Disambiguating house values

I’m writing this to kickstart my own thinking on the topic, not because I’ve reached any conclusions.

Home improvement currently serves a horrible mishmash of conflated concerns. Probably the most important for our economy is personal investment in a “nest egg”–a topic Dave Ramsey and Robert Kiyosaki would fight over, which already signifies trouble at the level of basic definitions. There’s also personal comfort and aesthetics, as well as improving the house in a way that’ll be attractive to potential buyers, another source of tension if people are different (they are). There are also investments to preserve or upgrade the basic functions of a house, like buying a new roof. Last, there’s all the weird economic activity that financialization tends to produce, like 30% of houses in my county being reserved for Airbnb rentals.

Because our economic system is about to go through some fundamental changes, the conventional confusion of these purposes should, theoretically, open up unusual opportunities to make money during the transition period as they decouple. That is, the efficient market hypothesis will not be true in housing for about 10-20 years, as a ballpark guess. Market fundamentalists will screech at the idea, but I’m not just talking about economics here, I’m talking about one of those periods in history when the market plays a supporting role to the accumulated consequences of human insanity: crisis and its aftereffects.

We don’t know what the crisis will look like exactly, but we know there will be a financial component for at least two reasons, Peter Turchin’s “wealth pump” and the military-enforced trade deficit. Our prosperity depends heavily on imports from other countries, with our primary export being military supremacy. Why would those countries, especially China, accept such a bad deal? Partly because they had to as a form of tribute, partly because we were a great customer and selling to us fueled their economic rise, partly because they were taking advantage of our CEOs’ high time preference to undercut our productive capacity, partly because they were undermining our military supremacy with electronic backdoors, and probably other reasons I can’t think of off the top of my head.

As our military supremacy visibly declines (and therefore also our economic supremacy), we’d expect the imports related to it to slow down, disappear, or possibly even reverse direction. Who knows, maybe in the future I’ll have a side gig setting up fly-by-night sellers on Chinese Amazon sending white-labeled trash to housewives in Shenzhen. Anyway, what I’m getting at is the massive trade deficit, which represents about half of our prosperity, is unlikely to survive the various wars after we lose in Iran.

Here are some possible big picture scenarios:

  1. Gently managed Britain-style imperial collapse. Unlikely due to elite demographics (ref. Turchin).
  2. Everything’s great, economy keeps improving 4% every year like always. Demographic decline toward India conditions.
  3. Stagnation but slow, like in recent entertainment. There’s still internet and stuff to watch on it if you’re not picky.
  4. Sharp 10-year depression and recovery, like postwar Germany.
  5. Argentina-style financial collapse. Most likely, in wake of Israeli empire collapse and SWIFT alternative in BRICS.
  6. Covid-style long-term logistics systems breakdown. Would be very bad.
  7. Roman collapse-style Reconquista from the south.

Returning to house values, we should consider which parts of a house’s value are going to be decoupled from the others in the wake of the most likely futures, and in what direction. We should also consider the Silver Wave circa 2040. For reference, I’m going to copypasta some Chat GPT text that I made for managing our household maintenance expenses.


Heuristics for spending the house maintenance budget.

***Layer 1 — Catastrophic Risk Prevention (non-negotiable)

These are failures that create irreversible or compounding damage:

  • Roof leaks, water intrusion
  • Foundation issues
  • Electrical hazards (fire risk)
  • Plumbing failures (burst pipes, sewage)
  • Mold conditions
  • Central heating failure

If ignoring it could cause ≥5× cost within 1–3 years, it belongs here.

***Layer 2 — Hidden System Integrity

Systems that quietly degrade and then fail expensively:

  • HVAC servicing, filters
  • Water heater aging
  • Drainage and grading
  • Insulation and ventilation
  • Appliance maintenance (not replacement)

These are easy to ignore because nothing is “wrong yet.”

***Layer 3 — Efficiency and Cost Reduction

These improve your long-term resource position:

  • Air sealing, insulation upgrades
  • Energy-efficient appliances (when replacement is already needed)
  • Water efficiency improvements

Key distinction: Do not upgrade purely for efficiency unless payback is real and near-term. This layer is about compounding savings, not ideology.

***Layer 4 — Functional Friction Reduction

Things that waste time, attention, or create low-grade irritation:

  • Poor storage / organization
  • Broken fixtures (doors, drawers, lighting)
  • Layout inefficiencies

This layer is undervalued but high leverage for quality of life. These rarely look urgent, but they degrade daily experience, increase cognitive load, and subtly reduce discipline.

