Simple, the whole is a hybrid. See my other explanation for more detail. Also, damaging an interface will severely impact whatever "listens" on that interface if the coupling is close enough.
Au contraire, the question is where you are getting this from. Your position is certainly not a scientific one and you are rather severely misinformed about what software can do (most people are). You are also severely misinformed about how complex a human brain is. Still, it is quite likely that a human brain is not enough to do what smart human beings can to. Just not enough computing power.
The problem is there is still absolutely no indication that human-equivalent intelligence can be implemented in machines. There is no credible theory how it can be done. The only known thing (automated theorem proving) does not scale up in this universe to what smart human beings can do when you assume a digital computer. There is just not enough matter, energy and time, so that cannot be it. And consciousness? The question just does not apply to software at all. It cannot do that. No really not. The best it can maybe do is a p-zombie. There are however about 50 years of intensive AI research that has completely failed in that direction. It has produced a lot of useful things, but they are not intelligence of the quality humans can do.
You also seem to be unaware that the term "emergent property" is a joke between scientists, not anything real. In physics, the whole is not more than it parts, as it cannot be. Expecting it to be _is_ some kind of mysticism.
There is no problem with empiric science without assuming physicalism. You make your hypothesis, you test it. If the test works, you get more confident. If it fails, you discard the hypothesis. No need for physicalism in any of that. Sure, as soon as an intelligent agent with free will is involved, things get a lot more fuzzy. But the same principles apply and it does not actually matter whether the free will is physical or not. What matters is that its effects can be observed.
Also do not forget, that "non-predictability" can be a good model for something. Quantum physics has some effects that are "random", like tunneling. Calling them "random" basically means "we do not know how they work, but they seem to follow a certain statistical model". It is a cop-out and a good physicist will admit that. As the statistical model is pretty good, this works so far. However, all physical theories so far have turned out to only be approximations and there is no reason to believe the current models are any better.
As to quantum mechanics, I doubt it is actually compatible with physicalism. The "magical" observer does not fit in there. Sure, most of physics and most of quantum mechanics _is_ compatible with physicalism, but you just need a few special effects that are not and physicalism is invalid, as it is an absolute faith. Also note that if dualism is the correct model, it may well be provable eventually. If physicalism is, then this is not provable as you cannot demonstrate that there are no circumstances where it is violated. The only way to argue for physicalism is to assume it as truth (without any hard evidence) and then to demand that others prove otherwise or shut up. That is pretty close to what many religions do. It is also not convincing at all.
Note that I am not arguing for dualism as the true model here (that is a different discussion), I am arguing that at this time, Science does not know whether it is one or the other and that assuming one of the two as truth is not scientifically valid.
Make sure the people are trustworthy, have them sign an NDA and that is it. How do you think really security-critical software gets external reviews? Also, even work on premises would let them steal the important parts of the code if they were so inclined, it is not that hard.
One thing you can do for trustworthiness is look for people and small (!) consultancies that already work with stuff of comparable sensitivity and difficulty.
Oh, and pay really well. Nothing makes contractors care less than being treated like cheap monkeys.
The neuro-"scientists" have this little problem that consciousness does not fit their models at all. (Neither does intelligence, but they have not noticed that little problem so far...) Hence they invent colorful non-explanations to misdirect others and themselves by claiming there is no problem.
The real problem is that they are not doing science. They assume physicalism as ground truth and that is a religious approach, not a scientific one. Actual scientists would realize that the question is still open at this time (but the more we know, the more it goes towards "some kind of dualism", although certainly not a religious one) and would search in both directions. They do not.
I do agree that the question does not make sense. Perhaps the strongest thing human beings find when entering this world is that they have a consciousness. (Well, unless this is solipsism where you are all p-zombies...) That is ground truth. Physics is less solid in comparison and physics actually does not seem to have a place for something like consciousness. It cannot explain it. In a rather strong sense, it does not apply to the question. Sure, there must be some kind of interface between consciousness and physical reality, but so far, it is completely unclear how that works. Interestingly, quantum-physics has the concept of an "observer", but the observer seems to be extra-physical as it can do "magic" and drag superposed quantum-states into a definite state. No purely physical object should be able to do that and yet it seems human beings can.
The Next AMD generations will offer that, on the same AM4 socket as the APUs. For the last offerings, the socket was different, but you still had both choices with AMD.
This whole thing only helps catching small-time criminals as it requires the criminal in question to be unaware or too stupid to understand what is going on. Hence it is a) not needed to fight small-time crime, as the stupid criminals usually make plenty of other mistakes and b) for competent criminals it does not help, as nothing will be on the phone.
This is not about fighting crime at all, unless law-enforcement has gotten terminally incompetent at their job.
It is complying with the orders when they _try_ to do it (which has no result as it is impossible). What they want to do is to be able to reject the orders outright, and that is the only sane thing to do.
Of course "sane" is not something most in the legal profession can do, as they are all living in their own little fantasy world where they are kings and define what reality is.
