A mysterious voice is haunting American Airlines’ in-flight announcements
waxy.org
waxy.org
It does not mention the "medical intercom jacks" in overhead bins that are mentioned at the end of the article
[1]:https://caisatech.net/uploads/XXI_3_AIRBUS_H51_A320_O_CCOM_R...
I wonder if someone replaced the standard announcement with a 5 hour long file with intermittent burps and groans....
video capture buffer0 40b37460
video capture buffer1 40a8dd60
wideo_capture buffer2 409e4610
geode_: gamma black+csccorrection
VAL2: Registered "NSC Geode V4LZ UBI driver" as char device B1, 224
VAL2: Registered "NSC Geode V4L2 Capture driver" as char device 81,
geode_v: Unknown standard
VAL2: Registered "NSC Geode V4LZ Overlay driver" as char device B1, 16
Warning: loading /lib/modules/realmagichul_kernelland.o will taint the kernel: non-GPL license - Proprietary. Copyright (c) 2002 Sigma Designs Inc. All rights reserved (version 103 with proc pci dolby css macrovision sm2288 osdric_memorycheck),
/lib/modules/realmagichul_kernelland.o:init_module:Invalidargument
Hint: insmod errors Can be caused by incorrect module parameters, including invalid IO or IRQ parameters
PCI: Setting latency timer of device 00:12.3 to 32
geodeoss: disabling PM: PM registration failed
geodeoss: version 1.2.0 time 12:37:36 Dec B 2004
geodeoss: use_pm=0 check_interrupt=0
Warning: loading /lib/modules/maspcu.o will taint the kernel: no license
usb.c: registered new driver pcu
MAS-eFX SEB USB PCU driver version 1.01
usb.c: registered new driver usbccr
usb_ccr open is being called (0xc165f000
) <6›ccrdev.c: Adding credit card reader: input1
input1: Panasonic Avionics Corp PAC USB Keyboard/Credit Card Reader Ver. 1.01 on usb1:2.1
usbccr.c: v0.59:USB HID Credit Card Reader driver
mice: PS/2 mouse device common for all mice
usb.c: registered new driver pacpcu
geode rev-7 pmr-03a834c4 mcr-00030250
using /dev/mtd1 /usr/app
--- mount succeeded ---
using /dev/mtd4 /usr/app2
--- mount succeeded ---I used to work adjacent to these exact systems...they're so bad.
Like, 10+ minute boot at times, or assigning multiple devices to the same IP, or the communication network just randomly dying... (and this was all in vitro testing)
It was still cool to see "the rack", aka an entire planes infotainment system setup on a rolling cart including however many seat backs each plane was supposed to have. They had multiple office buildings that were essentially full of these, because each airline and deployment was different enough that they didn't want to have to reconfigure when needing to test them.
Also elsewhere in the boot, you might notice they tend to be 80486 CPU emulation boards from Cyrix. They're not fast at all, but for what they do, they don't need to be.
Also, if you took video instead of a pic, there are a few interesting IP addresses might scroll by as the media program gets initialized.
https://twitter.com/actuallyemerson/status/15731071172546396...
I've seen it often, and somewhere on an old PeeCee buried in the darkest areas of my home office, I have a video of the whole sequence.
No announcements were made by the pilot or flight attendants afterwards, and everyone was left confused. I turned to the passenger next to me and we both shrugged.
>Then the master power button up at 1L for the cabin power kept getting I pushed so it would shut off all the IFE, seat power and outlets. I had to push the button on again about 3 times inflight. Then the 2 got a call with the same groaning noise and thought it was one of us playing a joke but it wasn't us.
>It certainly was freaking all of us and the passengers. They wanted off the plane as bad as we did. The 2 be FAs said they had been on the plane the other day and the same thing happened. I had a visual on each FA when the noise would come over the PA and one time when the FO was doing a break so unless the captain was playing a long sick, joke I don't have an explanation.
https://twitter.com/xJonNYC/status/1572358118985633795
Any ideas?
I can't even get passed the first sentence of this. What does any of this mean?
IFE is in flight entertainment.
This makes sense, because for safety reasons all the entertainment is paused when they use the public address system.
It's more common for the entertainment to be muted without being paused. I assume that's as a convenience to the passengers.
I had totally forgotten about this because I haven't flown long-haul in decades. I suppose these days I would just watch video on my tablet anyway.
Nobody said a feature implemented as a convenience to the user had to make things more convenient for the user. They're barely even related concepts.
