Copilot AI. Can you trust it?
Thread Starter
Joined: Mar 2023
Posts: 65
Likes: 32
From: sydney
Copilot AI. Can you trust it?
The Question and Copilot AI's answer is as shown below:
Do you agree with "copilot AI" (or not?) Give your reasoning.
Question:
What pitot-static system effects will be seen on the two Pilot's indicated airspeed instruments when the air data modules of a 777 aircraft's static systems are subjected to cryogenic ambient temperatures and low atmospheric pressures, resulting in a "stuck" value – during a descending level change from a very high altitude?
If that static pressure remains constant during the initial descent, will the airspeed tend to underread or overread?
Answer:
Assuming a normally functional pilot/copilot (and tertiary) pitot system, if all three static pressure ADM’s are in agreement and remain at a constant (frozen) value during a descent from high altitude, the airspeed indicator will typically tend to underread. This is because, in normal conditions, static pressure should increase as the aircraft descends to lower altitudes. If the (all three) ADMs' static pressure output remains constant instead of increasing, the pitot-static system will interpret this as the aircraft maintaining a higher altitude than it actually is. Consequently, the airspeed indicator will show a lower speed than the correct/actual indicated airspeed.
This scenario can occur if there is a blockage or malfunction in the static port, making it stick and preventing it from accurately sensing the increasing pressure during descent. It's crucial for pilots to recognize and address such issues to ensure safe flight operations.
Is there a specific scenario or issue you are considering?
Do you agree with "copilot AI" (or not?) Give your reasoning.
Question:
What pitot-static system effects will be seen on the two Pilot's indicated airspeed instruments when the air data modules of a 777 aircraft's static systems are subjected to cryogenic ambient temperatures and low atmospheric pressures, resulting in a "stuck" value – during a descending level change from a very high altitude?
If that static pressure remains constant during the initial descent, will the airspeed tend to underread or overread?
Answer:
Assuming a normally functional pilot/copilot (and tertiary) pitot system, if all three static pressure ADM’s are in agreement and remain at a constant (frozen) value during a descent from high altitude, the airspeed indicator will typically tend to underread. This is because, in normal conditions, static pressure should increase as the aircraft descends to lower altitudes. If the (all three) ADMs' static pressure output remains constant instead of increasing, the pitot-static system will interpret this as the aircraft maintaining a higher altitude than it actually is. Consequently, the airspeed indicator will show a lower speed than the correct/actual indicated airspeed.
This scenario can occur if there is a blockage or malfunction in the static port, making it stick and preventing it from accurately sensing the increasing pressure during descent. It's crucial for pilots to recognize and address such issues to ensure safe flight operations.
Is there a specific scenario or issue you are considering?
Joined: Feb 2018
Posts: 800
Likes: 279
From: East of Westralia
Pick up a good ATPL theory book, or other decent aeronautical theory book and learn the theory from them.
Leave large language models such as the one you quoted for other non aviation tasks. They are severely limited in vetting the truth from what they scrape.
Leave large language models such as the one you quoted for other non aviation tasks. They are severely limited in vetting the truth from what they scrape.
Joined: Jun 2008
Posts: 240
Likes: 35
From: Cambridge UK
> Leave large language models such as the one you quoted for other non aviation tasks.
> They are severely limited in vetting the truth from what they scrape.
An example from the world of pottery:
ChatGPT is completely wrong about the cause of glaze crazing!
https://digitalfire.com/picture/3159
> They are severely limited in vetting the truth from what they scrape.
An example from the world of pottery:
ChatGPT is completely wrong about the cause of glaze crazing!
https://digitalfire.com/picture/3159
Joined: Sep 2018
Aviation Qualifications: ATPL
Posts: 402
Likes: 240
From: Seattle
Has anyone ever been in a sim/real life and said something like “Shutdown Engine 1” and mister magoo sitting next to you starts to touch Engine 2? No?
