Nathan1123

Humans don't understand how we do most things

Some problems of mathematics like the Halting Problem and the Busy Beaver Problem are uncomputable, meaning that it is mathematically proven that any Turing-complete computer is physically incapable of solving the problem no matter how sophisticated its hardware or software is. Some algorithms on a Turing machine can be used to solve special cases of these problems, but the general case is known to be uncomputable. In the case of Busy Beavers, a computer program could check each combination of programs one-by-one, but this wouldn't work for BB(6) or higher because that is believed to be larger than the number of particles in the Universe.

Wikipedia has a full list of Undecidable problems.

To what extent does this suggest that a GAI will be limited in its capabilities, even in the most optimistic scenario of AI "takeoff"? Could this present an exploitable weakness that could be used to keep an uncooperative AI contained or shut down?

-3

The Stanley Parable: Making philosophy fun

Hello,

I find that a fundamental premise towards the development of AI is the ability to dissect all the internal calculations used by the human mind to make rational decisions, and then replicate those calculations in an electronic system. On a surface level, this works very well for many kinds of problem solving. In our everyday lives, we use logical reasoning to make rational decisions all the time, with the understanding that some decisions come from emotion and intuition instead. There are also some cognitive behavior that is completely subconscious or autonomous (such as walking or talking), which ends up being the most challenging skills to replicate for a machine.

For any kind of... (read 357 more words →)

Hello,

If you are unaware, The Stanley Parable is a video game originally released in 2013, and later re-released 2022 under the title The Stanley Parable: Ultra Deluxe. The player takes control of the titular character Stanley, as the present and future actions of his life are being dictated by a disembodied narrator (played by Kevan Brighting). The player is presented with a series of branching paths, in reaction to the suggestions or commands from the Narrator, and thereby making a deliberate impact on the outcome of the story.

The Stanley Parable is a game about choices, and each possible ending carefully deconstructs the nature of game design, story telling, decisions, and freewill from... (read 624 more words →)

Replying toEisenhower's Atoms for Peace Speech

Eisenhower's Atoms for Peace Speech

While this is an example of international cooperation in the face of mutually-assured destruction, there is some historical context that shows why this is effective, in my opinion: First, because the destructive power of nuclear weapons was already realized by Hiroshima and Nagasaki. Second, because the US no longer had a monopoly on nuclear weapons and determined that overcoming the USSR by force was no longer an option. Third, because the world was still recovering from the Second World War and shared a universal desire for peaceful resolutions.

Replying toIs there any literature on using socialization for AI alignment?

Is there any literature on using socialization for AI alignment?

The implication that I didn't think to spell out is that the AI should be programmed with the capacity for empathy. It's more of a proposal of system design than a proposal of governance. Granted, the specifics of that design would be its own discussion entirely

Is there any literature on using socialization for AI alignment?

Hello,

I was recently thinking about the question of how humans achieve alignment with each other over the course of our lifetime, and how that process could be applied to an AGI.

For example, why doesn't everyone shop lift from the grocery store? A grocery store isn't as secure as Fort Knox, and if one considered every possible policy that results in obtaining groceries then they may discover that shop lifting is more efficient than obtaining money from a legitimate job. That may or may not be the best example, but I'm sure LW is quite familiar with the concept that lies at the heart of the problem of AI alignment: what humans consider... (read 493 more words →)

Replying toA Confession about the LessWrong Team

A Confession about the LessWrong Team

You were the chosen one, Anakin!

Replying toPausing AI Developments Isn't Enough. We Need to Shut it All Down by Eliezer Yudkowsky

Pausing AI Developments Isn't Enough. We Need to Shut it All Down by Eliezer Yudkowsky

I think the harsh truth is that no one cared about Nuclear Weapons until Hiroshima was bombed. The concept of one nation "disarming" AI would never be appreciated until somebody gets burned.

Replying toCould the simulation argument also apply to dreams?

Could the simulation argument also apply to dreams?

I wonder if you could expand more on this observation. So you are saying that a dream is operating on a very limited dataset on a person, not an exact copy of information ("full description"). Do I understand right?

I sort of do intend of it as a kind of reductio, unless people find reason for this "Dream Hypothesis" to be taken seriously.

Could the simulation argument also apply to dreams?

We know that a dream can be real, but who ever thought that reality could be a dream? We exist, of course, but how, in what way? As we believe, as flesh-and-blood human beings, or are we simply parts of someone's feverish, complicated nightmare?