Only time will tell, but I'm almost certain I'm right on this one.
The FAA does not agree with that. The FAA probably thinks that easter eggs in avionics software are a crime. And I agree with that position! Aviation, medical, military, and any other critical software should not contain easter eggs, and it should be an offense to knowingly include one. Stuff like this (even though it's probably not an easter egg) is a great example why. Just imagine the cost to patch this if it's widespread... and it has to be patched, because this one can literally kill people if you don't fix it (think spooked flight crews making rushed or generally poor decisions if you don't think this is potentially lethal).
According to the researcher who found it, "they didn’t think it was very funny.”
I love easter eggs myself—classic Apple software in particular had a lot of these, and I really miss finding them—but they're admittedly a risk. I remember one obscure Easter egg in Mac OS 7.6 that occasionally caused crashes at boot time after displaying a bizarre message about how "bluets and granola bars / make a chewy snack". Cute, but probably not something Apple wants to be getting customer support requests about.
As far as the topic at hand, I don't believe these weird noises are an Easter egg, either. I'd guess it's a recording patched in somehow by someone who works for the airline.
And that's what it boils down to, isn't it?
Personally I draw the line somewhere around "software that's controlling something". If software can directly screw up or cause damage in the Real World™, that's a good reason it shouldn't have easter eggs in it. If the trouble has to indirect through one or more meatbags who know better before becoming reified, well, the harm caused by any easter eggs is probably much less than the harm caused by reported bugs already closed WONTFIX with a laugh. (Note: PAX do not in fact know better, ever, as any airline employee can and will tell you.)
So boring! I look forward to my pacemaker giving me palpitations to the rhythm of "Never Gonna Give You Up".
Turns out his pacemaker was shocking him because it thought his elevated heart rate was a danger! They fixed the setting.
The B-52H bomb/nav computers had a really rudimentary golf game. Think Atari 2600 quality.
The B-2A had a breakout-like game at one time.
Both were accessed via some really esoteric key presses.
https://aviation.stackexchange.com/questions/12776/how-are-t...
Jacking into that would not be inconspicuous.
An ESP32-PICO-D4 is 7mm wide, but based on internal images it appears you can safely grind it down to ~6.35mm. It contains a DAC, 4MB flash, and all needed peripherals plus bluetooth. The quality [1] would be poor, but this matches what we observed.
A circuit board would probably be out of the question given the size. Even 0.8mm would have very little usable area. Flex PCB could work well, but is probably overkill for the complexity required. I'd expect a project like this to be hand soldered, given the low (less than 6?) component count. It wouldn't be easy, but well within the capabilities of a hobbyist at home with a cheap USB microscope.
For power, I could imagine a repurposed AirPod battery (they fit in the 0.235 inch diameter "stem" of the headphone) with a deadbugged switching or linear regulator fitting both the power budget and the size constraint.
From a physical design perspective, a lot of traditional methods seem impractical. With this size constraint one possibility could be placing requisite components into a 1/4" jack shaped mold, and casting it in epoxy in a vacuum, completing the conductive surfaces with copper tape.
1. https://www.youtube.com/watch?v=lgDu88Y411o
I could definitely make one as a demo.
But overall I agree that it sounds very human. For example, the grunt at 0:13 doesn't sound like random noise at all, it could easily be a sound effect in a movie or game.
Suppose the two possibilities are ‘prank made from human’s voice’ and ‘weird electromechanical phenomenon’. The logic I believe you’ve implicitly applied is that most electromechanical phenomena do not sound human and therefore this human-like sound is likely human. However I put it to you that this logic is flawed because you have not heard a randomly sampled electromechanical phenomenon but rather a sound that was sufficiently human-like to warrant an article about hauntings so thinking about the ordinary distribution of electromechanical phenomena is misleading.
Furthermore I think that you’re assigning relativity equal weight to the two possibilities of a human prankster and any kind of PA system fault when I would argue that the latter is much more common. That is, I think you’re answering a question like ‘we flipped a coin and if it came up heads, we made silly noises into a dictaphone and if it came up tails we tortured the device in some way while it was recording. We’ll play the sound and you can guess which it was’ whereas I think the answer in this case hinges a lot more on the equivalent of the coin-flipping process (eg what if you needed 10 heads in a row for the first choice? What about 15? 20?)
One can formalise this too. I’ve tried to come up with some priors that illustrate the point and which I think are reasonably generous towards your conclusion.
