In Duran v. La Boom Disco, Inc., No. 17-cv-6331 (ARR) (CLP), 2019 U.S. Dist. LEXIS 30012 (E.D.N.Y. Feb. 25, 2019), Judge Duran sua sponte granted summary judgment to a TCPA defendant based on its absence of use of a ATDS.

The Second Circuit, however, only analyzed the meaning of “capacity” under “the statutory language itself.” As discussed earlier, the court declined to weigh in on the requisite functions of an autodialer. Importantly, the invalidation of the 2015 [*27]  Order does not have the same impact on “capacity” as it does on autodialer functions. Regarding “capacity,” the invalidation of the 2015 Order forces courts to analyze “capacity” under the statute, because the FCC had not offered prior guidance on the term.11 The FCC had, however, offered prior guidance on requisite autodialer functions, see 2003 Order, supra, ¶¶ 129-33; 2008 Order, supra, ¶¶ 12-13; 2012 Order, supra, ¶ 2 n.5, so the invalidation of the 2015 Order does not necessarily send courts back to the statute to analyze these functions. I understand the Second Circuit’s statement that “an administrative interpretation of the statute . . . has now been invalidated” and it must “consider the meaning of the statute independently, without an administrative interpretation to defer to” as limited to the meaning of the term “capacity.” King, 894 F.3d at 482 (holding that “the best interpretation of the statutory language is the one suggested by the D.C. Circuit’s discussion in ACA International: in the TCPA’s definition of an autodialer, a device’s ‘capacity’ refers to its current functions absent additional modifications”). I find support for this interpretation in the Second Circuit’s decision not to mention the prior FCC Orders or weigh in on the functions [*28]  an autodialer must possess. See id. at 481-82. Thus, the critical question for the courts in this circuit is whether invalidating the 2015 Order implicitly invalidates the analogous portions of the prior FCC Orders discussing autodialer functions. For the following reasons, I conclude that it does not. While courts in this circuit do not appear to have addressed the issue, a growing number of courts in other circuits have. See, e.g., Richardson, 2018 U.S. Dist. LEXIS 212558, 2018 WL 6622996, at *4-5 (collecting cases). Many courts have upheld the validity of the prior FCC Orders, in large part because ACA International does not clearly address the validity of these Orders. See, e.g., Reyes v. BCA Fin. Servs., Inc., 312 F. Supp. 3d 1308, 1321 (S.D. Fla. 2018) (“[N]owhere in the D.C. Circuit’s opinion are the prior FCC orders overruled.”); see also Richardson, 2018 U.S. Dist. LEXIS 212558, 2018 WL 6622996, at *4 (collecting cases); Pl.’s Reply 12-14 (same). Courts finding that the invalidation of the 2015 Order implicitly invalidates the prior FCC Orders generally engage in a two-part analysis. First, the courts cite to the D.C. Circuit’s holding that the 2015 Order is invalid because it offers competing views as to whether an autodialer must be able to generate random or sequential numbers. See Thompson-Harbach v. USAA Fed. Sav. Bank, No. 15-CV-2098-CJW-KEM, 2019 U.S. Dist. LEXIS 3687, 2019 WL 148711, at *10-11 (N.D. Iowa Jan. 9, 2019); Richardson, 2018 U.S. Dist. LEXIS 212558, 2018 WL 6622996, at *5; Pinkus v. Sirius XM Radio, Inc., 319 F. Supp. 3d 927, 934-35 (N.D. Ill. 2018); Sessions v. Barclays Bank Del., 317 F. Supp. 3d 1208, 1212-13 (N.D. Ga. 2018). Second, the courts [*29]  conclude that the 2003 Order must also offer a competing view, because as the D.C. Circuit noted, both the 2003 and 2015 Orders drew a distinction between numbers that are randomly or sequentially generated and numbers that come from lists. See Thompson-Harbach, 2019 U.S. Dist. LEXIS 3687, 2019 WL 148711, at *11; Richardson, 2018 U.S. Dist. LEXIS 212558, 2018 WL 6622996, at *5; Pinkus, 319 F. Supp. 3d at 935; Sessions, 317 F. Supp. 3d at 1212. This line of reasoning is incorrect. As discussed earlier, the 2003 Order clearly holds that equipment that calls from a list can meet the statutory definition of an autodialer. See supra Section II.A.i—ii; see also Maes v. Charter Commc’n, 345 F. Supp. 3d 1064, 1069 (W.D. Wis. 2018). While the D.C. Circuit did bring attention to the 2003 Order’s distinction between randomly or sequentially generated numbers and a list of numbers, it did so only to reinforce its understanding of the 2015 Order, i.e., that the FCC’s use of “dialing random or sequential numbers” meant “generating those numbers and then dialing them.” See ACA Int’l, 885 F.3d