Why Brady Violations Still Surface After Trial, Not Before

Here's the uncomfortable truth about Brady violations: you rarely find out about them before trial. The typical pattern is that a Brady violation surfaces after conviction, and the most common remedy is an overturned conviction that comes years too late, according to Cornell's Legal Information Institute. By the time the violation is litigated, your client has already served the sentence.

Consider what happened in Wyandotte County, Kansas. In 2025, the DA's office there overturned an 18-year-old murder conviction after its community integrity unit discovered that the prior administration had failed to disclose exculpatory evidence. Eighteen years passed between the crime and the disclosure failure being corrected, per news reports. That's the cautionary tale: nobody caught the problem in time, and the system only self-corrected after a new administration went looking.

Why does this keep happening? NACDL's analysis points to mundane administrative failure far more often than dramatic misconduct. The usual suspects include:

• Witness debriefings that were never memorialized in writing
• Notes that were lost, misfiled, or never turned over
• Brady review delegated to non-lawyers who don't know what counts
• Summaries substituted for original documents, which Judge Emmett Sullivan called "an opportunity for mischief and mistake"

The 2026 layer of digital evidence makes each of these failure modes worse. A single case now carries bodycam video, 911 audio, jail calls, and scanned reports, and nobody on either side cross-references all of it before trial. When a witness statement in a police report doesn't match what the bodycam actually captured, or a jail call contradicts a debriefing that was never written down, the inconsistency sits in the file until someone with time and a playback speed control finds it. Often that someone works on post-conviction.

Part of the problem is structural. Roughly 95% of cases end in guilty pleas, which is one reason Brady violation data is so scarce in public records, as NACDL has documented. Violations get resolved, or buried, without ever producing a published opinion you can learn from.

The practical takeaway for defense teams is that waiting for disclosure is a losing posture. The materiality standard is retrospective; as Judge Paul Friedman put it in United States v. Safavian, applying "reasonable probability" pretrial "borders on sophistry." The only pretrial question is whether evidence might be favorable, and answering it requires actually reviewing everything the prosecution has, cross-referenced against itself. That's the gap a grounded evidence-review workspace like Evidence Copilot is built to close: automated comparison of reports against bodycam and call audio surfaces candidate inconsistencies while there's still time to file a motion instead of an appeal. For teams building a systematic approach, a documented Brady list research workflow turns post-conviction discovery into pretrial diligence.

What Actually Counts as Brady Material in 2026

Most Brady violations don't come from a prosecutor hiding a smoking gun. They come from a note that was never memorialized, a debriefing that got misfiled, or Brady review delegated to someone who isn't a lawyer. By the time anyone catches it, the defendant has usually already been convicted, and the remedy is an overturned conviction years later, as Cornell's Wex entry on the Brady rule makes clear.

Start with the standard itself. The defendant bears the burden of proving the withheld information was both material and favorable, and materiality means a "reasonable probability" that the outcome would have been different. Two features of that standard matter enormously in practice. First, the duty is breached whether the withholding was intentional or accidental, though knowing suppression can expose the prosecution to sanctions. Second, under Kyles v. Whitley, you review everything collectively, not item by item. A single suppressed interview note might look immaterial in isolation. Stack it against the suppressed bodycam segment and the suppressed jail call, and the picture changes. That's why reviewing evidence in silos is a doctrinal mistake, not just a workflow problem.

Then widen the lens to Giglio. Impeachment material bearing on witness credibility is Brady material, and where information is both Brady and Jencks material, NACDL's guidance is blunt: Brady always trumps the Jencks Act, so disclosure comes on the earlier Brady timeline. In 2026, that matters more because states are reshaping how officer impairment data flows. Minnesota's HF 962, approved in March 2026, creates a uniform statewide process for defining Brady-Giglio impairments and bars "do not call" lists, while city attorneys have warned that new confidentiality provisions may limit defense access to personnel records. The practical upshot: your access to officer credibility information increasingly runs through the formal disclosure process, so a documented Brady list research workflow matters more than informal knowledge.

Two judicial observations should shape how you litigate disclosure. Judge Paul Friedman, in United States v. Safavian, called the retrospective "reasonable probability" standard something that "borders on sophistry" when applied pretrial. Pretrial, the only question is whether the evidence may be favorable to the accused. Don't let a prosecutor hide behind materiality arguments before trial; favorable is the test. Judge Emmett Sullivan, in the Stevens case, offered a second warning: summaries in place of original documents are "an opportunity for mischief and mistake." Insist on original-form production. A summary of a witness interview is the prosecutor's characterization of the evidence, and the impeachment value often lives in the exact words.

The scale problem is real. Roughly 95% of cases end in guilty pleas, which is partly why Brady violation data is scarce in public records. A 2010 Innocence Project study found 10% of 255 DNA exonerations involved alleged prosecutorial suppression. The lesson for 2026 is to treat disclosure as something you investigate and document, not something you wait for. Tools like Evidence Copilot can help by cross-referencing a police report against bodycam audio or a witness statement against jail calls, surfacing candidate inconsistencies with citations, but every output stays marked for human review. The judgment about what counts as Brady material remains yours.

