John Rizzo, formerly acting General Counsel to the CIA (and now a visiting fellow of the Hoover Institution, where he is working on a memoir) has a short piece on the practicalities of the reporting regime by the CIA to Congress. It's worth reading this short article (published in Hoover's house journal for disseminating work by its fellows, Defining Ideas) as well as this short (25 thousand word) book by Amy B. Zegart, Eyes on Spies, a political scientist's account of Congressional oversight of the intelligence agencies, which I reviewed for Lawfare here.
Rizzo, who retired from the Agency two years ago, notes that in November 2010, the House Permanent Select Committee on Intelligence announced, following a review of intelligence community Congressional notification practices, that "it was making a series of recommendations to improve the system. The Director of National Intelligence subsequently issued new guidelines in the area. So perhaps the communication lines between the CIA and the Congress are markedly better now." But speaking to past several decades, spanning multiple administrations, Rizzo observes:
[T]he conundrum has always been in trying to discern what information the CIA should report to the Hill, and when exactly it should be reported. During my many years at the Agency, we never seemed to get it quite right. Part of the problem can be traced to the statutory benchmarks Congress established for reporting matters to the intelligence committees, which consist of vaguely-couched terms like keeping the committees “fully and currently informed on all intelligence activities,” including “significant anticipated intelligence activities.” For new, presidentially-authorized covert action programs (“Findings”), it was always clear and understood that these had to be reported on a timely basis (in practice, virtually always within 48 hours of the presidential authorization). Beyond that, however, the CIA has over the years gotten little more in the way of concrete guidance from Congress .... in applying Congress’s cryptic standard of “fully and currently informed,” should we emphasize the “currently” or the “fully”? Put another way, do we risk reporting too little about something (or worse, something that may later prove to be inaccurate), or reporting it too late?
As Rizzo notes, to be sure, reporting requirements and internal processes might well be different, and better, today.