Teams accumulate significant gaps in test coverage and hence the risk of regressions across versions compounding cost and effort of devops.

Independent development and release of services leads to increased risks of bugs due to subtle mismatches in assumptions.

Software defects, especially regressions to existing workflows, quickly erode customers’ trust in vendors' ability to meet their needs.

If I were to choose an area of testing where we need significant improvements from using AI techniques, it would be functional API testing.

For many applications, functional correctness must be established before starting a canary.