The State Legislative Shift: Minnesota HF 962 and Brady-Giglio Impairments

For decades, the practical answer to a Brady violation was often informal: a defense attorney heard through the grapevine that an officer had credibility problems, or found a name on a "do not call" list. That safety net is being pulled away by state legislatures, and defenders need to adapt before trial, not after a conviction gets overturned.

The most significant development is Minnesota HF 962, approved by a house committee in March 2026. According to the League of Minnesota Cities, the bill creates a uniform statewide process for defining Brady-Giglio impairments and requires prosecuting agencies to adopt written policies on personnel data disclosure. It also prohibits "do not call" lists of impaired officers.

That last provision cuts both ways. On paper, uniform written policies sound like an improvement over the patchwork of informal practices most jurisdictions run on. But stakeholders, including city attorneys, have raised concerns that the new confidentiality provisions for personnel records may conflict with labor laws and restrict defense access to impeachment material. The informal channels defenders once relied on may close without adequate formal replacements.

Why this makes documented review of digital evidence more critical

If you can't count on learning about an officer's credibility problems from a list or a colleague's tip, the evidence the prosecution actually turns over becomes your only reliable source. That means the files in your discovery dump: the police report, the bodycam footage, the 911 audio, the jail calls. Per NACDL guidance, most Brady violations stem from mundane failures like unmemorialized witness debriefings and lost notes, and the materiality standard requires you to assess all undisclosed information collectively, not item by item.

A few numbers frame the stakes:

• Roughly 95% of criminal cases end in guilty pleas, which is why Brady violation data is so scarce in public records: most violations never surface at all.
• 16% of capital conviction reversals between 1973 and 1995 traced to prosecutorial suppression of favorable evidence.
• In 2025/2026, the Wyandotte County DA's Office in Kansas overturned an 18-year-old murder conviction after its community integrity unit found the prior administration had withheld exculpatory evidence.

The Wyandotte case is the cautionary tale. Nobody caught the problem at trial. It took 18 years and a post-conviction review unit to surface what a rigorous pretrial review of the file might have flagged much earlier.

The practical response is to treat every disclosure as something you verify against the underlying media. When a police report says your client consented to the search, does the bodycam audio match? When a witness statement is summarized in a report, does the original 911 call support it? Judge Emmett Sullivan called summaries "an opportunity for mischief and mistake," and that concern applies doubly to digital evidence nobody has actually watched or listened to.

This is where a structured Brady list research workflow earns its keep. Evidence Copilot's workspace cross-references each newly ingested file against earlier evidence in the case, surfacing candidate inconsistencies (report versus bodycam, statement versus call audio) as reviewable signals before anyone asks. Every output carries a citation to the exact source moment, and the original files stay immutable, so what you flag is defensible if you later litigate the disclosure failure.

HF 962 may or may not survive amendment in its current form, and other states will draft their own versions. Either way, the direction is clear: informal knowledge about officer credibility is disappearing, and the defense teams that win Brady fights in 2026 will be the ones who find the material in the evidence itself.

Building a Proactive Brady Review Workflow Before Trial

Standing Brady orders change the posture of disclosure before anyone has to litigate it. A growing number of judges now issue them in every criminal case, and those orders do real work: they clarify that Brady trumps other discovery rules, require original-form production rather than summaries, and make violations subject to contempt. Judge Emmett Sullivan called summaries "an opportunity for mischief and mistake," which is why the order should demand originals, not digests.

Your own demands should probe the prosecution's disclosure process, not just its output. Ask how the office identifies Brady material, who reviews it, and whether that review was delegated to non-lawyers, since unmemorialized witness debriefings and misfiled notes drive most violations. Then build the cross-reference layer the Kyles collective-materiality standard actually requires. You cannot assess withheld evidence item by item, so your review workflow has to hold the police report, the bodycam, the 911 audio, and the jail call side by side.

• Request a standing order requiring original-form production and a written disclosure protocol.
• Demand the identity of every person who conducted the *Brady* review.
• Cross-reference each report narrative against bodycam and dashcam footage of the same event.
• Compare every witness statement against jail call audio and 911 recordings.
• Log each gap as a documented follow-up demand, not a note in your head.

That last step is where manual review breaks down at scale. Evidence Copilot's Brady list research workflow compares every new file against earlier evidence in the case on ingest, surfacing candidate inconsistencies (report versus bodycam, statement versus call audio) as unreviewed signals before you go looking. Each signal carries both sides' quotes with citations into the exact video second or PDF page, and every AI output is marked for human review. The collective review the case law demands finally becomes mechanical instead of aspirational.

The stakes justify the effort. A study cited by the NACDL found that 16% of capital conviction reversals between 1973 and 1995 stemmed from prosecutorial suppression of favorable evidence, and a 2010 Innocence Project review tied 10% of 255 DNA exonerations to the same allegation. In Wyandotte County, Kansas, 18 years passed between the 2008 crime and the 2026 decision to overturn the conviction after a community integrity unit found undisclosed exculpatory evidence. With roughly 95% of cases resolving by plea, that kind of failure rarely surfaces until it is too late. Build the review before trial, and you are not waiting on a conviction to be overturned.

If you want to see how the ingest-time signal review works on your own discovery, book a 20-minute walkthrough and bring a case file.