An increasing number of judges, legislators, and scholars, particularly in the United States, have wrongly come to believe that the commitment that standard-essential patents be licensed on “fair, reasonable, and nondiscriminatory terms” (“FRAND”) was principally created to advance the interests of technology implementers, and have too often given a preference toward implementers’ interests in interpreting FRAND. That premise has led American courts to take a categorically hostile view toward awarding injunctions against implementers who infringe valid standard-essential patents, fearing that the injunctive remedy would give innovators undue leverage. Indeed, American courts have been so unilaterally concerned with innovators’ conduct that some have even allowed implementers to sue innovators simply for making an opening licensing offer that is later deemed “too high,” even if the implementer refused to make any counteroffer at all. An implementer–centric view of FRAND has also caused several courts to conclude that innovators are not entitled to any share of the commercial benefits arising from the standardization of their technologies, and that all such benefits must go to implementers alone. This Article argues that an implementer–centric view of FRAND’s origins and purposes is false. FRAND is a contractual agreement that reflects a voluntary reciprocal exchange of benefits and obligations driven by the need to solve significant coordination problems in the face of otherwise prohibitive transaction costs. As part of that bargain, innovators agree to disclose their latest, confidential discoveries to standard–development organizations and to waive their injunction rights as to eventual patents on those discoveries, in exchange for contractual protection against “patent holdout” by implementers. Those implementers are then permitted to use standard–essential patents on the condition that they agree to pay fair and adequate royalties for that use, with the royalty amount to be set through mutual good faith negotiations. Accordingly, this Article stresses that FRAND is not intended to be, and should not be interpreted as, a one-sided transfer from innovators to implementers. Rather, implementers too owe a significant duty to negotiate FRAND licenses in good faith—a duty that many courts have overlooked and underenforced. This Article demonstrates that implementers’ good faith obligations are a critical component of the basic FRAND architecture and that enforcement of those obligations is strictly necessary to the continued development of innovation–driven standards. This Article further observes that the FRAND bargain is not simply meant to give innovators a way to monetize their intellectual property. Rather, and perhaps more significantly, FRAND creates an agreed bargaining framework that allows implementers to access innovators’ otherwise confidential discoveries—inventions so recent that they are not otherwise disclosed in patents or published applications. In this way, FRAND supplies a solution to an iteration of Kenneth Arrow’s paradox of information, enabling the standards development effort to yield commercial benefits that would not exist absent innovators’ voluntary participation. Stated otherwise, innovators agree to give implementers access—and a fair license—to their most groundbreaking technologies because innovators believe that implementers will reciprocally later agree to take a license in good faith for using those highly valuable innovations. This Article shows both theoretically and empirically that courts’ failure to appreciate these aspects of the FRAND bargain, combined with their overreliance on liability rules (i.e., damages over injunctions) incentivizes the very patent holdout problem FRAND was intended to avoid. That “efficient infringement” outcome, in turn, has motivated innovators to reduce their participation in FRAND bargains, threatening to unravel a massive innovation–commercialization marketplace and its innumerable positive externalities for all parties. To reverse these harms, this Article recommends that courts automatically issue an injunction where an implementer is found to infringe valid FRAND–committed patents that it did not attempt to license in good faith. This Article also recommends that a proper FRAND licensing rate should include some portion of the benefits achieved through standardization of the innovations in question. More broadly, this Article suggests that courts, policymakers, and academic commentators have wrongly favored implementation over innovation—“things” over ideas—unwisely frustrating the emergence of an “ideas economy” that should rightly assign significant profits to upstream innovators and not to the low–margin manufacturing firms that specialize in turning those innovations into tangible products.