Say H = some human pranking a flight with silly noises,
M = some mechanical failure in the PA system,
Mh = human-sounding mechanical failure
Mm = mechanical-sounding mechanical failure
S = human-like sounds
And let’s invent some numbers for a random flight:
P(H) = 1e-6 (feels too big to me)
P(M) = 1e-3 (feels a bit small to me)
P(Mm | M) = 0.995 (I.e. 99.5% of mechanical sounds sound mechanical)
P(Mh | M) = 0.005
We read an article about S happening so we now wonder:
P(H | S). Apply Bayes theorem:
= P(H) P(S | H) / P(S)
= P(H) * 1 / P(S), and in our model
= 1e-6 / P(S).
Now what is P(S)? We can calculate in our model (assuming we’re describing all possible causes and for simplicity ignoring a flight with a human-sounding mechanical failure and a prankster).
P(S) = P(H) + P(Mh)
= P(H) + P(Mh | M) * P(M)
= 1e-6 + 0.005 * 1e-3
= 6e-6
And we can plug that back into the equation above:
P( H | S ) = 1/6
You're ignoring the fact that it's not just one production. I would consider your position if there was one continuous production with small to no variations.
Why are you ignoring the following?
- the sounds have the length of a yell or a moan.
- the silence between them is not fixed or related to any other sound that we can hear on the plane.
- there are no artifacts, except the weird pitch going up artificially in one of them, which might point to a prerecorded and manipulated sound
- if this was a electromechanical issue, it would have to mimic a significant part of the human anatomy, including a tongue (there are a few productions that sound like the throat is constricted and the tongue is rolling slightly).
Sure, a random electromechanical issue could have these characteristics, but isn't that extremely unlikely? You would expect other noises, not just the vocal-like sound.
The only problem for them is that AA could cross reference the travelers on those flights to figure out who it was, if it was a passenger using these ports to prank.
>The ACPs give a long-range visual indication about the system status (e.g. PAX call active). They are installed at the ceiling in the middle of the aisle between the passenger/crew doors, close to the attendant stations.
>The lighted segments are activated either continuously (steady) or flashing. In case of "normal" calls they are not flashing (steady), in case of "abnormal" or "emergency" situations they will flash. Only the blue PAX call field will never flash, this will be always a "normal" indication
These indicator lights would be helpful in narrowing down if the emergency system was being engaged or not. Could potentially point to exploitation/hack of the PA system rather than rogue utilization of normal functions.
If the device had a long time delay or a delay based on the number of flights they might have to check many flights before the affected flight. And there is no guarantee the person would use the same delay each time.
If they pick flights on routes that have a lot of repeat travelers and the delays on the device make it so the airline has to look at a fews days worth of flights to get the suspects for each incidents then I'd not be surprised if there are may people that were on all the planes sometime before the affected flights.
The big problem I see with it being done by someone not on the flight is that surely after the first couple of times the PA system is going to get thoroughly inspected and the device would be found. The mystery would then just be "who", rather than the "how" mystery that we currently (at least publicly) have.
If it’s a hack of any sort someone is taking their life in their hands because being caught tampering with any commercial airline systems could mean significantly jail time. Cleaning crews and other staff have enough time on the plane to do something… I’m thinking like an Adafruit Waveshield with a timer or something.
If it’s a ghost, it seems like a horny adolescent one.
not sure. A cursory reading of https://www.law.cornell.edu/uscode/text/18/32 doesn't mesh with a silly prank. Which law do you think they are breaking?
Sure, passengers weren't intended to use that port, but from a system perspective the system is being used as intended with no interference or exploitation.
I'm not sure this system can even be "hacked" short of disassembling parts of the plane and tapping into the communications wiring.
https://arstechnica.com/tech-policy/2021/10/missouri-gov-cal...
Yes, this is a state government and not federal. I’m sure you can find equally outlandish examples of federal cases.
Microcontrollers are small enough now that there could be lots of interesting ways to interface.
EDIT: Also, based on the OP, the PA gets power cycled repeatedly on at least one flight where this is happening. That’s much more intrusive.
Or again… horny ghost
Maybe it would make a difference if the prankster was an employee, and not just some random person, but I doubt it.
Yes.
We're well passed that these days, for better or worse.
Maybe I'm off-base and they're just risking an airline ban, but it seems (to my uninformed opinion) that doing that on an flying airplane is playing with fire that the airline won't sue and that, if they do, it'll be a clement judge which won't make an example out of them for messing with an airplane's computer, safety critical or not.
