This Reddit thread raises a technical question with genuine regulatory teeth: does 14 CFR 91.17, which prohibits operating a civil aircraft within eight hours of alcohol consumption or while impaired, apply to training conducted in an Advanced Aviation Training Device (AATD) used solely to log instrument currency under 61.57(c)(2)? The literal text of 91.17 references "civil aircraft," and an AATD is by definition not an aircraft—it's a ground-based simulator. On a narrow textual reading, the poster is correct that 91.17 doesn't directly criminalize flying a desktop or enclosed sim while intoxicated. But that's where the easy analysis ends and where pilots need to be far more careful than the question implies.
The more relevant regulatory hook is 14 CFR 61.53, which addresses medical certificate limitations and prohibits acting as pilot in command or in any other capacity as a required crewmember while the pilot knows or has reason to know of any medical deficiency, or while taking medication or receiving treatment for a medical condition that would make the pilot unable to meet the requirements of their medical certificate. Alcohol intoxication is squarely a medical/physiological deficiency. Logging approaches in an AATD to satisfy instrument currency under 61.57(c) is an act performed specifically to maintain privileges as PIC on future flights—it's not idle simulator time, it's a regulatory function tied directly to legal currency to act as PIC under instrument flight rules. An argument can be made that falsely or fraudulently logging currency obtained while impaired implicates 61.59 (falsification of records), since the logbook entry represents that the training/proficiency event was validly conducted, and a reasonable case exists that impaired performance undermines the legitimacy of that representation, especially if the CFI signing off the session is unaware of the pilot's condition.
Beyond the black-letter regulatory question, this scenario matters practically because instrument currency isn't a paperwork exercise—it exists to ensure a pilot retains the actual procedural proficiency to safely fly an ILS or RNAV approach to minimums in IMC, potentially in low visibility with an aircraft full of passengers. An impaired currency session doesn't just risk violating a technicality; it undermines the entire purpose of the currency requirement, producing a pilot who is "legally current" on paper but who never actually rehearsed the procedures with the attention and cognitive function needed to retain them. If that pilot later flies an actual approach in IMC relying on skills "practiced" while drunk, the safety gap between logged currency and real proficiency becomes a chain-of-events risk factor identical to what NTSB reports repeatedly cite in CFIT and loss-of-control accidents following instrument approaches.
For working pilots—particularly CFIIs who conduct these sessions and sign off training—the liability exposure is significant even if enforcement under 91.17 itself is questionable. A CFI who knowingly allows an impaired pilot to log currency exposes their own certificate to 61.53-adjacent scrutiny, and any accident investigation following an approach flown on impaired-derived "currency" would almost certainly examine the AATD session records, raising falsification and reckless operation questions well beyond the narrow 91.17 gap the original poster identified. The broader lesson echoes a persistent theme in FAA enforcement history: regulatory gaps in specific rule text rarely provide real cover once the FAA or NTSB reconstructs intent and outcome after an incident. Professional pilots and flight schools should treat AATD currency sessions with the same sobriety standards applied to actual flights, both because the surrounding regulatory framework (61.53, 61.59) likely still applies and because the entire value of currency training depends on the pilot being fully capable of learning and retaining what's being practiced.