That is uninformed bullshit. Breaking block ciphers with Quantum Computers (if they turn out to be feasible at all for the sizes required) _halves_ the number of bits. AES256 would go down to 128 bit strength, which is still quite unbreakable.
Anyone who thinks AES 256 (what iPhones are encrypted with) can be cracked by any computer doesn't understand the math.
There are a lot of those people, and usually they are Dunning-Kruger sufferers too, i.e. they have no clue that they have no clue.
Without a severe vulnerability in the cipher or its use, AES 256 can likely not be broken in this universe. Even a working Quantum Computer of sufficient size (and it is doubtful that one is possible, _and_ the model may break down at these complexities anyways) would not be able to do it.
One problem: It is not possible for most calculations and it is unknown whether it ever will be. And another problem: Even if possible with a sufficiently general set of operations, its impact will be rather limited.
Email, the WWW, computers, computer networks, etc. are all a much, mucg more important than this specialized solution for a problem that has other solutions as well.
The question can be answered with a resounding "No, and why are you asking stupid questions?"
It is as the IoT people never even have heard of the, by now, 30+ years of history of Internet security fails. These must be the dumbest, most arrogant and most clueless developers, lead by managers of the same quality. It is high time that we get legally actionable gross negligence for manufacturers that ignore Internet security best practices.
Nice additional fact: Because there are now quite a few refugees coming into Europe that claim to be underage (and if they are they have quite a few advantages), a larger European newspaper looked into how precisely a medical professional can determine age when having full access to the person (i.e. _not_ just pictures or video). Turns out that even with the person standing right before you and all modern medical equipment at hand, you can still be off and regularly will be by up to two (!) years. With just video or pictures, it will be more.
Unfortunately, this has indeed all the tell-tales of a witch-hunt. Just like the "war on drugs" has, for example, and with the same people driving it and benefiting from it. With this latest step by the FBI, I have serious doubt anybody there actually cares much about the victims. They do quite obviously care very much about increasing their "business", and are willing to do whatever it takes. Pretty bad.
If that is true (and it might be, as economic theory rather strongly suggests it), the the FBI and others are actively encouraging and increasing child abuse, in particular for for the case of drawings and other material that falls under CP, but where no actual child was hurt in order to create it. This whole issue is really a lot murkier than they would have us believe.
Slashdot Mirror
Simple, the whole is a hybrid. See my other explanation for more detail. Also, damaging an interface will severely impact whatever "listens" on that interface if the coupling is close enough.
Au contraire, the question is where you are getting this from. Your position is certainly not a scientific one and you are rather severely misinformed about what software can do (most people are). You are also severely misinformed about how complex a human brain is. Still, it is quite likely that a human brain is not enough to do what smart human beings can to. Just not enough computing power.
The problem is there is still absolutely no indication that human-equivalent intelligence can be implemented in machines. There is no credible theory how it can be done. The only known thing (automated theorem proving) does not scale up in this universe to what smart human beings can do when you assume a digital computer. There is just not enough matter, energy and time, so that cannot be it. And consciousness? The question just does not apply to software at all. It cannot do that. No really not. The best it can maybe do is a p-zombie. There are however about 50 years of intensive AI research that has completely failed in that direction. It has produced a lot of useful things, but they are not intelligence of the quality humans can do.
You also seem to be unaware that the term "emergent property" is a joke between scientists, not anything real. In physics, the whole is not more than it parts, as it cannot be. Expecting it to be _is_ some kind of mysticism.
There is no problem with empiric science without assuming physicalism. You make your hypothesis, you test it. If the test works, you get more confident. If it fails, you discard the hypothesis. No need for physicalism in any of that. Sure, as soon as an intelligent agent with free will is involved, things get a lot more fuzzy. But the same principles apply and it does not actually matter whether the free will is physical or not. What matters is that its effects can be observed.
Also do not forget, that "non-predictability" can be a good model for something. Quantum physics has some effects that are "random", like tunneling. Calling them "random" basically means "we do not know how they work, but they seem to follow a certain statistical model". It is a cop-out and a good physicist will admit that. As the statistical model is pretty good, this works so far. However, all physical theories so far have turned out to only be approximations and there is no reason to believe the current models are any better.
As to quantum mechanics, I doubt it is actually compatible with physicalism. The "magical" observer does not fit in there. Sure, most of physics and most of quantum mechanics _is_ compatible with physicalism, but you just need a few special effects that are not and physicalism is invalid, as it is an absolute faith. Also note that if dualism is the correct model, it may well be provable eventually. If physicalism is, then this is not provable as you cannot demonstrate that there are no circumstances where it is violated. The only way to argue for physicalism is to assume it as truth (without any hard evidence) and then to demand that others prove otherwise or shut up. That is pretty close to what many religions do. It is also not convincing at all.
Note that I am not arguing for dualism as the true model here (that is a different discussion), I am arguing that at this time, Science does not know whether it is one or the other and that assuming one of the two as truth is not scientifically valid.
Make sure the people are trustworthy, have them sign an NDA and that is it. How do you think really security-critical software gets external reviews? Also, even work on premises would let them steal the important parts of the code if they were so inclined, it is not that hard.
