(dissenting and voting to reverse the judgment of conviction, on the law and the facts, and to dismiss the indictment): During summations in this bench trial, the prosecutor urged the trial court to convict the defendant on a theory that was not charged in the indictment. When the defendant objected and clearly stated the ground for his objection, the court overruled the objection, opaquely adding later that it was bound by the case file and would make a decision based on the evidence. Because the prosecutor’s change in theory violated the defendant’s constitutional right to be prosecuted by indictment voted by a grand jury (see NY Const, art I, § 6; People v Perez, 83 NY2d 269, 273 [1994]; People v Grega, 72 NY2d 489, 495 [1988]), I vote to reverse his conviction, and, because any conviction on the theory that was charged in the indictment was against the weight of the evidence, I vote to dismiss the indictment.
Penal Law § 235.22 (disseminating indecent material to minors in the first degree) provides in pertinent part:
“A person is guilty of disseminating indecent material to minors in the first degree when:
1. knowing the character and content of the communication which, in whole or in part, depicts . . . sexual conduct . . . which is harmful to minors, he intentionally uses any computer . . . to . . . engage in such communication with a person who is a minor; and
2. by means of such communication he importunes, invites or induces a minor to engage in . . . sexual contact with him, or to engage in . . . sexual conduct for his benefit.’’[emphasis added].
The indictment, which contained five counts of attempted dis*917seminating indecent material to minors in the first degree, charged the defendant only under the theory that the defendant sought to lure a person he believed to be a minor into sexual contact with him (hereinafter the physical-contact theory). The bill of particulars was consistent with the indictment, limiting the charges to the physical-contact theory. The indictment did not charge that the defendant sought to have a person he believed to be a minor engage in sexual conduct for the defendant’s benefit (hereinafter the defendant’s-benefit theory). The defendant’s-benefit theory could, if charged, have permitted a conviction without evidence that the defendant sought to be in the minor’s physical presence at any time. In other words, the defendant’s-benefit theory would have been satisfied with proof that the defendant importuned, invited, and induced the person he believed to be a minor to engage in sexual conduct, e.g., masturbation, during their cyber chats, for the defendant’s benefit, e.g., the defendant’s arousal.
Nevertheless, perhaps concerned that the evidence on the physical-contact theory was too weak to sustain a conviction—a concern that, as I discuss below, would have been well founded— the prosecutor urged the Judge in this bench trial to find the defendant guilty on the uncharged defendant’s-benefit theory. By overruling the defendant’s precise objections to this impermissible alteration in the theory of the prosecution, and thereby allowing the People to constructively amend the indictment, the County Court erred (see People v LaPetina, 34 AD3d 836, 839 [2006], affd 9 NY3d 854 [2007]; People v Taplin, 1 AD3d 1044, 1045 [2003]; People v Brown, 221 AD2d 353, 353-354 [1995]; People v Chicas, 204 AD2d 476, 477 [1994]).
My colleagues in the majority do not state any contrary view as to the impropriety of the People’s change in theory. Instead, they conclude that, because this was a bench trial, and in light of the court’s comment after summations, there is no danger that the fact-finder convicted the defendant on an improper basis (but see People v Haines, 139 AD2d 591, 592 [1988]; but cf. People v Giocastro, 210 AD2d 254 [1994] [reversal after bench trial on weight of the evidence]). I cannot share their confidence in that conclusion because it fails to take into account the fact that the court overruled the defendant’s objection to the improper amendment and never disclaimed reliance on the defendant’s-benefit theory. In short, because the Trial Judge, as the trier of fact, never abjured the uncharged defendant’s-benefit theory, he may have convicted the defendant on an accusation for which the grand jury failed to indict him (see CPL 200.70 [1]; People v Grega, 72 NY2d at 498; People v Kaminski, *91858 NY2d 886, 887 [1983]; cf. People v Martinez, 83 NY2d 26, 32-35 [1993], cert denied 511 US 1137 [1994]; People v Shealy, 51 NY2d 933, 934 [1980]; People v Barnes, 50 NY2d 375, 379 n 3 [1980] [nonjury trial]; People v Rodriguez, 164 AD2d 832, 833 [1990] [nonjury trial]; People v Haines, 139 AD2d 591, 592 [1988] [nonjury trial]). Thus, I conclude that the convictions must be reversed.
