Meridian is standing up its computer-vision ramp-safety program and multiple teams — IT, safety, and ground operations — all touch the camera-footage data differently. Who should the PM ensure is explicitly assigned to own data quality, access rules, and definitions for this dataset?

Data stewardship establishes clear accountability for a dataset's quality, definitions, and access rules — critical when, as at Meridian, multiple functions (IT, safety, ground ops) interact with the same camera-footage data. A designated steward resolves ambiguity about who decides labeling standards, who approves access, and who is accountable when quality slips. The option claiming no one needs to own the data is the failure mode governance exists to prevent — undefined ownership reliably leads to the 'data swamp' problem from Q144. The option saying whichever team touches the data last should own it describes de facto, accidental ownership rather than deliberate stewardship, which produces inconsistent standards as different 'last touchers' rotate. The option assigning ownership to the AI vendor incorrectly places internal governance responsibility on an external vendor; the vendor built the model, but Meridian owns and is accountable for its own operational data, including under any regulatory scrutiny. The exam distinction to hold onto: stewardship is a named, assignable role within data governance, not a synonym for 'whoever happens to be using the data.'