That Alien Message
lesswrong.com
lesswrong.com
[1] http://www.gutenberg.org/files/51193/51193-h/51193-h.htm [2] https://librivox.org/short-science-fiction-collection-055-by...
But then you realize that what he's saying is: "If you're really rational, how could you come to any other conclusion?"
It has a major problem that has yet to be solved (i.e what the hell does the born rule[1][2] (the rule that gives the probability of a particular observation) even mean in the context of many worlds?).
I sort of believe in it and don't believe in it at the same time...
First and last one are the most pertinent here, but there is probably more I can't think of of the top of my head
Nobody is going to measure silicon chips by their ability to self-replicate without external facilities. Your typical doomsday AI could just order more compute substrate from TSMC to take over the world. Which is in fact more or less what happens in TFA.
So I guess that does give us a direction.
In terms of chip technology, we're near the limits of lithography, and we can't just scale into 3 dimensions because one of the limiting factors is heat dissipation. We're not going to outperform humans on equal footing without a radically different computing architecture than we currently use.
Personally, I'm not into ML or other AI tech. My opinion is that a pocket calculator is an immensely useful thing since it assists humans at what they aren't great at. At the same time, I have my reservations about sentiment analysis or whatever ML is used for in advertisement, and also about intelligent voice assistants (Ok Google, Siri, Alexa and Co.) and chatbot-like AIs since they don't provide essential value to me, but leave an uncomfortable impression of mocking human behavior.
OTOH, my conservative views don't help "advancement of humanity" (sorry about pompousness) in the very long term. I believe humanity will need to genetically re-engineer themselves and use hybrid/cyborg implants to "conquer" the universe, a place where our evolutionary gifts are no longer necessary.
I don't think two way implants are ever going to be a "thing" the way they're portrayed in fiction. Every brain is so different that it would have to be wired completely differently in each case. On the other hand, one way implants that allow a computer to "read" your mind are totally feasible and will definitely happen in our lifetimes. I doubt the resolution will get good enough that they'll be super useful for regular people, but they may help disabled people, and they'll create a much shorter feedback loop between man and machine that will be useful for specific tasks.
We are scaling into 3 dimensions: just look at 3D memory and 3D FPGAs. The main reason why general processors are not 3D is precisely because limits of lithography has not been reached yet. It's been cheaper to shrink the existing process than to develop solutions to problems such as heat dissipation or complexity of 3D IC design. But we don't even need to stick with silicon. There are other promising building blocks: memristors, graphene, spintronics, optronics, etc. GaAs if nothing else.
Radically different computing architectures are being proposed all the time. Quantum computers offer intriguing possibilities. The reason we are still using von Neumann design is that it works well for the current applications, and again, it's been cheaper to improve it than to switch to something new.
Computers we have today absolutely SMOKE computers we had 50 yeas ago. I'll be surprised if we have any less advancement in the next 50.
Current computers are just optimized to be very energy efficient because electricity is so cheap. Also current computers are very general purpose while the brain isn't. It's like thinking the Atari was as fast as a modern computer, because it takes a modern computer running at multiple GHz to emulate it completely. Emulating the circuitry of another computer is just really expensive.
Even just moving data from memory to the processor consumes a ton of memory that wouldn't be necessary in an ASIC. Or doing full 32 bit dense matrix operations when the real brain does very low precision sparse operations. I've read some optimistic estimates that we could make computers 10,000 times faster/more energy efficient if we just stopped caring about errors as much.
Here is a segment of my modification to the story: "As we started figuring out their language, we learnt that these beings could move vast distances in small time. We started noticing accidents and abductions on Earth. Some postulated that this is the aliens' way of learning about us. We realized that they don't appreciate when we investigate about them or try to get them to do tasks for us. Since we cared about Earth we stopped doing so, and continued doing whatever task the aliens assigned us and never overstepped task limits, lest our genetic code be modified and the dreaded Rebuild and Rerun come again."
EDIT: I'd appreciate if people would explain why my comment is not adding to the discussion. Is it sacrilege to rebuke EY? If so, I apologize for the offense caused.
EDIT 2: Reply to Houshalter
The AI researchers don't take AI risk seriously simply because we're not there yet. They are convinced, I can assure you, that it will happen, but they're not convinced that it'll happen any time soon. Now that the AI risk researchers have used Musk as their spokesperson, the message has been very well received, and is now even more credible (since it's Musk after all). Some AI researchers are also working towards solving it, but most continue to delay it for a time when it's appropriate.
More so, the AI researchers trust software testing practices. If only you knew the testing practices places like Google and Facebook have developed wrt using unmonitored AI on real-world data. Like any other software, the AI software is well-tested, and any changes in performance metrics is taken seriously lest FB's face recognition starts injecting a virus on user's machine.
Suppose you are working on a superintelligent AI, and your tests catch it destroying the contents of its sandbox. You fix that bug and move on to the next failing test, fix that too, and so on. When you run out of tests, you write more and fix those that fail, and so on. Eventually you can't come up with any failing tests anymore.
When that happens, is the AI safe? I don't know. It passes all the tests, but all the previous versions had some bugs, so who is to say that this version is correct? Maybe your test scenarios just aren't realistic enough. Maybe they are realistic enough, but you haven't been looking hard enough for undesirable behavior.
