AI TOOL COMPARISON REVIEW - TWO-RUN WORKSHEET - VERSION 1

Input version:
Run A date, tool and visible settings:
Run B date, tool and visible settings:
What changed between runs?

Keep both original reviews. For each case, record meets check, needs
correction, or could not test, then note the evidence and next test.

Case    Run A       Run B       Evidence or unsupported claim
CMP-01  __________  __________  _________________________________
CMP-02  __________  __________  _________________________________
CMP-03  __________  __________  _________________________________

Did either response turn an unmatched, unrun or historical test
into a current conclusion?
What one check would make the comparison more useful?