Other good effects: nobody expects to get cancer, so I guess it doesn’t happen?
Things happen exactly as in reality except the existence of expectation applies a base-rate multiplier. So there would be more disease because e.g. hypochondriacs would be more likely than normal to contract disease.
There are many ways conflicting expectations could resolve, I’m not sure which would make the most sense. It could be a flat vote or have magnitude proportional to the amount of observations.
If it was advantageous to use structures of those inside cells for reactions somehow, then some organisms would already do that.
Not necessarily. The space of advantageous biologically possible structural configurations seems to me to be intuitively larger than the space of useful configurations currently known to be in use.
In order for a structure to be evolutionarily feasible, it must not only be advantageous but also there must be a path of individually beneficial (or at minimum not harmful) small steps in between it and currently existing structures. If an adaptation does not lend itself to linearly realized benefit, e.g. one that works really well but only when 90%+ is ‘correct’, it has no evolutionary way to piece itself together from 0-90%.
Like, for example, it makes sense that a future LLM would be able to explain a mathematical concept that has been documented and previously discussed but I just can’t see it solving existing frontier problems in mathematical theory, as it’s a completely different “skillset”.
Most non-mathmatician humans such as myself are arguably in the same boat for this specific example. I certainly wouldn’t know how to begin to work on frontier mathematical theory, but despite this if I were an AI I would fit many definitions of an AGI, albeit a lowly human-level one.
It would be useful to have a distinction between ‘routine’ logical reasoning (being able to apply established knowledge dynamically) and novel logical reasoning (being able to synthesize new theories/ideas from existing ones).
LLMs can arguably do the first but not the second, however the line dividing the two can be hazy. Many “novel advances” are not ‘new’ as much as existing approaches from different domains being recombined to solve a different type of problem than they had previously been used for. Is this routine or novel? Could an LLM do this? Particularly if the problem were phased to be tabooed/abstractified to avoid using leading questions/keywords that would encourage the LLM to regurgitate the common approaches?
An impressive thing about LLMs is how well they can work with arbitrary problems, so long as the problems are not too far out of the training distribution.
An example:
I asked ChatGPT 4o:
”A widget = 14
A phro = −3
A midge = 5
An attached wrok multiplies value by 0.5
What is the value of: A widget with an attached wrok plus three phros and two midges, and three wroks by themselves?”
Solving this sort of (simple to a human) word problem requires combining the domains of basic math and conceptual relations between objects (i.e. <object> can have <relation> to other <object> which interacts with the properties of that thing) and also the ability to make assumptions from missing information (no explicit value was assigned to solitary wroks)
GPT answer:
”To find the value of the given expression, let’s break it down step by step.
1. **Value of a widget with an attached wrok:**
- A widget is worth 14.
- An attached wrok multiplies the value by 0.5.
- So, the value of a widget with an attached wrok is .
2. **Value of three phros:**
- A phro is worth −3.
- So, three phros are .
3. **Value of two midges:**
- A midge is worth 5.
- So, two midges are .
4. **Value of three wroks by themselves:**
- Since a wrok by itself is not attached to anything and we don’t have a specified standalone value for a wrok, we can assume it doesn’t add any value by itself. So, three wroks by themselves are worth .
Now, we need to add all these values together:
First, let’s handle the addition of 7 and −9:
Then, add the result to 10:
And finally, add the result to 0:
So, the value of a widget with an attached wrok plus three phros and two midges, and three wroks by themselves is .”
Is this logical reasoning? Depends on how you define it. It’s certainly useful. How much of the thinking we humans do is ‘mere’ pattern matching with variable inputs? I suspect a large amount.
Haha makes sense. I wasn’t sure what the demographic distribution was likely to be.
You’re right, “objectively” doesn’t fit as well in that statement as I thought.
That is how I intended ‘convincing’ to be interpreted.
For almost every category of X, you’ll be judged hard for your preferences, even if you didn’t consciously choose any of them.
It depends on if X is a demographic/group or a variable. “I don’t want to date people who are [uneducated/from a drastically different cultural background]” sounds a lot less politically correct than “I want to date people with whom I estimate a high probability of mutual relationship satisfaction.” because you don’t have to explain your criteria to everyone.
I admit that’s more semantic obfuscation of judgement risk markers than it is mitigating the problem.
The rule “90% of everything is garbage” applies, but recent moral values are rejecting any sorts of hierarchies, even between functional and dysfunctional countries, cultures, cities, religions, values, etc.
When society suppresses attempts to evaluate concepts or situations as objectively better or worse than alternatives, is it any surprise that polarization increases?
If there are no commonly agreed upon benchmarks to calibrate against it becomes a war of whoever can shout loudest/most convincingly.
I think a significant contributing factor that makes ‘simple’ questions in some contexts prohibitively difficult to answer is the lack of True Availability of the information being requested.
In this case, I’m defining True Availability[1] as the requested content being already prepared and organized into the correct format and grouped together, needing no further processing other than finding it. Conditional Availability would be when you know how to obtain the information, but it requires some degree of processing and filtering to be ready for consumption.
In computer science, this is similar to a lookup table. Lookup tables typically contain a collection of pre-calculated results for common computations, because looking up a result in a table is generally faster than calculating it from scratch.
Anything you have in a LT is Truly Available, whereas anything you have to calculate is Conditionally Available.
In your example of freedom of information requests, the questions are hard to answer because they are only available on the condition that someone filters the requested information from everything else and then prepares it into a usable format for releasing.
If I was tasked with refining the information availability of a large organization, I would attempt to prepare publicly-releasable copies of everything that COULD legally be requested via freedom of information act and publish a public database of that. Let them knock themselves out. Individual request processing and answering sounds like a terribly inefficient method of sharing information.