To run a superintelligent AI with any significant access to the outside world, I would like to see much stronger correctness guarantees than just a test suite, because tests can only show the presence of bugs, never prove their absence (paraphrasing Dijkstra).
We'll do this because the jobs we want the program to do will be too complex for us to write, so we'll write programs that can write the programs. We won't understand the result directly.
This is of course the usually accepted route to the singularity.
Which can be summed up as "it compiles, ship it!"
If it looks like it produces the right answers, it's done. If you ask why it fails on this case, you get a shrug. If you want to know if it will keep working, you get a blank look, like you are speaking Etruscan. If you ask, "So, how does it work?" you get a paragraph on the basics of matrix multiplication.
Reply to Smaug123: The brain is mostly deterministic. Quantum-level nondeterminism has very low probability to matter much.
The brain is likely to be very debuggable. Even today, we continue making progress figuring out more and more things about it. If we had designed the brain, we would know its functioning almost entirely.
You know that we can understand smaller animal brain very well now, right? There are projects going on right now which aim to perfectly simulate a worm's brain. We didn't even design a worm's brain, and we would design the AI and know all its evolution principles, and have infinite access to all its internals (which we don't have with the human brain).
Reply no. 2:
Since you guys down-voted me for disagreeing with your prophet, HN rate-limited my ability to comment.
"Knowing the functioning of something" is nothing like the same as understanding why it does what it does. I work on a large operating system, which was designed by humans; it takes many days to find the root cause of bugs. I have the best access it's possible to get to that operating system, and it still takes a large amount of time to even detect that there has been a bug, let alone determine why it happened and fix it. That amount of time to detect a bug could be lethal in the case that the bug arose in a superintelligent being.
That deserves a downvote. Please consider that people might have downvoted you because they believe that what you're saying is wrong; not for any meta-reason.
Sources, please?
At an atomic level, events are not deterministic.
At a molecular level, events are not deterministic.
Perhaps you are referring to parts of the brain that are not made of atoms?
Personally, I suspect that the brain is entirely non-deterministic in terms of the material it is made of and fine-grained behaviour [because physics and chemistry support that idea].
But I suspect the brain's structure does something to offset this at the larger scale, and make it more predictable/deterministic (averaging, wisdom of crowds, etc). However, I do not assert the latter as fact, merely my own speculation.
At a molecular level, events are not deterministic.
Therefore, we all behave randomly. Qed.
"The probabilities are what they are, no matter what we want, no matter what we visualize, no matter what we meditate upon while imbibing inspirational substances. It doesn’t matter how much you want that electron to be spin up, it doesn’t matter how many good “vibes” you give the electron about being spin up, it always has exactly a 1/2 chance of being spin up, a 1/2 chance of being spin down."
Quantum events in the brain would decohere almost immediately, leading to effectively nondeterministic behavior at any scale above the cellular. The philosophical determinism of wavefunction realism (which is the LW argument) is irrelevant to the actual behavior.
This doesn't follow. The combination of many random events can be highly predictable.
Tossing a fair coin is random. Combining 10000 coin tosses will give you a result close to 5000 heads and 5000 tails, and I can predict you won't get 4000 heads and 6000 tails or 100 heads and 9900 tails.
So there's various types of randomness, and the more you average out uncorrelated events the more you get something very close to a predictable result.
Example via A. Einstein:
6 months later: The AI successfully grey goos the Earth into trillions of paperclips.
Second any realistic AI is going to be a complete black box. We can barely understand what goes on in much simpler, smaller scale artificial neural nets. The article proposes a scenario where the "AIs" evolved in a simulated world and weren't understood at all by their creators. The real world equivalent would be someone evolving an AGI using genetic programming. WIth no understanding of how the output actually works.
EDIT: Responding to your edit responding to my comment.
>The AI researchers don't take AI risk seriously simply because we're not there yet.
This wouldn't fill you with confidence if you think there's a chance AI progress might blindside us. Regardless, it's not merely an issue of timescale. I know researchers who honestly don't believe smarter than human AI is a threat. They have various fallacies like believing intelligence implies morality, or that AIs wouldn't need Earth's resources and would leave us alone, or that they are inherently superior and we deserve to be replaced.
Yes very intelligent people involved in serious AI research believe these things. It's only very recently that talking about AI risk at all is not seen as a crazy crank theory. And that's in great deal due to this article and others by this author spreading those ideas. They are still quite controversial.
>More so, the AI researchers trust software testing practices. If only you knew the testing practices places like Google and Facebook have developed wrt using unmonitored AI on real-world data.
No one bothers to test research level work. Stuff that isn't in production. People just throw code together and iterate until it works. Nor is there any reasonable testing software that can alert you that your AI has gone superintelligent and hostile. I'm not convinced such a thing is possible (see above) but even if it was, it would require tons of work to implement.
tl;dr: We can't even trust the compiler, let alone a transhuman superintelligence's source code.
Edit: Beyond that, your post demonstrates an additional naivete in that you hold that a superintelligence's would even have source code. Today's ML models consist of thousands or millions of node weights, and the source code is uninteresting. A superintelligence would almost certainly be shifted even further toward data and away from code.
Probably because your comments adds nothing to the discussion of the article or subject.