I would be happy to agree that the CFAA is overbroad. But disruptive actions toward flights and aircraft are quite reasonably punishable under other theories, when they could plausibly create a risk for aviation, regardless of the means.
So yes this is legally a hack in the way it’s used by the media, it’s also a hack of in the normal since of the word regardless.
Law enforcement use phrases like "unlawful access to a protected system" and whatnot.
The "part" is the PA system. While the device is playing the prank sound, it is temporarily "made unworkable" for its intended purpose of allowing the crew to communicate with the passangers. This "endangers the safety of the aircraft" because in certain situations it's possible the crew would need to communicate quickly with passangers or have silence so that they can understand each other and focus on solving an issue.
Edit: Looks like captain could override, flight attendants couldn't. That still causes a safety risk because in an emergency the cockpit might want to focus on flying the plane while the flight attendants instructed the passangers. Plus the distraction factor remains.
I’ve got to believe it’s safety critical.
49 U.S. Code § 46504 - Interference with flight crew members and attendants
An individual on an aircraft in the special aircraft jurisdiction of the United States who, by assaulting or intimidating a flight crew member or flight attendant of the aircraft, interferes with the performance of the duties of the member or attendant or lessens the ability of the member or attendant to perform those duties, or attempts or conspires to do such an act, shall be fined under title 18, imprisoned for not more than 20 years, or both. However, if a dangerous weapon is used in assaulting or intimidating the member or attendant, the individual shall be imprisoned for any term of years or for life.
You might argue that flight crew could be intimidated by the knowledge that their aircraft is behaving unusually, but I don't think that's particular core territory of "assaulting or intimidating".
- law that says you aren’t allowed to vandalise other peoples stuff (for the verdict); and
- the thing being vandalised is the thing used to coordinate safe evacuation of passengers on a vessel (for a hefty sentence).
Like scribbling on a painting vs a bus stop — let the justice be done in the sentencing.
https://i.stack.imgur.com/5w241.jpg
* Unless someone had made dongle that once inserted looked like the same plugs, but raised enough to hide the dongle and match the color/texture of the surrounding plastic. The fake plugs on the outside would not even need to be functional and the device could be set to start on a delay, so whoever is placing them is not physically on the flight when it starts. It could even have an accelerometer to turn on after take-off and turn off on landing. If the person intentionally booked a seat at end of the plane, very likely they could easily add it when everyone is trying to leave the plane.
Credit card skimmers do something similar all the time:
https://www.google.com/search?q=how+to+spot+credit+card+skim...
____________
This guess is based on information provided by two prior HN comments:
[1] PDF of flight systems:
- https://news.ycombinator.com/item?id=32958432
[2] Photo of medical audio jacks:
Wise, although wiser to make that assumption for most sources. Until this starts being reported independently by more reliable sources then I assume it’s fake.
I feel like that would only target a very small percentage of potential customers.
https://mediashower.com/blog/how-clowns-did-the-heavy-liftin...
I mean, the only thing I could see them doing is asking the flight attendants or AA, but both of those are likely hard to get an actual answers for (AA would most likely say 'we are investigating' and would never follow-up). If something is on social media and there are enough corroborating accounts, it's probably enough to write a news article about.
>American Airlines inspected the Boeing 737-800 as well as the P.A. system itself after these reports. They tell me that maintenance “determined the sounds were caused by an issue with the PA amplifier. There was no external access to the system.”
https://viewfromthewing.com/no-theres-no-ghost-in-the-machin...
Hard to agree. Those sounds are almost certainly a human voice, presumably digitized. Unless the issue with the PA amp is that someone connected a sample player to it.
Former pro audio engineer here.
Longtime guitarist, producer, film fanatic here and this is really going to be interesting.
It does remind me of last Halloween where I had my jumpsuit and hockey mask to be Jason Voorhies and I got a small mp3 player and looped an hour of “ki-ki-ma-ma-ma” lifted from YouTube and played thru a little amp on my porch. Play scarecrow and only move to scare the big kids or their parents haha.
As an audio person, musician, producer blah blah, that is totally human. Likely recorded and filtered a few times. By the inflections I think Eastern European. The sound of it reminds me of a Slavic based vocal upbringing. Doesn’t seem like North Korean, Latin, or North American to me.
Probably about a couple minutes of audio looping somehow at the source. Far too many for it to be a hardware exploit because honestly who wants to invest that much in this kind of prank? This isn’t Zug Super Bowl level[0].
