Why Brady Violations Keep Happening in 2026

More than sixty years after Brady v. Maryland, disclosure failures remain a structural feature of American criminal practice rather than a solved problem. The reasons are not mysterious, and they are not fixed by good intentions.

The most detailed account comes from NACDL research built on interviews with 35 state prosecutors across seven offices. That study identifies three causes: negligence driven by caseloads and underfunding, cognitive bias, and intentional misconduct. Urban prosecutors in the sample handled roughly 60 to 80 felony cases per year, and the study found few offices conduct any retrospective review or audit of disclosure decisions.

Cognitive bias is the hard one. NACDL describes it as largely unconscious and not overcome by will, good intentions, or training. Prosecutors naturally read reports for confirming evidence rather than for what disconfirms the charge. Ben Gershman, a former prosecutor turned academic, offers the blunt version of why the practice persists: "because it works." Convictions follow, sanctions rarely do.

That is why trust-based review keeps failing on the defense side. Public Brady lists help, but they lag and narrow the field. Milwaukee County's list held 217 entries involving 190 officers as of September 2025, per Wisconsin Watch reporting, yet it excludes adverse judicial credibility rulings and can trail internal investigations by years. Defense attorney Caitlin Firer describes the practical response: when bodycam footage "strikes me as something's not right," she runs the officer's name herself.

The issue is live in 2026. A Cleveland Guardians pitcher asked a New York federal judge in September to sanction prosecutors over an alleged failure to promptly disclose exculpatory evidence, according to Law360's report on the sanctions motion.

If bias is unconscious, verification has to be systematic:

• Cross-reference every testifying officer against the jurisdiction's Brady list, including the ones who don't look problematic.
• Compare the police report against the bodycam and the 911 call, line by line, for omissions and contradictions.
• Build a Miranda and custodial-interrogation timeline from the recordings themselves, not from the report's summary of them.
• Log every inconsistency as it surfaces, with the exact source moment attached, so it survives the next continuance.

That last point is where a Brady list research workflow earns its keep: grounded citations into the exact second of video or page of the PDF, every AI output marked for human review. Evidence Copilot is built for exactly that verification layer, and you can see it on a 20-minute walkthrough before deciding anything.

Why You Can't Rely on the Prosecutor's Brady List Alone

Milwaukee County's public Brady list sounds like exactly the resource you'd want: a searchable roster of officers with documented integrity problems. As of September 2025, it held 217 entries involving 190 individual officers. But if you treat it as a complete picture of officer credibility, you're missing things the list structurally cannot capture.

Start with what the list excludes. Milwaukee County only adds officers with a pending criminal charge, a past conviction, or an internal investigation that brings the officer's integrity into question. It does not track adverse judicial credibility rulings, which means an officer a judge has effectively discredited in open court can still be off the list. Milwaukee DA Kent Lovern defends that line, arguing that "credibility determinations, which are frequently made by courts, don't constitute judgments of untruthfulness." University of St. Thomas law professor Rachel Moran sees it differently: "An officer who has misstated information in his police report, that's exculpatory regardless of whether the officer intended to do it." You can weigh that dispute yourself, but either way, a court's finding won't show up in your search.

Then there's the timing problem. Internal investigations take years, and the list only updates when they conclude. Officer Eian West made the list in 2025, two years after the April 2023 domestic violence calls that triggered his investigation. For two years, prosecutors had no list entry to disclose because the underlying inquiry hadn't finished.

The list also gets things wrong. The first public version, released in September 2024, contained what reporters called "inaccurate, incomplete and outdated information": an officer wrongly described as involved in a custody death, another listed for an expunged case, and several deceased officers. A corrected list followed in October 2025.

So what does a working defense practice look like? Milwaukee defense attorney Caitlin Firer runs the officer's name against the Brady list whenever bodycam footage strikes her as off. State Public Defender Angel Johnson puts the broader frustration plainly: "It should be happening from the DA's office, but we are still finding ourselves doing that legwork and it's not our obligation or ethical duty to do so."

That legwork means pulling records the list doesn't hold:

• Adverse credibility findings from suppression hearings and prior trials
• Civil judgments and settlements naming the officer
• Fire and Police Commission decisions, including discipline that got overturned (Officer Burtch's removal from the list after such an overturn shows how fluid these entries are)
• Media reports and internal-affairs outcomes still pending

The practical takeaway: treat the public list as one input, never the answer. A systematic Brady list research workflow pairs the list with your own review of the discovery in front of you, so a Brady violation claim rests on what you can cite, not on what someone else chose to publish.

What Counts as a Brady Violation and What It Takes to Get Dismissal

A Brady violation is more than a missed document. Under Brady v. Maryland and United States v. Bagley, the duty covers evidence that is favorable to the accused and material to guilt or punishment, and evidence counts as favorable "either because it is exculpatory or impeaching," according to a defense-side analysis of the dismissal remedy. Two elements make the violation: the government willfully or inadvertently failed to produce the evidence, and that suppression prejudiced the defendant.

That's the standard for the violation itself. Dismissal is a different, much higher bar. A judge may dismiss only when the failure is willful or flagrant and substantially prejudices the defendant. Simply showing a Brady violation is never enough.