Humans can’t make mistakes? Why hold AI to a higher standard than magoo number two? AI just has to be as good as. And shortly (in days/weeks/months) it will be better than humans.
Humans can’t make mistakes? Why hold AI to a higher standard than magoo number two? AI just has to be as good as. And shortly (in days/weeks/months) it will be better than humans.
Joined: Jul 2013
Aviation Qualifications: Non-Aircrew
Posts: 5,682
Likes: 3,346
From: Everett, WA
It's all pretty much a moot point, at least for the near future. The FAA is on record as stating that they will not certify any airborne software that incorporates AI (and I believe EASA has said something similar).
The issue is pretty basic - all the testing done to certify airborne software requires that the outputs for a given input are always the same - and AI doesn't always do that.
Now, at some future point, I expect that AI will progress to the point where the regulatory authorities will need to re-access that position - and it will become apparent that the advantages of AI outweigh the disadvantages.
But I don't see that happening for a long time - probably decades.
The issue is pretty basic - all the testing done to certify airborne software requires that the outputs for a given input are always the same - and AI doesn't always do that.
Now, at some future point, I expect that AI will progress to the point where the regulatory authorities will need to re-access that position - and it will become apparent that the advantages of AI outweigh the disadvantages.
But I don't see that happening for a long time - probably decades.
Joined: Mar 2005
Posts: 83
Likes: 11
From: Feet on the rudder pedals
The Question and Copilot AI's answer is as shown below:
Do you agree with "copilot AI" (or not?) Give your reasoning.
Question:
What pitot-static system effects will be seen on the two Pilot's indicated airspeed instruments when the air data modules of a 777 aircraft's static systems are subjected to cryogenic ambient temperatures and low atmospheric pressures, resulting in a "stuck" value – during a descending level change from a very high altitude?
If that static pressure remains constant during the initial descent, will the airspeed tend to underread or overread?
Answer:
Assuming a normally functional pilot/copilot (and tertiary) pitot system, if all three static pressure ADM’s are in agreement and remain at a constant (frozen) value during a descent from high altitude, the airspeed indicator will typically tend to underread. This is because, in normal conditions, static pressure should increase as the aircraft descends to lower altitudes. If the (all three) ADMs' static pressure output remains constant instead of increasing, the pitot-static system will interpret this as the aircraft maintaining a higher altitude than it actually is. Consequently, the airspeed indicator will show a lower speed than the correct/actual indicated airspeed.
This scenario can occur if there is a blockage or malfunction in the static port, making it stick and preventing it from accurately sensing the increasing pressure during descent. It's crucial for pilots to recognize and address such issues to ensure safe flight operations.
Is there a specific scenario or issue you are considering?
Do you agree with "copilot AI" (or not?) Give your reasoning.
Question:
What pitot-static system effects will be seen on the two Pilot's indicated airspeed instruments when the air data modules of a 777 aircraft's static systems are subjected to cryogenic ambient temperatures and low atmospheric pressures, resulting in a "stuck" value – during a descending level change from a very high altitude?
If that static pressure remains constant during the initial descent, will the airspeed tend to underread or overread?
Answer:
Assuming a normally functional pilot/copilot (and tertiary) pitot system, if all three static pressure ADM’s are in agreement and remain at a constant (frozen) value during a descent from high altitude, the airspeed indicator will typically tend to underread. This is because, in normal conditions, static pressure should increase as the aircraft descends to lower altitudes. If the (all three) ADMs' static pressure output remains constant instead of increasing, the pitot-static system will interpret this as the aircraft maintaining a higher altitude than it actually is. Consequently, the airspeed indicator will show a lower speed than the correct/actual indicated airspeed.
This scenario can occur if there is a blockage or malfunction in the static port, making it stick and preventing it from accurately sensing the increasing pressure during descent. It's crucial for pilots to recognize and address such issues to ensure safe flight operations.
Is there a specific scenario or issue you are considering?
