Taxiway speed management sits at an interesting intersection of regulation, airline policy, and pilot judgment—and the Reddit thread's premise is essentially correct: there is no single published "speed limit" painted on taxiway signage the way there is on a highway. The FAA does not mandate a universal numeric taxi speed in 14 CFR Part 91 or 121; instead, the governing principle is the broad "careless or reckless operation" standard along with each operator's own Flight Operations Manual (FOM) or General Operations Manual (GOM) guidance. Most airlines and fractional/charter operators internally target roughly 15-20 knots on straight taxiway segments, slowing to 8-10 knots or less in turns, and down to walking pace in congested ramp areas, near gates, or in low-visibility conditions. Airport authorities can and do post advisory speed placards in some ramp or non-movement areas (particularly on FBO ramps or cargo aprons), but on the movement area controlled by ATC, taxi speed compliance is fundamentally a crew responsibility, not a tower-enforced metric.
For working pilots, this is a meaningful distinction between "movement area" and "non-movement area" operations. In the movement area, ATC issues taxi clearances and route instructions but does not typically monitor or call out speed—there's no radar-derived speed enforcement analogous to highway patrol, and ground controllers have no practical way to gauge an aircraft's groundspeed from the tower cab except by eye. That means taxi speed discipline is almost entirely a function of company SOPs, aircraft type performance (a widebody's inertia and brake dynamics differ substantially from a regional jet's), and individual pilot technique. Passenger perception of "fast" taxiing—the very observation that prompted this question—is often a mix of actual speed variance, aircraft mass and suspension characteristics, and how aggressively a crew manages thrust and braking through turns rather than any regulatory violation.
Where this becomes operationally significant is in the accident and incident record. NTSB and FAA data show that excessive taxi speed is a contributing factor in a nontrivial number of runway excursions, taxiway excursions, and ground collisions, particularly during turns onto taxiways with tight radii, in wet or icy conditions, or when a pilot misjudges deceleration distance approaching a hold-short line. High-profile ground incidents—wingtip strikes, tug and ground-equipment conflicts, and excursions off paved surfaces—frequently cite "taxi speed excessive for conditions" in post-incident analysis, which is why FOMs increasingly specify numeric target speeds by segment type (straightaway vs. turn vs. ramp) rather than leaving it to pilot discretion alone. Flight data monitoring (FOQA/FDM) programs at major and regional airlines now routinely flag taxi speed exceedances as part of routine safety-data review, turning what used to be an informal "bad habit" into a trackable, coachable metric with real safety-management-system (SMS) implications.
This ties into a broader industry trend toward quantifying and standardizing behaviors that were traditionally left to airmanship and habit—taxi speed, stabilized approach criteria, flap/config callouts, and sterile cockpit discipline are all areas where operators have moved from "good practice" to codified, monitored standards as SMS and data-driven safety programs mature. For business aviation and Part 135 operators without the fleet-wide FDM infrastructure of major carriers, this remains more reliant on training culture, chief pilot oversight, and simulator/checkride emphasis, but the trajectory across all segments—airline, corporate, and GA—is toward more explicit, measurable taxi procedures rather than leaving speed control purely to "feel." For pilots, the practical takeaway is that even though ATC rarely says a word about taxi speed, it is very much being watched—just after the fact, through data, rather than in real time by a controller.