The Ninth Circuit's 2020 decision in United States v. Bundy, 968 F.3d 1019, shows how rare that is. Mid-trial disclosure failures in that case led to dismissal with prejudice, affirmed on appeal, in a prosecution stemming from a 24-year failure to pay federal grazing fees. Federal courts weigh three supervisory-power factors before dismissing an indictment:

• Remedying the violation of a recognized statutory or constitutional right
• Preserving judicial integrity so a conviction rests on validly presented considerations
• Deterring future illegal conduct

The conduct must also be "so grossly shocking and outrageous as to violate the universal sense of justice" under United States v. Kearns (9th Cir. 1983). As Amado v. Gonzalez put it, the prosecution's interest "is not that it shall win a case, but that justice shall be done."

The practical lesson: don't wait for disclosure you may never get. Milwaukee County's public Brady list held 217 entries involving 190 individual officers as of September 2025, yet it excludes adverse judicial credibility rulings and can lag internal investigations by years, per investigative reporting on the county's list. Defense attorney Caitlin Firer describes running an officer's name the moment bodycam footage "strikes me as something's not right." That kind of independent verification, built into a repeatable Brady list research workflow, is how you catch the issue before trial instead of mid-trial.

Evidence Copilot's grounded evidence-review workspace keeps every finding tied to the exact bodycam second or PDF page it came from, so your Brady and Giglio review rests on citations a court can check, not memory. Every AI output is marked for human review; the legal call stays yours.

A Defense-Side Brady Review Workflow You Can Actually Run

Start with the moment that makes most defenders reach for a second screen: you are watching bodycam footage and something does not add up. Caitlin Firer, a Milwaukee defense attorney, describes exactly that instinct, saying she will run an officer's name on the Brady list when footage "strikes me as something's not right." That reflex is the seed of a repeatable workflow.

Here is one you can run on every case, in order.

• Ingest the discovery dump first. Load bodycam, dashcam, jail calls, 911 audio, interrogation footage, and police report PDFs into a single case file so reports and recordings sit side by side.
• Compare the paper against the audio and video. Read the report, then read the transcript next to it. Flag every place the narrative and the recording diverge, and note the timestamp for each.
• Cross-reference testifying officers against public sources. Run each name through your jurisdiction's Brady/Giglio list, then check media reporting, commission decisions, and civil judgments, because lists lag and exclude credibility rulings.
• Document everything with citations. Tie each inconsistency to the exact second of video or page of the report, and keep a running memo you can attach to a motion or hand to co-counsel.
• Re-run the check when new discovery arrives. Officer records change mid-case, so a name that was clean at intake may not be clean at the preliminary hearing.

Why build the workflow at all? Because the lists alone will not carry you. Milwaukee County's public Brady list held 217 entries involving 190 individual officers as of September 2025, yet it does not track adverse judicial credibility rulings, and internal investigations can take years to surface. Officer Eian West was added in 2025, two years after the April 2023 calls that triggered his investigation.

That gap is why Angel Johnson, of the State Public Defenders Office, says the legwork "should be happening from the DA's office" but defenders keep doing it themselves. A structured Brady list research workflow turns that burden into a checklist instead of a scramble.

Evidence Copilot was built around this loop. Auto-signals on ingest compare each new file against earlier evidence in the case and surface candidate report-versus-bodycam inconsistencies before anyone asks. Timestamped diarized transcripts and named-party records keep officer mentions tied to the exact moment in the recording, so your citation is already written when you need it. Every AI output is marked for human review, and originals stay immutable.

Turning What You Find Into a Brady Motion or Impeachment

You've found the discrepancy: the report says consent was given at 9:14, the bodycam shows the question was asked at 9:22. Now what? A finding sitting in your notes does nothing for your client until it becomes a motion, a cross-examination, or a preserved record for later relief.

Start with the motion itself. A Brady violation requires showing the government failed to produce evidence that was "both favorable to the accused and material to guilt or to punishment," and favorable reaches beyond exculpatory facts to impeaching material as well. The elements are straightforward: the government willfully or inadvertently failed to produce the evidence, and the suppression prejudiced the defendant. What trips up most motions is vagueness, so anchor every paragraph to a specific document, timestamp, or page. If your workflow already runs a Brady list research workflow, pair those officer findings with your in-case inconsistencies so the motion tells one connected story.

Cross-examination is where the same evidence earns its keep twice. Build questions directly from quoted sources, and make the tension do the work:

• Read the officer the exact line from the report, then play the bodycam segment from the exact second. Ask which is accurate.
• Tie each question to a quoted source on both sides, so the impeachment rests on the record rather than your characterization.
• Stop at the contradiction. No credibility conclusions from you; the jury draws its own.

Preservation matters just as much, especially when disclosure arrives late. Dismissal is an extreme remedy: you must show a willful or flagrant violation that caused substantial prejudice, not merely that a Brady violation occurred. But it happens. In United States v. Bundy, the Ninth Circuit in 2020 affirmed dismissal with prejudice after mid-trial disclosure. Sanctions motions are also live in 2026: a Cleveland Guardians pitcher indicted in an alleged bet-rigging scheme asked a New York federal judge to sanction prosecutors over alleged failure to promptly disclose exculpatory evidence, per September 2026 reporting in Law360. If you want that remedy available later, document the disclosure timeline now: when you requested, what you received, when each late file landed, and what prejudice followed.

This is where tooling earns its keep without replacing your judgment. In Evidence Copilot, every inconsistency signal and cross-exam draft carries citations into the exact second of video or page of the PDF, so the motion