Move Fast With AI Agents Without Breaking Prod
A live conversation on recovery, blast radius, and what good looks like when anything, a person, a script, or an AI agent, can wipe production in seconds
Our speakers
The mistake takes seconds. The recovery takes hours.
An AI agent can wipe a production database in seconds, recovery copies and all, and to every system watching, it looks like a routine API call.
This isn't a rare horror story anymore. Agents are running in testing or production at 81% of teams, but only 14% had security sign-off before going live (Gravitee, 2026). Everyone's handing agents real access faster than anyone's checking what they can reach, and it's going wrong more often than teams admit.
It's not just agents, either. A fat-fingered engineer, a runaway script, ransomware, they all end the same way: something legitimate touched production, and your recovery is the only thing between you and a very bad week. Agents just make it faster and quieter, and when one gets something wrong, it'll usually tell you it went fine.
The part people miss: this is an IAM and blast-radius problem, not a prompt problem. Guardrails and prompts buy you time, but they don't hold. Most teams find that out the hard way, in the restore, not the policy. In Eon's 2026 Cloud Data Infrastructure Report, 60% of teams said a full restore takes six or more hours. Only 5% can do it in under an hour. Snapshots aren't built for how fast recovery has to happen now.
Join this session live to learn how to:
- Spot how legitimate access, human or agent, actually breaks prod
- Treat recovery as an IAM and blast-radius problem, so one compromised credential can't take your recovery with it
- Prove backup coverage across every account and resource type, and catch drift before it's a gap
- Restore at the row, file, or table level instead of rehydrating everything
- Keep read-only agents off production, and what to point them at instead
- Get ready for what's next as AI-native threats reshape recovery