The only remaining question is the appropriate corrective action {see CPL 470.10 [3]). Because any conviction on the physical-contact theory was against the weight of the evidence, I conclude that the indictment must be dismissed, rather than that the case be remitted for a new trial on that theory. I agree with the majority that a conviction on the physical-contact theory was supported by legally sufficient evidence, if the Internet communications are read in fragments, isolated from their full context. In determining whether the evidence was legally sufficient, we must, of course, consider whether “after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt” (People v Contes, 60 NY2d 620, 621 [1983], quoting Jackson v Virginia, 443 US 307, 319 [1979]; see People v Conway, 6 NY3d 869, 872 [2006]; People v Santi, 3 NY3d 234, 246 [2004]; People v Rossey, 89 NY2d 970 [1997]). “Ultimately, so long as the evidence at trial establishes ‘any valid line of reasoning and permissible inferences [that] could lead a rational person’ to convict, then the conviction survives sufficiency review” (People v Santi, 3 NY3d at 246, quoting People v Williams, 84 NY2d 925, 926 [1994]; see People v Bleakley, 69 NY2d 490, 495 [1987]). Thus, I agree that, measured by that standard, the evidence of the defendant’s guilt under the physical-contact theory is legally sufficient.
Nevertheless, I conclude that any verdict of guilt at this trial on that theory—the only one we may consider because it was the only one with which the defendant was charged—was against the weight of the evidence {see CPL 470.15 [3] [b]; [5]), particularly when the communications are considered in their entirety and in full context. As the Court of Appeals has so recently reaffirmed, the first question we must ask on factual review of a verdict is whether an “acquittal would not have been unreasonable” (People v Danielson, 9 NY3d 342, 348 [2007]; see People v Romero, 7 NY3d 633, 643 [2006]; People v Bleakley, 69 NY2d at 495). Here, I conclude that an acquittal would not have been unreasonable. Therefore, we must “weigh conflicting testimony, review any rational inferences that may be drawn from the evidence and evaluate the strength of such
*919conclusions. Based on the weight of the credible evidence, [we must then] decide[ ] whether the [fact-finder] was justified in finding the defendant guilty beyond a reasonable doubt” (People v Danielson, 9 NY3d 342, 348 [2007]). Doing so in this case, I agree with the words of Judge Frederick Crane, who, writing for the Court of Appeals in People v Crum (272 NY 348, 350 [1936]), said, “[A] reading of this record causes me to hesitate. I am not convinced.” First, it must be observed that, unlike in most weight-of-the-evidence reviews, credibility of witnesses plays little part in the analysis here; the evidence was almost entirely supplied by the actual written communications between the defendant and Investigator Pascal Storino, writing as “John in Yonkers 914,” purportedly a 14-year-old boy. Thus, our factual review is composed almost entirely of “reviewing the] rational inferences that may be drawn from the evidence and evaluating] the strength of such conclusions” (People Danielson, 9 NY3d 342, 348 [2007]). Upon my reading of the communications between the defendant and Investigator Storino, I conclude that it was the investigator who importuned, invited, and induced the defendant to meet for the purpose of actual sexual contact and not the other way around. The defendant was reluctant, and he agreed to meet at all only after Investigator Storino persisted over a lengthy period of time. Consequently, I am persuaded that any verdict of guilt under the physical-contact theory was against the weight of the evidence. Given this conclusion, a retrial on this theory is forbidden by statute (see CPL 470.20 [5]; People v Romero, 7 NY3d at 644 n 2). Instead, the only proper corrective action is to dismiss the indictment (see CPL 470.20 [5]).
By vivid contrast, the evidence that the defendant importuned, invited, or induced a person he believed to be a 14-year-old boy to engage in sexual conduct for his benefit is overwhelming; it is all over this record, and, had the defendant been charged under this theory, convictions on all five counts charged in the indictment would have been almost certain. The grand jury, however, did not charge the defendant under this theory—and the People never took the simple step of obtaining a superseding indictment—and it is fundamental, therefore, that the fact-finder could not convict him of any count under this theory (see People v Haines, 139 AD2d 591, 592 [1988]).
I respectfully dissent.