Federal Rule 707 and AI-generated evidence: what changes in US federal court
Law firms and in-house teams put more digital evidence into federal litigation every month: screenshots of executive chats, recordings of video depositions, documents pulled from the web and, increasingly, output produced by AI systems. The problem is that the evidentiary standards used to weigh this material were written before generative AI existed.
Draft Federal Rule 707 is meant to close that gap. FRE 707 extends to machine-generated and AI-produced content, when it is offered as evidence without an expert witness to testify, the same reliability standard that Rule 702 applies to expert testimony. The existing authentication rules, Rule 901 and Rule 902(13), stay anchored to a pre-synthetic era: they show that a record came from a system, not that a plausible-looking piece of content was never fabricated.
So the operational question for anyone bringing evidence into a federal courtroom comes down to one thing: how do you produce digital evidence that holds up to the new reliability standard? The answer is to move the integrity guarantee to the moment of acquisition, instead of trying to reconstruct it once authenticity is already in dispute. That is the ground TrueScreen works on, certifying photos, video, screenshots, video calls and documents at the source, with a qualified eIDAS timestamp and a seal applied through an integrated third-party QTSP.
What is Federal Rule of Evidence 707 and what changes for AI-generated evidence
Federal Rule of Evidence 707 is a proposed rule, not yet in force, that applies the Rule 702 reliability standard to machine-generated and AI-generated evidence. When a party offers output produced by a system, without a human expert vouching for it, that output has to be as reliable as an expert's testimony would be.
The draft of FRE 707 was published by the Advisory Committee on Evidence Rules in August 2025, with public hearings in January 2026 and the public comment window closing on February 16, 2026. It is now under revision as part of the adoption process. The core of the rule is its reference to Rule 702: machine-generated content offered as evidence has to rest on sufficient facts and data and on reliable methods applied correctly. Basic scientific instruments stay exempt, things like thermometers, scales and clocks, whose operation raises no real doubt. FRE 707, the variant most people actually search for, shifts the center of gravity: it is no longer enough for a piece of data to exist, it has to be demonstrably reliable.
Where it comes from: the gap left by pre-AI standards
Federal Rule 707 exists to fill a grey zone. Until now, output from a language model, a synthetic video or a summary produced by an algorithm fell between two categories that do not describe it well. On one side, Rule 702 was built for human expert testimony and assumes a person who can explain the method. On the other, Rule 901 and Rule 902(13) were built to authenticate documents and electronic records: they check origin, not the reliability of the content itself.
The effect was that AI-generated evidence could enter a proceeding with no real reliability screening, simply because no rule caught it specifically. FRE 707 closes that opening and extends the Daubert factors to machine output as well. The deeper concern is the admissibility of digital evidence once its genuineness can no longer be assumed.
Where the adoption process stands
Federal Rule 707 follows the Rules Enabling Act path, which is long and runs through several stages. After the Advisory Committee draft and the close of comments in February 2026, the text moves to the Standing Committee, then the Judicial Conference, then the Supreme Court, and finally Congress, which has a window to act.
Realistically, the rule would not take effect before December 1, 2027. For litigators and in-house counsel, that means it is not applicable yet, but the direction is clear. Anyone building digital evidence today that needs to hold up over the coming years should already be moving toward acquisition processes that can demonstrate reliability and integrity.
The limits of current authentication standards against synthetic content
Today's authentication rules are not equipped for synthetic content. Rule 901 asks, in general terms, for proof that evidence is what its proponent claims it is; Rule 902(13) allows self-authentication of records generated by an electronic process or system, through a qualified person's certification. Both answer the question "where did this file come from," not the question "is this content genuine or was it fabricated to look that way."
Unlike Federal Rule 707, the authentication standards in force predate generative AI on their digital side and address a different problem. Rule 901 and Rule 902(13) were written to certify that a record comes from a trustworthy system: server logs, database extracts, file metadata. Neither was conceived for a synthetic video that looks shot on a real camera, or a screenshot reproducing a conversation that never happened. The forensic expert tasked with authenticating digital evidence is left with tools that verify the data's transmission chain, not its genuineness at the source. That is the weak point the academic debate, including the "Deepfakes on Trial" paper circulated in uscourts circles, has brought into focus: self-authentication holds up only until the fake becomes indistinguishable from the real thing.