Charles Beaumont

In reading about the summary of Bostrom's Simulation Argument, it seems to speculate on the realization of future technology which may or may not ever exist. That is, computer technology that is capable of making simulations that are not only vividly accurate to the behavior of people from the distant past, but also contain a consciousness that are convinced they are the very people they are imitating. And... (read 873 more words →)

Replying toWhat is the probability that a superintelligent, sentient AGI is actually infeasible?

What is the probability that a superintelligent, sentient AGI is actually infeasible?

I don't see anything in that scenario that prevents a human-level AGI from using a collection of superintelligent tool AIs with a better interface to achieve feats of intelligence that humans cannot, even with the same tool AIs.

At that point, it wouldn't functionally be different than a series of tool AIs being controlled directly by a human operator. If that poses risk, then mitigations could be extrapolated to the combined-system scenario.

What fundamental law of the universe would set a limit right there, out of all possible capacities across every possible form of computing substrate?

I'm not trying to imply there is something about the human mind specifically that forces a limit to computing... (read more)

What is the probability that a superintelligent, sentient AGI is actually infeasible?

Hello,

I am not at all adverse to discussing contingencies for things that are either uncertain or unlikely, but I'm curious what the general consensus is for how likely a superintelligent AGI scenario actually is.

To be clear, I am certainly aware that advances in AI have made leaps and bounds throughout my own lifetime alone, with public feats such as AlphaGo far exceeding the expectations of its creators.

But just because there is a trend over a certain period of time doesn't mean that it will stay that way forever, or develop in a way that one expects intuitively. For example, population models tend to grow quickly at first, but flatten out over time.... (read more)

-3

An Uncanny Prison

In reading about proposals for AI Boxing, I notice two excerpts jump out at me. First, is a proposed scenario to uncover an unfriendly AI:

A virtual world between the real world and the AI, where its unfriendly intentions would be first revealed

And the second excerpt of interest is a technique by which an AI can trick its owners into letting it escape:

Other, more speculative ways include: threatening to torture millions of conscious copies of you for thousands of years, starting in exactly the same situation as in such a way that it seems overwhelmingly likely that you are a simulation

The first scenario, attempting at containment, would fail if the AI is smart... (read 425 more words →)

Infant AI Scenario

Hello,

In reading about the difficulty in training an AGI to appreciate and agree with human morals, I start to think about the obvious question, "how do humans develop our sense of morals?" Aside from a genetically-inherited conscience, the obvious answer is that humans develop morality by interaction with other agents, through gradual socialization and parenting.

This is the analogy that Reinforcement Learning is built off of, and certainly it would make sense that an AGI should seek to optimize approval and satisfaction from its users, the same way that a child seeks approval from its parents. A paperclip maximizer, for example, would receive a stern lecture indicating that its creators are not angry,... (read 841 more words →)

Dissected boxed AI

This potato only generates 1.1 volts of electricity. I literally do not have the energy to lie to you - GLaDOS (Portal 2)

In reading about proposals about AI Boxing, one thought immediately comes to mind: Why would we want the risk of having an Unfriendly AI turned on in the first place? After all, inasmuch as a Boxed AI poses reduced risk, a collection of source code in non-executable text files poses no risk at all.

At first, I hesitated to write this post because it felt like a really dumb question. Surely, if we knew that the AI was unfriendly, then there would be no question to shut it down. And yet, the... (read 270 more words →)

-8

Replying toDo advancements in Decision Theory point towards moral absolutism?

Do advancements in Decision Theory point towards moral absolutism?

Isn't a deceptive agent the hallmark of unfriendly AI? In what scenarios does a dishonest agent reflect a good design?

Of course, I didn't mean to say that TDT always keeps its promises, just that it is capable of doing so in scenarios like Parfit's Hitchhiker, where CDT is incapable of doing so.

Do advancements in Decision Theory point towards moral absolutism?

From my understanding reading through the site, each successive decision theory builds off of previous models in an upward progression: CDT/EDT is surpassed by the "winning" paradigm of TDT, which in turn is surpassed by the updateless paradigm of UDT/FDT. And the goal that these decision theories move towards is one that enables an agent to "understand, agrees with, and deeply believe in human morality" (from Superintelligence FAQ). In other words, creating a mathematical model that correctly reproduces the same impetus of human morality.

With that in mind, I start to notice certain trends across the advancements in Decision Theory that reflects what kind of ideology a truly moral AGI would use. CDT,... (read 942 more words →)

Replying toHow would two superintelligent AIs interact, if they are unaligned with each other?

How would two superintelligent AIs interact, if they are unaligned with each other?

Thanks, I'll be sure to check them out

Replying toAre ya winning, son?