My guess is they’re coming through the least secure provider for a third party service - probably the WiFi enterprise thru being on the inside of them - and AA is their first playground where people are noticing. How they get from there to the intercom is probably buried deep in a couple audio wires plugged into overlapping sources unintentionally or there is some kind of frequency band that isn’t shielded properly so it’s possible to push the audio signal over something else and cause garbled noise and almost impossible to find.
Like, in the past it was funny when picking up FM radio signals through my guitar amp, but it sucked when trying to have a serious jam session and talk between playing. Audio signals can be really powerful compared to other delicate systems IMO.
Definitely curious and want to see how this unfolds!
I started noticing some odd sounds in the Student Union building. One day I had some time and decided to investigate. I realized it was coming from a fire alarm. I stood under the alarm and listened for a while.
The alarm was quietly whistling "Harder, Better, Faster, Stronger" by Daft Punk.
But, I'm going off of 8 year old memories, so the only thing I can say for certain is that it happened.
Maintenance and cleaning crews love to listen to radio/music on speakers while working. Someone from this staff plugged in a receiver to listen while servicing the plane, and forgot it. Or they left it there for convenience and didn't think anyone would notice. Someone else unrelated, while on the flight, found this receiver on their phone/laptop, and played these sounds, possibly unwittingly.
AA may have found these devices after the incidents and just not reported it. Or they didn't take the reports seriously enough to comb the plane for problems. Or they gave up and would rather not lose the money by grounding the plane and tearing it down. (I don't have any aviation expertise, so I don't know exactly how disruptive it would be for them to to do this.)
Seems like whoever's doing this knows those systems unusually well, given that their own personnel don't know how it's done.
The audio might be analog, but they still have to "key" the mic which would be a digital input at the very least, or something even more complex such as a phone-like system where handsets have actual keypads and internal numbers.
I wouldn't expect the amplifier to be active at all unless a "call" is in progress (otherwise you might hear a constant noise, plus energy saving/thermal concerns and all that), so even if there was interference around, it shouldn't be getting amplified to an audible level unless something is actually "keying" the mic.
The traditional design uses a 70v or higher bus and a transformer at each speaker.
I would be really surprised if the PA speakers we anything but that-- electronics at each speaker is a repair/failure rate headache.
There's no way this is a sanctioned AA marketing program.
And yes, it is there so the attendants can talk directly to medical personnel on the ground while attending to the person needing care.
PS - “intercom” is probably going to give people the wrong idea. It’s a headset with a thick black 50-foot long cord, and when you plug it in, it dangles down on top of passengers before getting to the aisle. Also, you can’t hear it - only the person wearing the headset hears anything…
Well, if you make use of the complete system, sure. But if it just takes a regular TRRS jack; and if someone finds the socket and plugs their own headphones directly into it...
Flight attendants can plug in an aviation headset with a very, very long cord, and the pilots patch it into one of the radios in the cockpit. This allows the flight attendants to talk to a doctor on the ground while they're physically next to a passenger's seat.
That is pretty damn cool.
They should be able to run these sounds through the same software to see if it properly emulates a throat and mouth or not.
Back in 2003 or 2004 I was at University and my best friend went there too, a year behind me so different lives.
He’s a brilliant software developer today and we ran amok digitally before the age of 18 and things really got serious. Free Kevin. We had that spirit.
I was playing music and hanging out with Mary Jane and he got a job with campus IT. Through his personal explorations he gained access to the Admin of the campus AV system. I mean the entire thing - if it was connected, he could control it. I doubt that was in his job description or assigned permissions, but I didn’t ask.
His way of proving it is you could tell him “I’m in Calculus II at 11-12:30 on Tuesday in Building X Room XXX” and at 11:55 the overhead projector would turn on magically but because it wasn’t connected to anything locally it was just the light. Off, then on again, then off for good.
A little disruptive but hey it’s higher education and that taught me a valuable lesson about willpower and mischievousness with a Kokopelli spirit.
[0] https://www.errorinstruments.com/a-51153210/new-synthe-noise...
Directly, this should not be possible (input vs output on the amp), but in case the PA mic signal is routed to common amp after keying and preamp, this technically may be possible.
Now, if the "keying" includes some specific signal to allow the routing to PA, the question is if the "keying" signal can be sent from the seat's console. Then just play back the annoying track into the headphones jack.
>> Professional reverse engineer/hardware hacker/security analyst Andrew Tierney (aka Cybergibbons) dug up the Airbus 321 documents in this thread, and is stumped too.
