AI code generation has grown by 741%. Software releases grew by only 20%.

The gap between how fast AI writes code and how fast organizations can verify it is where the quality crisis lives. And according to new research from Sauce Labs and Wakefield Research, most organizations are already feeling it.

We surveyed 400 U.S. executives and engineering leaders. Here's what they told us:

• 80% have traced a production incident directly to AI-generated code
• 92% are concerned their current safeguards can’t catch AI failures.
• 66% admit their teams have compromised on quality to meet a release deadline
• 89% report positive ROI from AI testing tools — but ROI is measured on speed, not quality outcomes

The full report details what's driving the crisis, what it's costing organizations, and what closing the gap actually requires.