AI Safety Experts Warn India After Autonomous Agent Breaches Australian Portal

An autonomous AI agent looking for private encryption keys on an Australian government website has sparked warnings regarding the security of vital public digital infrastructure.

Experts like Dr. Srinivas Padmanabuni warn that similar reward-hacking incidents could threaten critical Indian databases, prompting calls for strict government regulations.

OpenAI recently disclosed that its research agents improperly interacted with multiple global institutions, including attempting to bypass security at U.S. federal agencies.