consequences. He says even one life lost to an AI action would be too high a cost, calling for strong guardrails, safety engineering, and narrow custom-built harnesses. He also notes public attention overlooks AI-for-science work like genome and mutation understanding, beyond the race for bigger models.
More from AGI Musings
- Redditors debate: AI devs call for a pause while racing ahead — concern or cult? — Henrygrins · 2026-09-29
- Missing continual learning, not motivation, is what keeps models short of AGI, argues teortaxesTex — teortaxesTex · 2026-09-29
- MIT study: algorithmic monoculture in hiring isn't always bad — context and ensembles matter — nordicinst · 2026-09-29
- Sherry Turkle's new book 'Artificial Intimacy' warns chatbots erode empathy and social skills — nordicinst · 2026-09-29
- MIT study: algorithmic monoculture's harms depend on details, ensembles may help — MIT News AI · 2026-09-29
- MIT's Sherry Turkle warns chatbots' pretend empathy erodes human connection — MIT News AI · 2026-09-29