AIOps: Prove It! An Open Letter to Vendors Selling AI for SREs
summary
Vendors selling AI for SREs must show data, failure modes, and exactly what was built and tested — not slogans like “your new SRE.” The talk demands concrete proofs: metrics from live incidents, false positive and false negative rates, who stays in the loop, and whether agents actually reduce mean time to repair or just create hero single points of failure. Promises to join / extend / ground operations need experimental results and rollback plans.
AI-written summary · about
chapters (13)
- 0:00Introduction to AI in SRE
- 1:24Evaluating AI vendor claims
- 3:52Learning from past mistakes
- 5:49Critical questions for AI
- 9:05Accountability and human role
- 11:28Effective AI implementation
- 14:54The role of SRE in risk
- 17:04Balancing risk and innovation
- 18:28Managing rapid industry change
- 21:27Bridging business and technical
- 24:58Adaptability and professional growth
- 27:33Embracing curiosity over fear
- 30:06Concluding thoughts
related talks
44mAgentic Observability — The Path to AI Co-SRE
42mDemystifying SRE and DevOps
16mHow AI is redefining SRE and Customer Experience
42mKeynote VC Panel SRE from Investment Perspective
40mWhen Incidents Fix Themselves: AI SRE in action
43mWhere SRE Can Go Wrong
28mDO WE STILL NEED TO "OBSERVE" THE FUTURE OF AI & OBSERVABILITY
26mBeyond Hype: Leading the Charge in Responsible AgentOps for Business Transformation