28Add-on treatments are the new black. They are provided (most frequently, sold) to people undergoing 29 in vitro fertilization on the premise that they will improve the chances of having a baby. However, the 30 regulation of add-ons is consistently minimal, meaning that they are introduced into routine practice 31 before they have been shown to improve the live birth rate. Debate over the adequacy of this light-touch 32 approach rages. Defenders argue that demands for a rigorous approval process are paternalistic, since 33 this would delay access to promising treatments. Critics respond that promising treatments may turn 34 out to have adverse effects on patients and their offspring, contradicting the clinician's responsibility to 35 do no harm. Some add-ons, including earlier versions of PGT-A, might even reduce the live birth rate, 36 raising the prospect of desperate patients paying more to worsen their chances. Informed consent 37 represents a solution in principle, but in practice there is a clear tension between impartial information 38 and direct-to-consumer advertising. Because the effects of a treatment can't be known until it has been 39 robustly evaluated, we argue that strong evidence should be required before add-ons are introduced to 40 the clinic. In the meantime, there is an imperative to identify methods for communicating the associated 41 risks and uncertainties of add-ons to prospective patients. 42 43 Capsule 44 How should IVF add-ons be regulated? Is it ethical to provide unproven treatments? How can we inform 45 patients about the risks and uncertainties? 46. 47