This month the Supreme Court will hear oral arguments in two consolidated cases, Turner and Overton, that deal with prosecutorial misconduct and the State’s Brady duty to disclose exculpatory evidence very often. Here is an excellent analysis from Bidish Sarma in the American Constitutional Society Blog:

“Last year, I wrote for the ACS here about one of the chief concerns underpinning the case: that prosecutors decide whether to disclose exculpatory information pre-trial based on their prospective assessment of whether withholding it would make a “material” difference at trial. To recap the Brady test courts use post-trial: a new trial must be granted where the defendant has proven: (1) suppression—that the State actually failed to turn over the information at issue; (2) favorability—that the information would have helped the defendant; and (3) materiality (also known as prejudice)—that, had it been disclosed before the trial, there was “any reasonable likelihood” it could have “affected the judgment of the jury.” (Wearry v. Cain (2016) (internal citations omitted)). Tying prosecutors’ disclosure determinations to their own predictions about materiality is problematic for a bevy of reasons … But, what is truly worrisome is that the Supreme Court justices have not acknowledged or fully grasped how their Brady rulings influence prosecutorial disclosure decisions.
In the oral argument in Smith v. Cain in 2011, several justices expressed dismay when the prosecutor arguing the case informed them that the lawyers in her office making disclosure decisions did not simply consider whether information is exculpatory, but also gauged whether it was material. Some vocalized their concerns with this approach. Justice Kennedy posited a distinction between a Brady “obligation” to turn over evidence before trial and a Brady “violation” based on an appellate court’s post-trial assessment of materiality:
And with all respect, I think you misspoke when you – you were asked what is – what is the test for when Brady material must be turned over. And you said whether or not there’s a reasonable probability . . . that the result would have been different. That’s the test for when there has been a Brady violation. You don’t determine your Brady obligation by the test for the Brady violation. You’re transposing two very different things. And so, that’s incorrect.
If there is a distinction—and there surely should be one—the Court’s jurisprudence provides no guidance on how to identify it.
Indeed, consider what the Solicitor General has argued in Turner and Overton. Although there is no dispute that the prosecution failed to turn over several categories of exculpatory evidence, the Government’s brief on the merits states, “[t]he government complied with is obligations under Brady v. Maryland, 373 U.S. 83 (1963). Nondisclosures violate Brady only when withheld information is both favorable and material.” See how easily the Government elides that distinction between a Brady obligation and a Brady violation? In the first two sentences in the summary of its argument, the Government exploited a jurisprudential problem that the justices should address.