> Undermining the credibility of computer science research is the best possible outcome for the field, since the institution in its current form does not deserv

> Undermining the credibility of computer science research is the best possible outcome for the field, since the institution in its current form does not deserve the credibility that it has. This is obviously somewhat tongue in cheek, but I’m not sure the author appreciates the full implications of how this would play out. Research funding for AI exists in competition with funding for all other fields. Revealing deep-seated BS would serve mostly to decrease funding commitments to AI projects in favor of other fields, or away from R&D entirely. It would be bad for all researchers (but if the true level of fraud is that high, and I have no doubt that it is, this may simply be an inevitable market correction). Other comments in this thread seem to miss the point as well: “why don’t universities select for quality not quantity”? Well, the change has to start with the funding sources for the unis to care.