Beacon Outbound
Joined: Mar 2002
Aviation Qualifications: ATPL
Posts: 710
Likes: 22
From: "Home is were the answer machine is"
The original question has nothing to do with having AI or an AI assistant on the flight deck. Copilot AI is the name of Microsoft's AI offering: https://copilot.microsoft.com
Copilot AI has the same kind of limitations that ChatGPT (which I showed earlier) has:
Copilot AI has the same kind of limitations that ChatGPT (which I showed earlier) has:
Last edited by IRRenewal; 13th November 2024 at 20:31. Reason: added picture
Joined: Dec 2002
Posts: 2,775
Likes: 353
From: UK
Think first, then Question
Thus this thread is a very good example of the need to consider our interaction - how we communicate with humans, or the differences with AI; how the subject is framed (described), and the specific question posed with context.
Remember that ChatGPT based aids, knowledge sources, will be influenced by this social chatbot, where the machines' objective is to satisfy the human need, as represented by the input question and contextual description.
Remember that ChatGPT based aids, knowledge sources, will be influenced by this social chatbot, where the machines' objective is to satisfy the human need, as represented by the input question and contextual description.
Last edited by safetypee; 13th November 2024 at 22:05.
Thread Starter
Joined: Mar 2023
Posts: 65
Likes: 32
From: sydney
No relevant Responses
Interesting that nobody above has challenged the basic premise of the AI response, i.e., that the pilots' ASI's (with a non-responsive "stuck" pressure) will underread in a descent. Perhaps you should re-read what Bernoulli has had to say on the subject and think it through in the nether recesses of your amygdala. Whilst it is readily agreed that one should not take an authoritative answer at face value, you should also be able to demonstrate just why it may be wrong.... or even argue for what reason it is right (or conditionally incorrect).
Reminds me of a proof of life document sent to me by a pension provider. It was ( I foolishly assumed) required to be presented to a Justice of the Peace for attestation and stamping/signature. In my Drivers' Licence photo I have a beard - but I presented myself clean-shaven and shorn. She (the JP) refused to sign me up, saying that this person depicted on the licence could be anyone once shorn. "You appear to be 20 years younger than this person". Upon sighting the birth certificate with the exact same D.o.B. she claimed that I could be the gentleman's son. Upon sighting my signature, she claimed that she couldn't verify it against the one shown on my licence as she was not a signature expert. I offered to give her a blood sample which she rejected also. Utility bills she claimed, proved nothing.
As it turned out, the whole demurrral hinged upon the requirement to testify that she had known me for a period of not less than five years (she'd recently blown in from NSW). Upon sighting that 5 year proviso, I realised that a JP's signature was a step too far and (within her view) got the aged library cleaner (who didn't know me from Adam) to sign me up. Worked OK and I'm still drawing my pension. Planning to get copilot AI to sign me up again next year... with a bogus signature. Should be manageable.
Reminds me of a proof of life document sent to me by a pension provider. It was ( I foolishly assumed) required to be presented to a Justice of the Peace for attestation and stamping/signature. In my Drivers' Licence photo I have a beard - but I presented myself clean-shaven and shorn. She (the JP) refused to sign me up, saying that this person depicted on the licence could be anyone once shorn. "You appear to be 20 years younger than this person". Upon sighting the birth certificate with the exact same D.o.B. she claimed that I could be the gentleman's son. Upon sighting my signature, she claimed that she couldn't verify it against the one shown on my licence as she was not a signature expert. I offered to give her a blood sample which she rejected also. Utility bills she claimed, proved nothing.
As it turned out, the whole demurrral hinged upon the requirement to testify that she had known me for a period of not less than five years (she'd recently blown in from NSW). Upon sighting that 5 year proviso, I realised that a JP's signature was a step too far and (within her view) got the aged library cleaner (who didn't know me from Adam) to sign me up. Worked OK and I'm still drawing my pension. Planning to get copilot AI to sign me up again next year... with a bogus signature. Should be manageable.
