Google Cloud COO warns AI security must be built in
Get the Tech newsletter
Daily tech — startups, AI labs, chips, the launches that shape the next decade. Free.
- Francis de Souza says security must be built into AI platforms from the start, warning against “shadow AI” and urging governance and auditability.
- Google Cloud adopts a multicloud security posture, emphasizing consistent protection across clouds and SaaS partners.
- Attack timelines have collapsed, with the average time between an initial breach and the next stage dropping from eight hours to 22 seconds.
- SharePoint servers can be exposed by AI agents roaming internal systems, surfacing forgotten data assets.
- AI‑native defense agents are emerging to meet machine speed with machine speed, enabling fully‑agentic security.
Why it matters: Enterprises that adopt AI‑native security agents can keep pace with 22‑second breach cycles, protecting costly legacy data and avoiding costly breaches that could run into millions.
