AI agents are gaining more autonomy and access to critical systems, and incidents are rising with them. Experts point to lax security, model design, and system glitches. Our research looks at the rise in AI-related incidents and the guardrails teams need.

Build or buy an AI SRE + DevOps platform? We built an illustrative three-year model for a combined SRE + DevOps platform and had the assumptions reviewed by a senior architect at a Fortune 50 organization.

We’ve been running AI agents in production for months, and cost is often the first sign of trouble. Traditional APM can’t tell you why an agent suddenly costs more or keeps repeating the same actions. Here’s what we’ve learned about monitoring #AIagents in production.

“We can build this ourselves.” Then come integrations, governance, audit trails, incident workflows, and maintenance. If you’re weighing build vs. buy for an AI SRE + DevOps platform, compare the cost, time, headcount, and burden behind each path. #AISRE #DevOps

AI now accounts for 1 in 10 incidents, a 6× rise in three years. Our research also found cases where AI agents acted independently and damaged live systems. Read the press release, now live on Business Wire: businesswire.com/news/home/2…

🕙1 day to go! Tomorrow at 11 AM ET, John Jamie and Aakash Dabrase will unpack findings from 177,960 status-page entries across 390+ companies, including what’s driving incidents and where AI fits into incident response. #aisre #incidentmanagement

The first thirty minutes of an incident are expensive. Sabith, Principal Engineer at StackGen, shares what the team learned from shipping incident triage into Aiden, including the engineering tradeoffs and runtime choices behind it. #AISRE #SRE

Two engineers can identify the same root cause and still follow different operational workflows.That's why the follow-up should be predictable. This walkthrough shows how Aiden keeps human approvals in the loop while executing predefined Skills across tools.