State of Iowa v. Max Amyda

Issue

What must the State show to authenticate a digital video that reached it by way of social media, where no witness saw it recorded and no witness can describe the device or process that made it? And does a defendant's assertion that the video is a deepfake raise a genuine question about the original's authenticity that bars admission of a duplicate under the best-evidence rule?

Rule

Iowa Rule of Evidence 5.901 requires evidence sufficient to support a finding that an item is what its proponent claims. The Iowa Court of Appeals held that rule 5.901(b)(1), testimony of a witness with knowledge, and rule 5.901(b)(9), evidence describing a process or system, are permissible methods but not the only ones: rule 5.901(b)(4) allows authentication by circumstantial evidence drawn from distinctive characteristics and other contents of the item together with all the surrounding circumstances. The court relied on the Iowa Supreme Court's decision in State v. Holderness, which upheld authentication of a photograph almost entirely on circumstantial proof. On the best-evidence side, Iowa Rule of Evidence 5.1002 requires an original to prove content, and rule 5.1003 makes a duplicate admissible to the same extent as an original unless a genuine question is raised about the original's authenticity or the circumstances make it unfair to admit the duplicate. Review of both rulings was for abuse of discretion.

Application

The video in Amyda ran about a minute and depicted the sexual abuse of a sleeping woman. It was never recovered from Amyda; it arrived in the victim's hands weeks after the night in question, forwarded through Facebook Messenger by her sister, who had received it from another family member. No witness had seen it filmed and no witness could describe the camera or process that produced it. The district court held an evidentiary hearing on authentication as a preliminary question under Iowa Rule of Evidence 5.104, heard from the victim, her sister, and the detective, and denied the motion in limine in a written ruling four days later, agreeing with Amyda that no witness could authenticate the video by the two methods he insisted on but concluding that circumstantial evidence sufficed. The Iowa Court of Appeals agreed, and its account of the circumstantial evidence is the useful part of the opinion. The victim testified she invited Amyda over, that they drank together, that they went to her bedroom, that she rejected his advances and fell asleep, and that she woke with him beside her. That placed him alone with her while she slept, and because she testified this was the only time she was ever alone with him, a jury could find it was the only occasion on which the recorded conduct could have occurred. She recognized her bed, pillow and clothing, and testified the clothes she wore in the video were the clothes she wore that night. She recognized her own body and a set of details she could describe with particularity. She recognized the hand in the video and the distinctive tattoo on it as Amyda's. She testified that she and Amyda communicated through Facebook Messenger, that the video circulated through Facebook Messenger, and that when she messaged him asking for an explanation he blocked her. Her sister corroborated the circumstances of the visit and independently recognized the victim, the clothing, the bedroom and the bed sheets. The court noted that Amyda offered no argument against the sufficiency of any of this, resting entirely on his position that only an eyewitness to the filming or a witness to the recording process could authenticate the video, which the court held was a wrong reading of rule 5.901. It collected decisions from Nebraska, Washington, Missouri and New Jersey that authenticated videos of sexual offenses largely on circumstantial proof, and it cited Mooney v. State, 321 A.3d 91 (Md. 2024), for the proposition that this remains so even though video is susceptible to alteration and image-generating artificial intelligence presents new challenges. The deepfake argument was pressed under the best-evidence rule rather than under rule 5.901. Amyda argued at the motion-in-limine hearing and throughout trial that the video may have been created through generative artificial intelligence and that the underlying conduct never occurred, which he said raised a genuine question about the original's authenticity and made the duplicate inadmissible under rule 5.1003. The Iowa Court of Appeals assumed without deciding both that the best-evidence rule reaches this kind of video and that the file the victim received was not an original, and it still affirmed. Amyda pointed to nothing in the record supporting the theory, and the court found nothing in the video itself suggesting artificial creation. It held that bare conjecture does not raise a genuine question about authenticity or make admission of a duplicate unfair, citing its earlier decision in Manning for the related point that editing or manipulation of a video generally goes to weight rather than admissibility.

Conclusion

The Iowa Court of Appeals affirmed Amyda's conviction for third-degree sexual abuse, holding the district court did not abuse its discretion on either the authentication objection or the best-evidence objection. Judge Langholz wrote the opinion after oral argument before a panel of Chief Judge Tabor and Judges Ahlers and Langholz; there was no separate writing and no dissent. It is an intermediate appellate decision, so it binds Iowa district courts but not the Iowa Supreme Court, and it carries no North Western Reporter citation yet.

Circumstantial evidence based on the “distinctive characteristics” and other contents of the video and all the surrounding circumstances can also be sufficient to support a finding that the video is what it claims to be

Majority, Langholz, J., summarizing the authentication holding

Amyda pointed to no evidence in the record supporting this speculative theory. And nothing about the video gives even the slightest hint that it is an artificial creation.

Majority, Langholz, J., rejecting the deepfake claim

speculative claims that the video was a “deepfake” without any evidentiary support did not raise a genuine question about the video’s authenticity

Majority, Langholz, J., stating the best-evidence holding

Why it matters. Amyda is the cleanest available statement that a deepfake objection has to be more than an assertion. The defendant raised it in limine and repeated it throughout trial, so this is not a case where the court brushed off a late or undeveloped argument, and the court still called it bare conjecture. Read next to Matter of M.S., decided three weeks later, it marks the two poles of how courts are currently handling the same argument.

Read the full opinion on CourtListener

This summary was written from the opinion itself. Every quotation above was located in the text of the decision it is attributed to. It is a summary for public education and is not legal advice.

Related reading