The tweet was deleted by the author.
But we saved everything 🙂.
Questions about the readiness of enterprises to deploy AI agents with full autonomy are gaining attention.
Nikesh Arora expressed doubt that organizations are equipped to manage so-called agents with ''agency,'' citing hurdles such as visibility, governance, the ability to intercept actions, and audit capabilities. Currently, most AI agents in business settings are limited to functions such as data summarization, collation, and analytics, rather than operational independence. The comments highlight a cautious stance on pushing for broader uses of autonomous AI before robust oversight frameworks are in place.
Arora previously reported that AI adoption is a main focus for European customers, following more than 200 meetings across the region (link). He has also described how Lightwell uses virtual patching to address open source vulnerabilities within an hour (link). These discussions reflect ongoing attention to practical and secure AI deployment in enterprise environments.