[mceachen]

Nope, current flagship models are very happy to make huge missteps across the whole development stack of design, planning, implementation, and testing -- but playing different models against each other can help catch more egregious issues.