at 702. Only the 2015 Order contained a contradiction, however, by stating that equipment must have the capacity to “dial random or sequential numbers” in order to qualify as an autodialer. See supra Section II.A.ii; see also Maes, 345 F. Supp. 3d at 1068 (“[T]he flaw in the 2015 ruling was not that it reaffirmed the 2003 order, but that it both reaffirmed the 2003 order and contradicted it.”). In fact, the D.C. Circuit [*30]  expressly declined to endorse either interpretation of an autodialer. See ACA Int’l, 885 F.3d at 702-03 (“The 2015 ruling . . . gives no clear answer (and in fact seems to give both answers). It might be permissible for the Commission to adopt either interpretation. But the Commission cannot, consistent with reasoned decisionmaking, espouse both competing interpretations in the same order.”). Because the logic behind invalidating the 2015 Order does not apply to the prior FCC Orders, I conclude that the invalidation of the 2015 Order does not implicitly invalidate the prior FCC Orders. See also Maes, 345 F. Supp. 3d at 1068-70 (finding the same). I will therefore continue to rely on those Orders to interpret the definition of an autodialer. B. ExpressText and EZ Texting As discussed earlier, I interpret the prior FCC Orders as holding that equipment can meet the definition of an autodialer if it pulls from a list of numbers, so long as the equipment also has the capacity to dial those numbers without human intervention. See supra Section II.A.i. Plaintiff concedes that the programs at issue lack the ability to generate randomized or sequential numbers. See Pl.’s Br. 16 (“Their websites do not indicate that these systems have the capacity to actually [*31]  ‘produce’ telephone numbers ‘using a random or sequential number generator.’ Rather, the automated text messaging works off of a pre-established list of numbers that have been uploaded into the system.” (citation omitted)); see also Patel Aff. ¶¶ 8-9 (stating that ExpressText “does not have the ability to generate randomized or sequential lists of contact cell phone numbers” and “does not dial cell phone numbers or send text messages sequentially or randomly”); Ramos, 334 F. Supp. 3d at 1265 (“[T]he CEO of . . . the company that owns the EZ-texting program[] confirms that the program can only be used to send messages to specific identified numbers that have been inputted into the system by the customer . . . [and] does not have the ability to . . . generate phone numbers.”). Because I do not understand the prior FCC Orders as requiring random or sequential number generation, I do not find that the programs’ reliance on a database of numbers disqualifies them from TCPA coverage; rather, I conclude that they do not qualify as autodialers because they are not capable of dialing numbers without human intervention. There is no dispute that for the programs to function, “a human agent must determine [the time to send [*32]  the message], the content of the messages, and upload the numbers to be texted into the system.” Pl.’s Reply 16; see also Pl.’s Br. 14-16; Patel Dep. 10:10-12; Najera Dep. 40:7-11. Plaintiff argues that these programs operate without human intervention because “the user does not have to ‘click’ before each number is dialed. Rather, the user can simply direct the system to fire off thousands of texts at a designated time.” Pl.’s Reply 16. Neither the FCC Orders nor the relevant case law support plaintiff’s understanding of what it means to operate without human intervention. When the FCC expanded the definition of an autodialer to include predictive dialers, the FCC emphasized that “[t]he principal feature of predictive dialing software is a timing function.” 2003 Order, supra, ¶ 131 (emphasis added). Thus, the human-intervention test turns not on whether the user must send each individual message, but rather on whether the user (not the software) determines the time at which the numbers are dialed. See Blow v. Bijora, Inc., 191 F. Supp. 3d 780, 788 (N.D. Ill. 2016) (“The uncontested facts show that the technology in this case does not possess . . . the sine qua non of a predictive dialer—the ‘timing function.’ The FCC considers predictive dialers to fall within [*33]  the definition of an ATDS because the dialer itself, and not the user, decides when to call.” (citation omitted) (citing 2003 Order, supra, ¶ 131)), aff’d on other grounds, 855 F.3d 793 (7th Cir. 2017);12 see also Ramos, 334 F. Supp. 3d at 1275 (concluding that the EZ Texting program used by defendant “was not an ATDS” because “no text message would have been sent” if defendant “had not ultimately pressed ‘send’ to authorize the EZ-Texting platform to send the text message”). The Ramos court examined four cases dealing with EZ Texting or a similar program. See id. at 1274-75. In each case, the court granted summary judgment for the defendant and found that the program required too much human involvement to be an ATDS. See Herrick v. LLC, 312 F. Supp. 3d 792, 803 (D. Ariz. 2018) (“[Defendant] . . . had to . . . log into the system, create a message, schedule a time to send it, and perhaps most importantly, enter a code to authorize its ultimate transmission. As such, the text was not sent automatically or without human intervention and thus was not sent using an autodialer . . . .”), appeal docketed, No. 18-16048 (9th Cir. June 7, 2018); Luna v. Shac, LLC, 122 F. Supp. 3d 936, 941 (N.D. Cal. 2015) (“[H]uman intervention was involved in several stages of the process prior to Plaintiff’s receipt of the text message, and was not limited to the act of uploading the telephone number to the [*34]  . . . database . . . . As explained above, human intervention was involved in drafting the message, determining the timing of the message, and clicking ‘send’ on the website to transmit the message to Plaintiff.”), appeal dismissed, No. 15-16790 (9th Cir. Nov. 20, 2015); see also Jenkins v. mGage, LLC, No. 1:14-cv-2791-WSD, 2016 U.S. Dist. LEXIS 106769, 2016 WL 4263937, at *5-7 (N.D. Ga. Aug. 12, 2016); Blow, 191 F. Supp. 3d at 788.13 To support his understanding of human intervention, plaintiff relies on a number of cases dealing with predictive dialers. The programs at issue in this case, however, are not predictive dialers, and the predictive-dialer cases cited by plaintiff actually support the notion that a program does not qualify as an autodialer unless a computer determines when to send the message. See Espejo v. Santander Consumer USA, Inc., No. 11 C 8987, 2016 U.S. Dist. LEXIS 142259, 2016 WL 6037625, at *4 (N.D. Ill. Oct. 14, 2016) (holding that “there is little [human intervention] to speak of” as the “‘dialer’—not the agents—makes the calls ‘by dialing numbers from the uploaded list'”); Strauss v. CBE Grp., 173 F. Supp. 3d 1302, 1309 (S.D. Fla. 2016) (“To determine whether a dialer is a predictive dialing system, and therefore an ATDS, ‘the primary consideration . . . is whether human intervention is required at the point in time at which the number is dialed.'” (quoting Brown v. NRA Grp., No. 6:14-cv-610-Orl-31-KRS, 2015 U.S. Dist. LEXIS 73065, 2015 WL 3562740, at *2 (M.D. Fla. June 5, 2015));  [*35] In re Collecto, Inc., No. 14-MD-02513-RGS, 2016 U.S. Dist. LEXIS 16319, 2016 WL 552459, at *4 n.9 (D. Mass. Feb. 10, 2016) (distinguishing between programs that require human intervention solely “to activate the process (by assembling a list of numbers and uploading them to the dialer)” from programs where “human intervention [is] required to dial the target telephones”).14 While I agree with plaintiff that “[a]n unsolicited text delivered at, say, 5:00 PM will be equally obtrusive whether it was a human being or a computer who determined that it should be delivered then,” Pl.’s Reply 16, the requirement that a computer determine the time in order for the device to qualify as ATDS is not arbitrary. The FCC decided to include predictive dialers under the statutory definition of an ATDS because “[t]he legislative history . . . suggests that through the TCPA, Congress was attempting to alleviate a particular problem—an increasing number of automated and prerecorded calls.” 2003 Order, supra, ¶ 133. Presumably, programs with computer-run timing functions have the capability to barrage consumers at a higher rate than programs requiring more human involvement. In sum, because a user determines [*36]  the time at which the ExpressText and EZ Texting programs send messages to recipients, they operate with too much human involvement to meet the definition of an autodialer. Summary judgment for plaintiff is therefore improper. Further, since plaintiff’s motion turns on whether the programs qualify as autodialers, I conclude that “all of the evidentiary materials that [plaintiff] might submit in response to a motion for summary judgment [on this issue] are before the court.” Bridgeway Corp., 201 F.3d at 140. Because the record reveals no material dispute as to how the programs work, and I find that the programs are not autodialers as a matter of law, a sua sponte grant of summary judgment against plaintiff is appropriate. See id.