Something shifted in security operations over the last two years: AI stopped being a pilot program and became the plan.
And if you survey 500 security professionals on whether that's going well – as Omdia did, commissioned by Rapid7 – you get a remarkable level of consensus: 97% report positive outcomes, 98% say AI reduces alert fatigue, and 95% say it's helping address staffing shortages.
Those numbers are high enough that the story could stop there; AI is working, everyone agrees. Move on. But there's a more interesting finding sitting underneath that consensus, and it tells you something important about where security operations is actually headed.
The confidence gap nobody is talking about
While frontline SOC teams report strong confidence in AI, executive security leaders like CISOs, CSOs, and VPs are taking a considerably harder look.
The research found that executive leaders are 1.6 times more likely than operational security managers to be highly concerned about how AI vendors handle their organization's security data. This isn't a contradiction of the 97% positive sentiment, but rather a maturity signal.
When AI was experimental, the question was: does it work? Operational teams answered that. Now that AI is embedded in production SOCs, a different question is arriving at the executive level: how do we govern it? Who is accountable when something is wrong? What happens when the model misses a critical threat, and whose job was it to catch that?
These are board-level questions, and the research suggests that the organizations who don't have good answers are about to feel that gap acutely.
The human-AI balance is the real differentiator
92% of respondents said AI enhances rather than replaces human analysts. But 41% are actively worried about over-reliance, or that teams could start deferring to AI in moments where an experienced analyst would have caught something the model missed.
The best security teams aren't choosing between AI and human expertise. They're building operating models where AI handles the volume – triaging, pattern matching, initial investigations – while analysts lead the decisions that require context, creativity, and accountability.
The question for any SOC leader isn't 'should we use AI?' It's 'where does AI create capacity without creating new blind spots?'
MDR is being redefined
86% of respondents believe AI-enabled MDR has a clear advantage over traditional approaches. And when asked what they actually expect from an AI-enabled MDR provider, the top answer wasn't faster detection or higher automation rates. It was transparency.
55% said their top expectation is visibility into how AI decisions are being made. 53% want AI explicitly integrated with human analyst expertise. 52% want regular updates on model performance and accuracy. 'We use AI' is not a differentiator anymore. What buyers now want to know is: can you show me exactly how?
What this means for your security strategy
The Omdia research gives security leaders something most AI content doesn't: an independent benchmark for where the market actually is.
Use it to pressure-test your AI governance posture, to reframe conversations with your board, and to evaluate whether your MDR provider can answer the transparency questions that 55% of buyers are now asking.
The full report covers all of this in detail, including where organizations remain most cautious, what makes AI adoption succeed or stall, and what the next phase of AI-enabled security operations looks like.