The table below summarizes what each rule solves and what it leaves uncovered.
| Rule | Question it answers | What it checks | Limit against synthetic content |
|---|---|---|---|
| Rule 707 (proposed) | Is the evidence reliable? | Reliability of machine/AI output under the Daubert factors (Rule 702) | Applies only to evidence a party admits is AI-generated |
| Rule 901 | Is the evidence what it claims to be? | General authenticity and origin of the material | Does not separate genuine content from a well-made synthetic one |
| Rule 902(13) | Does the record come from a reliable electronic system? | Self-authentication via a qualified person's certification | Certifies the process, not the genuineness of the captured content |
Reliability and authenticity stay two distinct requirements: Rule 707 addresses the first, Rule 901 and Rule 902 the second. Anyone bringing evidence into federal court has to satisfy both, because evidence that is reliable but not authenticated, or authenticated but not reliable, is still open to challenge. Getting both right is what an end-to-end chain of custody and certification process is built to deliver.
What Rule 707 requires from those presenting digital evidence in federal court
Federal Rule 707 asks a party to show that machine-generated evidence is as reliable as an expert's testimony, by meeting the four requirements of Rule 702. This is where the difference between Rule 707 and Rule 702 turns practical: the same criteria built for the human expert now apply to the machine's output.
To admit machine-generated evidence under Rule 707, a party has to satisfy four requirements drawn from Rule 702(a)-(d):
- Helpfulness: the evidence helps the judge or jury understand a fact or decide a disputed one.
- Sufficient factual basis: the output rests on sufficient facts and data, not on incomplete or unrepresentative material.
- Reliable methods: the content is the product of reliable principles and methods whose operation can be examined.
- Reliable application: those methods were applied reliably to the specific facts of the case.
In practice, this opens new discovery burdens around how the system that produced the evidence works, its training data and its error rates. For anyone offering a screenshot, a video or a video call, it comes down to one thing: the integrity of the material has to be demonstrated, not assumed. To meet the reliability standard Rule 707 sets, evidence integrity has to be guaranteed at the moment of forensic acquisition, an approach TrueScreen takes by generating a verifiable evidentiary log.
Certifying at the source: how to meet Rule 707's requirements
TrueScreen certifies every photo, video, screenshot, video call and document at the source, with a qualified eIDAS timestamp and a seal applied through an integrated third-party QTSP. Digital evidence holds up to the reliability standard only when its integrity is guaranteed at the moment of acquisition, not reconstructed after the fact once genuineness is already contested.
The operational answer to Federal Rule 707's requirements is to certify the evidence at the source. Every piece of content captured with TrueScreen receives a qualified eIDAS timestamp and a digital seal applied by a qualified third-party QTSP integrated via API. One point deserves to be clear here: TrueScreen does not issue certificates and does not act as a certificate authority, it integrates the seal of a qualified provider. The result is a verifiable evidentiary log that documents three things together: that the content was not altered after capture, where it came from, and exactly when it was captured. These are the reliability and authentication requirements that FRE 707, alongside Rule 901 and Rule 902, is set to reinforce in US federal litigation.
What it means to certify content at the source
Certifying at the source means acting at the instant the content is captured, not afterward. The difference matters: TrueScreen certifies that a given photo or video was acquired at that moment and stayed intact, it does not unmask an already existing deepfake. Deepfakes are the context that explains why a new approach is needed, not the object of the work.
With certified acquisition, integrity becomes a property the content carries from birth. That flips the problem Federal Rule 707 poses: instead of proving after the fact that material is reliable, its reliability is documented from the first instant, narrowing the room for challenges at the admissibility stage. It is the same logic that underpins Digital Provenance, the documented history of where a piece of content comes from.
Qualified timestamp and QTSP seal: the evidentiary log
The qualified timestamp and the third-party QTSP seal are the two elements that turn an acquisition into defensible evidence. The qualified eIDAS timestamp fixes a certain date for the acquisition with legal value; the seal applied through a qualified third-party QTSP attests that the content was not changed after that moment. Together they build a documented, verifiable chain of custody.
Consider an in-house counsel who has to produce, in federal litigation, a screenshot of a web page and the recording of a video call. Captured with TrueScreen, each element is born with a qualified timestamp and a third-party QTSP seal, so integrity and a certain date become demonstrable without having to reconstruct provenance after the fact. Organizations and law firms use TrueScreen for exactly this: to acquire digital evidence with a certain date and a documented chain of custody, reducing their exposure to authenticity challenges during e-discovery and litigation.
