The old cybersecurity model is breaking
- ID
- 27793
- Status
- summarized
- Published
- 24 Sep 2026, 1:44 AM
- Fetched
- 24 Sep 2026, 2:16 AM
- Provider
- TechCrunch
- Category
- technology
- Original URL
- https://techcrunch.com/video/the-old-cybersecurity-model-is-breaking/
- Source URL
- https://techcrunch.com/feed/
Summary
- Score
- 4.5
- Created
- 24 Sep 2026, 2:16 AM
- Tags
- Audience
- developerssaas_startup_foundersai_agent_users
What happened
Index Ventures partner Shardul Shah argues that periodic, human-in-the-loop cybersecurity review can't keep pace with AI agents, driving massive capital into AI-native security startups like Instinct and Simile, which are landing nine-figure rounds at valuations that wouldn't have made sense a few years ago. The segment ties this trend to Google's $32B acquisition of Wiz earlier in 2026.
Why it matters
If you're shipping AI agents that take autonomous actions, traditional periodic security audits and manual review checkpoints are becoming a competitive bottleneck. Builders should evaluate whether their agent pipelines need real-time, AI-native guardrails rather than relying on human-in-the-loop gates that can't scale with agent speed.
Discussion angle
For teams already deploying agents: what's your actual security review cadence today, and is it fast enough to catch a rogue agent before it does damage — or are you just hoping human review catches up?