Given the rationalization the scientist makes, its probably wise to be skeptical of his judgements overall.

A direct quid pro quo / payment for services rendered is more honest than this model which is Google buying the professor vs. buying specific research. That's a serious ethical issue.