GECKO CEO: HUMANS MUST CONTROL PHYSICAL AI

Loosararian says the next frontier for AI safety involves maintaining human oversight of autonomous systems deployed in robots across military and industrial environments. The partnership with Nvidia focuses on creating guardrails that allow AI agents to operate within defined parameters while preventing unintended actions. The initiative addresses growing concerns about autonomous systems in high-stakes settings. Rather than positioning safety as an obstacle to deployment, Loosararian contends that robust safety frameworks and rapid advancement can coexist. Gecko's open agent safety platform aims to establish industry standards for constraining AI behavior in physical applications. The approach reflects a broader shift toward responsible AI deployment as robotic systems become increasingly prevalent in critical infrastructure and defense sectors. The work underscores ongoing tension between accelerating AI capabilities and implementing protective measures before widespread real-world deployment.