It's not clear how the "seats" bus is isolated. I assume it's a bus, so that seats are addressable. The seats should at least be able to send the "need attendant" signal.
I hope a qualified technical analysis already pointed a few possible vectors for this. The question is in remediation and, well, identification of the perp(s).
Either way, at this moment the story seems to act more as an attracting magnet than repelling for the airline. Back during summer, it certainly was another level of annoying and disturbing.
We're a little more professional and we actually test stuff properly these days, so maybe it isn't that.
But that's a moot point if it really is happening on both B737s and A321s.
Planes fly higher and higher for better fuel efficiency and are starting to capture some of the lower ghosts.
Boeing is suppressing this.
“Do you believe in ghosts?”
“No, do you?”
“Yeah, a bit. Think about all the times people saw and heard ghosts. Every single one of those times was wrong? Seems unlikely.”
“Do you believe in angels?”
“No.”
“Then what about all the times people saw an angel?”
“Those were just ghosts.”
"Then how do you explain all the dead unicorns?"
Which causes us to need more climate change as we get more flights. Sad but true.
How helicopters can turn directly upside down, and fly that way[1], without reversing the direction of their props, that I don't get. (Thinking about it now, I think it turns from a helicopter to a hovercraft? Eg, this is a ground effect?)
[1] https://www.youtube.com/watch?v=gdDESYBdeSE
Also, tangentially, Magnus effect planes are a thing:
Obviously not the most probable explanation, but I found the story interesting.
https://mysteriesrunsolved.com/2020/07/ghosts-of-flight-401....
I'd like to say that even back then I mostly didn't really believe all that stuff, but I think maybe I kinda did, at least to a degree. Now I think it's all bollocks, and maybe that's one of the things I miss about being a kid... that sense of fascination with the mysterious and unknown and the sense of possibility that comes with allowing yourself to believe in the Bermuda Triangle, ghosts, yadda yadda.
I would like to have that again, if only for a few moments.
We thought people were bad on planes before yikes!
That seems unlikely. Definitely sounds human
But anyway, I'm playing devil's advocate with myself. I agree that's definitely a human voice on the intercom.
That shit is fucking aliens
Back when I was in college some twenty years ago, we somehow managed to stumble upon a phone number we could call that would connect us directly to passengers in some unknown elevator car in some distant city. We had great fun with that for a while.
Also, in the video linked in the blog post there doesn't seem to be much panic visible at all. I bet most people are laughing their asses off once the initial confusion is over.
I don't know if this is true or not either, but I don't think we can reject it on it's face.
Fortunately, planes are extremely robust and somehow Mexican airlines have an excellent safety record.
I was on a flight from Chicago to Raleigh a few years ago, and we were just starting to taxi out to the runway when the plane stopped and the pilot got on the PA and said something like "Sorry folks, we have a maintenance indicator (kind of like a 'check engine' light) that has come on. We have to go back to the terminal and get maintenance to check things out."
So we go back to the terminal, wait approximately 3 times longer than the expected lifetime of the universe (OK, that part might be an embellishment, but it felt like it at the time. The rest of this story is true though). Anyway, after a bit the pilot comes back on the PA and says something approximately like:
"OK folks, a $FRIZGIBBIT on this plane is broken. Thankfully these planes have two $FRIZGIBBITs and are certified to fly with just one working, so we're going to go ahead and get underway and get you all to Raleigh."
And that's just what they did. Frankly I was a bit nervous about the whole thing. I don't know what the hell a $FRIZGIBBIT is or what it does, and they say it's OK to fly with just one, but the way I figure it, somebody decided to put two in the first place for a reason. What if the remaining one breaks mid-flight??
Anyway, it all worked out in the end, or I wouldn't be here to tell this story, but the point is, planes do apparently fly on occasion with broken something-or-others.
It turns out that all three independent systems malfunctioned and the landing gear was fine.
I think I’d probably believe that it’s just some elctromechanical issue. I doubt fixing the PA system is going to be in the pilot’s manual and I’d expect flight attendants to be basically incompetent in dealing with such an issue.
I think I would guess either an issue with some prerecorded announcements system (eg playing bits of messages at strange speeds and at random) or some kind of electrical interference or short-circuiting problem (eg some glob of solder bouncing around inside some unimportant component).
It's obviously someone messing around. I assume AA claimed it's a glitch because PR doesn't want stories going around about how their planes are hackable, but it just makes them sound untrustworthy.