TechCrunch AI’s report reveals a paradox: Google Cloud’s push for AI-native defenses coincides with its own security missteps, such as unauthorized API access to Gemini models. While de Souza advocates for multicloud security and agentic defenses, Google’s automated billing policies and delayed key revocation reportedly expose developers to significant risks. This disconnect suggests that even as AI accelerates defensive capabilities, the industry may be racing to patch vulnerabilities faster than it can create them. The broader question is whether platform providers like Google can balance innovation with accountability.