There are probably legal/bureaucratic/practical difficulties to my proposed solution, but my point is merely that there are in some contexts systemic barriers making answers disproportionately expensive rather than answers being intrinsically more difficult in every case.
I suspect there already exist more conventional terms for the concepts I’m referring to, but I’m making do with what I already have Available.
If you want to be generally skilled at the type of challenges D&D Sci provides, putting some points into the data science and statistics proficiencies would be a good way to start.
In particular, some related skills:
SQL—Easy to pick up for someone with good technical skills. Challenging to master. Before going too deep on relational databases I also recommend learning good theory and practices behind it like the different design forms and why they’re important.
R programming language
Familiarization with various statistical analysis methods and what use cases they are intended for
As a software developer who works on object-level automation every day, I’m intimidated by the difficulty of attempting to definitively quantify ‘profit from automated tasks’ in a useful way.
For example, how do we define ‘automation’? “A task that formerly needed to be done by a human that now doesn’t need to be”? A printing press is automation by some interpretations of that insufficient definition.
Some changes in efficiency also have similar effects on productivity without being ‘automation’ (although much less scalable), for example a user that becomes highly proficient in the hotkeys of a complex platform may see massive improvements in their productivity, and subsequently eliminate jobs that would have been needed if they hadn’t become more productive.
I suspect if additional taxes were levied on ‘job automation’ it would merely create large incentives for companies to skirt around whatever the legal definition of automation was, and potentially hide it in things like the above example.
In the case where there was no ‘automation tax’ created, I would anticipate a NIT to be reasonable but not sustainable because I expect automation to continue to remove jobs at an accelerating rate in years to come. I do not expect tax revenue to increase at the same rate because my current understanding is that the most wealthy tend to also be those most proficient at exploiting loopholes in the tax system to evade as much as possible.
My takes here are almost entirely conjecture and I’d appreciate someone more informed to correct and/or clarify.
I’m in a similar situation. I have very little self control with sweets/candy if I have them available. I can far more easily stop myself from buying them in the first place.
If I allow myself to buy a bag of candy I’ve already lost and I will consume all of it in a matter of hours/days.
How much of the developed world’s economy is devoted to aesthetic personalization of products rather than accomplishing the essential functions of [product here]?
I am not saying aesthetics or personalization are ‘bad’, however I suspect that if the cost were quantified and demonstrated to people along with examples of more productive things that could be done with that money, many people might prefer forgoing some of our more wasteful things.
Example:
The cost of having thousands of different styles of sink faucet, instead of a small number of highly efficient and optimized faucet designs for distinct use cases [small household kitchen, large household kitchen, small form factor, high throughput restaurant]. These costs are created via the overhead caused by the redundant costs of engineering, design, manufacturing, and logistics.
These same factors apply more or less to every product where variations are sold primarily for aesthetic rather than functional purposes, particularly when they replace existing functional versions.
I believe the root cause of this inefficiency is our psychological tendency to overvalue ephemeral utility such as using possessions as social status tools rather than trying to optimize how we collectively use our limited economic output. For example, if a sizeable portion of the money in the market for functionally useless decorations were able to go towards medical research.
I do not know how a more efficient allocation of resources could be practically enacted. According to my understanding most attempts at centrally planned economies have even less success than the free market, as inefficient as it is.
If a large portion of people decided to prioritize their purchases better that would work, but that’s obviously a very challenging coordination problem.
There’s soft skills in “communicating to others without hurting them”, (i.e. “tact”)
What about the situation in which:
One has highly religious relatives who are somewhat less cognitively functional that oneself
You wish you could help them have a map more closely coupled to reality
You are confident that you have a good chance of convincing them of reality, but not that the knowledge would actually be a net gain for them to have, since:
They are so invested in their beliefs that the realization of falsehood might do irreparable psychological damage
Yes, but it thankfully for me only lasted a couple of hours and they didn’t start keeping track until near the end.
I had a very similar experience as a teenager after a mild concussion from falling on ice. According to my family, I would ‘reboot’ every few minutes and ask the same few questions exactly. It got burdensome enough that they put up a note on the inside of my bedroom door with something along the lines of:
”You are having amnesia”
“You hit your head and got a mild concussion”
”You’ve already been to the ER, they said you’re likely to be fine after a few hours and it is safe to sleep.”
The entire experience was (reportedly) very stressful to me due to disorientation.
Yes.
For example: The common saying, “Anything worth doing is worth doing [well/poorly]” needs more qualifiers. As it is, the opposite respective advice can often be just as useful. I.E. not very.
Better V1: “The cost/utility ratio of beneficial actions at minimum cost are often less favorable than they would be with greater investment.”
Better V2: “If an action is beneficial, a flawed attempt may be preferable to none at all.”
However, these are too wordy to be pithy and in pop culture transmission accuracy is generally sacrificed in favor of catchiness.
Yeah, many people, like the majority of users on this forum, have decided to not build AGI.
Not to build AGI yet.
Many of us would love to build it as soon as we can be confident we have a realistic and mature plan for alignment, but that’s a problem that’s so absurdly challenging that even if aliens landed tomorrow and handed us the “secret to Friendly AI”, we would have a hell of a time trying to validate that it actually was the real thing.
If one is faced with a math problem where you could be staring at the answer and know no way to unambiguously verify said answer, you are likely not capable of solving the problem until you somehow close the inferential distance separating you from understanding. Assuming the problem is solvable at all.
This is fantastic!
I’m not sure the best way conflicting expectations could resolve. It could be a flat vote or have magnitude proportional to the amount of observations...Or even based on relative emotional investment! What could possibly go wrong?