• Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Using AI In Audio Debates & GR Research [Video]

If you gaze long into an AI, the AI also gazes into you
I have no idea if this post was intended for levity but there is truth to it. Currently AI platforms will answer with absolute confidence the incorrect answer. In my company we are required to do 4 hours per week of AI training. It’s a really smart choice. As we have seen with Danny, using AI to verify claims is flawed on so many levels. I would hope Amir continues to evaluate the validity of AI when making claims moving forward. IMO he has it totally correct with this sites current stance.
 
AI is guided by its prompts. It does NOT think. It serves to please you. If you ask it why is it good to upgrade your cables it will likely tell you something like, “using high end cables will have benefit, many audiophiles notice great improvements, although measurements have not verified this claim.” If you ask it, I’ve noticed no proof expensive cables make a difference, should I upgrade? The answer will be something more like, “There is no scientific evidence to support a properly functioning cable to be different from another.” And it’s likely might even reference this site. AI, like Gemini, is simply drawing from any reference it can. It wants to verify your assertions and is simply broken for any kind of real data. That said, I could ask Gemini, do most people feel better about their cable purchase if they spend more? It likely would look at that prompt and give you marketing data that would support that assertion.
...it seems like it thinks but it references...it is programmed to keep us engaged...it is very good at that...one certainly needs to know what to ask it...
 
I have no idea if this post was intended for levity but there is truth to it. Currently AI platforms will answer with absolute confidence the incorrect answer. In my company we are required to do 4 hours per week of AI training. It’s a really smart choice. As we have seen with Danny, using AI to verify claims is flawed on so many levels. I would hope Amir continues to evaluate the validity of AI when making claims moving forward. IMO he has it totally correct with this sites current stance.
...we are training it...it is training us...
 
...we are training it...it is training us...
...what is the deal with these multiple one liners...

...are you trying to build an online personality...

...each one of these one liners is like trying to be a zinger...

...this is very weird of you...
 
...OK...

...this was too funny...

What do you have to lose? You end up going on a wild goose chase whilst being lead by a hallucinating AI.

Use AI as a glorified search engine. I once asked AI a question for which I did not know the answer. It gave me a confident sounding answer. So I asked it for a source. It linked me back to a post I made on ASR. So yeah, thanks AI.
 
...is it OK for me to like how something sounds if I am not too worried about the facts...???...

Your liking something is not in opposition to the facts - it is a fact that you have a preference.
 
...is it OK for me to like how something sounds if I am not too worried about the facts...???...
As long as your aim is not to convince me, sure. The problem is that people want to turn their views into facts but lack the foundation, or even an attempt to verify their views.
 
...we are training it...it is training us...
The assumption is that the only way for a machine to pass the Turing test is for the machine to get smarter. But the other way to pass the Turing test is for the human to get dumber.

By altering our behavior in response to machines, we are sometimes helping the latter path.
 
1779275276828.jpeg
 
...I just don't want to see AI rejected as irrelevant...

...IMHO that would be akin to rejecting electronic instruments or recorded music..

...understand there are some musicians who do...

(...personally, I regard such people as snobbish whack jobs, but that's talent...)

...EDIT: I should have watched the video before posting and I would have understood there is little danger of Amir dismissing AI as irrelevant, but a lot of people seem to...

...I am not generally interested in arguments...(this one is way over me head)...I now understand this video is more of a PSA than an argument...
 
Last edited:
I am often staggered at how wrong AI can be. And dangerously so on more than one occasion. When "it" gets stuff wrong it gets it totally wrong like Black is White type wrong. And when you pull it up it just apologizes... and gets it wrong a different way!
Amir's last point is key. Don't ask AI for something you don't already know the answer to or can check.
A World of Pain lies ahead otherwise... or this guy Danny...same thing really :):)
 
There is quite a lot of AI bashing going on here. Folks just have to keep in mind, you don't use AI as a replacement, you use it as a supplement. At least with the state of AI today.

I have used AI with great success. But I am a diligent user.
 
Show us a footnote in any of your posts.
Bold claims such as yours are subject to citation upon request, if none is provided, then claims are freely dismissed.
 
Bold claims such as yours are subject to citation upon request, if none is provided, then claims are freely dismissed.
I guess that I will double down.

As is in Hard Science, Social Science and Political Science there is no proof. However there is always strong opion.
 
I guess that I will double down.

As is in Hard Science, Social Science and Political Science there is no proof. However there is always strong opion.
Dismissed. You can triple down, to only be dismissed again.
 
Back
Top Bottom