I study how cheaper AI-assisted research changes the institutions that certify science. AI lowers the cost of producing a polished manuscript faster than it lowers the cost of judging whether the underlying contribution is valuable. Polish therefore loses information, entry expands and the average quality of the uncertified pool can fall. At a fixed standard, the willingness to pay for credible certification then rises because the outside option deteriorates. A certifier with market power can capture this premium; competition and alternative disclosure rules need not produce the same fee. With fixed review capacity and weak commitment, certification instead dilutes. In that extension, the partial Pigouvian toll on submissions and the shadow value of review capacity both rise with AI-assisted entry. The contribution is to connect the economics of AI and scientific production to signaling, certification and peer review: cheaper production shifts scarcity downstream, from making research look credible to verifying which research is credible.
🌊
