aviation

Day 17 of having no clue what happened to Air India 171.

Take your time, India, it’s not like we need to know for safety reasons or anything

Day 26 after the Air India 171 crash and we are starting to receive some indications that the fuel switches may have been moved to the CUTOFF position at some point during the takeoff or initial climb sequence.

That info is still subject to confirmation in the coming weeks, but it would be a major head-scratcher, involving either a gross error or murderous intent

Day 26 after the Air India 171 crash and we are starting to receive some indications that the fuel switches may have been moved to the CUTOFF position at some point during the takeoff or initial climb sequence.

That info is still subject to confirmation in the coming weeks, but it would be a major head-scratcher, involving either a gross error or murderous intent

https://theaircurrent.com/aviation-safety/ai171-investigation-fuel-control-switches/

Aviator, diver, Etherean, nexialist, PhD in aviation and climate science, host of /diving /ethfinance /forteantimes /geopolitics /science /singapore /thomas

If that was confirmed as the reason for the crash, would that be good or bad in your opinion?

EVM. ETHFinance. Home Staker. Believer in decentralisation.

All good points. I’m ambivalent about AI — not about flight automation itself (most flights are already flown, and some even landed, on autopilot), but about preventing malice from someone in the cockpit who still needs ultimate control.

You either leave enough authority to the crew to override automations, or you take the human entirely out of the equation (which, by the way, is only being vaguely considered as a distant end-game beyond single-pilot operations).

It’s the “mixed mode” that’s thorny, and has been thorny since at least Airbus introduced fly-by-wire in the early 80s (see Saint-Odile) and as recently as Boeing’s 737 MAX debacle (automation overriding pilots leading to two accidents).

In this specific Air India 171 instance, *maybe* an automation could have determined that shutting off the fuel to either or both engines in the absence of a detected engine fire was malicious and could have overridden the pilot’s action.

But for the one-in-a-thousand case where the pilot was acting maliciously, there are a thousand-minus-one cases where pilots will shut off a fuel valve with good reason, and now we need to make sure none of them get denied, or we’ve just compounded the original problem.

Not unsolvable, but certainly tricky and will require a lot of soul-searching by the aviation community and clever work by the design authorities / OEMs

I don’t know why, but if human intervention was the reason for the crash it would be „positive“ for me cause it would restore trust in the tech „plane“. And maybe with AI down the road there is something we can add to decrease the amount of successful suicides?

Terrible from a human perspective, of course — 9.5B people entrusted their lives to pilots last year, and as much as we can reduce mechanical or environmental hazards, there’s very little we can do against pilot murder-suicides, just prevention and psychological care