One thing you can do for trustworthiness is look for people and small (!) consultancies that already work with stuff of comparable sensitivity and difficulty.
Oh, and pay really well. Nothing makes contractors care less than being treated like cheap monkeys.
The neuro-"scientists" have this little problem that consciousness does not fit their models at all. (Neither does intelligence, but they have not noticed that little problem so far...) Hence they invent colorful non-explanations to misdirect others and themselves by claiming there is no problem.
The real problem is that they are not doing science. They assume physicalism as ground truth and that is a religious approach, not a scientific one. Actual scientists would realize that the question is still open at this time (but the more we know, the more it goes towards "some kind of dualism", although certainly not a religious one) and would search in both directions. They do not.
I do agree that the question does not make sense. Perhaps the strongest thing human beings find when entering this world is that they have a consciousness. (Well, unless this is solipsism where you are all p-zombies...) That is ground truth. Physics is less solid in comparison and physics actually does not seem to have a place for something like consciousness. It cannot explain it. In a rather strong sense, it does not apply to the question. Sure, there must be some kind of interface between consciousness and physical reality, but so far, it is completely unclear how that works. Interestingly, quantum-physics has the concept of an "observer", but the observer seems to be extra-physical as it can do "magic" and drag superposed quantum-states into a definite state. No purely physical object should be able to do that and yet it seems human beings can.
I expect they will know enough on what Nvidia infringes to give them a quiet but effective warning to not go down that road.
The Next AMD generations will offer that, on the same AM4 socket as the APUs. For the last offerings, the socket was different, but you still had both choices with AMD.
From actual performance comparisons, it does not look like AMD has given up at all. I also would say that HBM is not a sign of having given up either.
This whole thing only helps catching small-time criminals as it requires the criminal in question to be unaware or too stupid to understand what is going on. Hence it is a) not needed to fight small-time crime, as the stupid criminals usually make plenty of other mistakes and b) for competent criminals it does not help, as nothing will be on the phone.
This is not about fighting crime at all, unless law-enforcement has gotten terminally incompetent at their job.
Or alternatively, there will be a lot of smuggling of iPhones into China in the future.
It is complying with the orders when they _try_ to do it (which has no result as it is impossible). What they want to do is to be able to reject the orders outright, and that is the only sane thing to do.
Of course "sane" is not something most in the legal profession can do, as they are all living in their own little fantasy world where they are kings and define what reality is.
That is uninformed bullshit. Breaking block ciphers with Quantum Computers (if they turn out to be feasible at all for the sizes required) _halves_ the number of bits. AES256 would go down to 128 bit strength, which is still quite unbreakable.
Anyone who thinks AES 256 (what iPhones are encrypted with) can be cracked by any computer doesn't understand the math.
There are a lot of those people, and usually they are Dunning-Kruger sufferers too, i.e. they have no clue that they have no clue.
Without a severe vulnerability in the cipher or its use, AES 256 can likely not be broken in this universe. Even a working Quantum Computer of sufficient size (and it is doubtful that one is possible, _and_ the model may break down at these complexities anyways) would not be able to do it.
Oh, you mean you want a propaganda-platform for everybody? That will turn out well, I am sure...
Don't tell me, you like to blame victims, no?
Indeed. And hence bad security must be made a significant cost factor for those making devices with it.
Can't argue with that ;-)
Possible, but will take a while ;-)
One problem: It is not possible for most calculations and it is unknown whether it ever will be.
And another problem: Even if possible with a sufficiently general set of operations, its impact will be rather limited.
Email, the WWW, computers, computer networks, etc. are all a much, mucg more important than this specialized solution for a problem that has other solutions as well.
The question can be answered with a resounding "No, and why are you asking stupid questions?"
And that makes the devices more secure how?
It is as the IoT people never even have heard of the, by now, 30+ years of history of Internet security fails. These must be the dumbest, most arrogant and most clueless developers, lead by managers of the same quality. It is high time that we get legally actionable gross negligence for manufacturers that ignore Internet security best practices.
Nice additional fact: Because there are now quite a few refugees coming into Europe that claim to be underage (and if they are they have quite a few advantages), a larger European newspaper looked into how precisely a medical professional can determine age when having full access to the person (i.e. _not_ just pictures or video). Turns out that even with the person standing right before you and all modern medical equipment at hand, you can still be off and regularly will be by up to two (!) years. With just video or pictures, it will be more.
Unfortunately, this has indeed all the tell-tales of a witch-hunt. Just like the "war on drugs" has, for example, and with the same people driving it and benefiting from it. With this latest step by the FBI, I have serious doubt anybody there actually cares much about the victims. They do quite obviously care very much about increasing their "business", and are willing to do whatever it takes. Pretty bad.
If that is true (and it might be, as economic theory rather strongly suggests it), the the FBI and others are actively encouraging and increasing child abuse, in particular for for the case of drawings and other material that falls under CP, but where no actual child was hurt in order to create it. This whole issue is really a lot murkier than they would have us believe.