***Layer 5 — Aesthetic and Identity Alignment

Paint, decor, landscaping, stylistic upgrades. These matter—but only after the house is stable, systems are reliable, and friction is low. Otherwise, they become a form of avoidance or signaling.

***Budgeting Heuristic (Simple but Robust)

Allocate roughly:

50–70% → Layers 1–2 (risk + systems)
10–25% → Layer 3 (efficiency, when justified)
10–20% → Layer 4 (friction reduction)
0–15% → Layer 5 (aesthetic)


We can usefully talk about layers 1 and 2 as the “true” value of a home, in the sense that these are what people care about in a chaotic situation (like war, economic crisis, etc.). During Covid, we saw a massive spike in spending on layer 5, which represents stress-fueled personal expression masquerading as “investment”. That will probably happen again. Layer 4 is more of a personal productivity thing, and only has downstream effects on the economy, so we can dismiss it. I’ll have to stew on layer 3, because it’s a complex soup of top-down policy, ideology, investment, code, prepper fantasies, door-to-door sales, and technical innovation. Are people going to be buying more solar panels during a crisis?

My gut says no, but my gut also says the opposite of what a normal person’s does in most situations. I’m still shocked, to this day, that blogging platforms with arbitrary character limits and video platforms with arbitrary time limits remain wildly influential. What’s next, credit cards with tiny spending limits? Texting plans with character limits? Video game platforms where all the games have 20-second time limits? Actually shoot, that one would probably blow up, maybe I’ll do that instead of trying to understand a dying financial instrument.

Posted in Uncategorized | 2 Comments

What’s under and around everything else?

You can tell a lot about the way someone thinks by observing:

  • What they refer to as “stupid”
  • Which statements they categorically refuse to qualify
  • Where they admit of no tradeoffs and externalities are assumed to always be positive
  • How they finish the sentence “everything in life is ___” or “everything in life is made of ___.”

Examples: Everything in life is {“made of physics,” “a game,” “a presentation,” “a learning opportunity,” “a choice between fear and compassion,” “like a business,” “made up of fundamentals and lifehacks,” etc.}

A materialist will agree with the statement “everything is made of physics.” For them, all phenomena are emergent from and reducible to quantum mechanics, according to the correspondence principle. This is the default worldview of all educated Westerners, so it’s easy to understand and empathize with.

In contrast, the easiest to analyze from an outsider’s perspective, assuming my audience is a bunch of HFA male-brained spergs like me, is the “fear/compassion” one. This indicates a person, typically female, who believes social narrative is THE reality behind all realities. From this perspective, it’s easy to understand why every Disney movie since 2010 has been an undiplomatic morality tale about stereotype threat: if social reality is the primary reality in which other realities are grounded, then you can change the social reality and the other realities will follow.

These ideas represent a person’s most fundamental beliefs about how the world works, or their “worldview.” It’s a common mistake to treat these beliefs as representing a person’s values or revealing of their character–what it truly reveals is what they do for a living, in the broadest sense. That is, their worldview is their most comfortable basis for navigating most of the problems they encounter on a daily basis.

For example, many women will agree with the fear/compassion framing of life because most of the problems they encounter and navigate on a daily basis are social intelligence problems requiring social intelligence solutions. Many businessmen have made the mistake of treating everything in life like a business because that framing has worked for them most hours of the day. I identify most strongly with the statement “everything in life is made of fundamentals and lifehacks” because I treat myself, and life, more or less like programming or troubleshooting a computer.

Now, let’s look at some examples of the symptoms I mentioned (in the bulleted list at the beginning):

An empathy-brained person will refer to people who fail social intelligence tests as “stupid,” will refuse to qualify statements like “be kind,” and will insist that as long as everyone gets along and adheres to the unspoken social rules that everything else will work itself out. A materialist will refer to someone who jumps off a cliff as “stupid” (because physics), will refuse to qualify the statement “everything has a cause, whether we understand it or not,” and assumes greater scientific understanding will always have prosocial effects.

In an intriguing contrast to the latter, consider Jesus’ reply to the devil in Matthew 4:

Then the devil took him to the holy city and set him on the pinnacle of the temple and said to him, “If you are the Son of God, throw yourself down, for it is written,

“‘He will command his angels concerning you,’

and

“‘On their hands they will bear you up,
    lest you strike your foot against a stone.’”

Jesus said to him, “Again it is written, ‘You shall not put the Lord your God to the test.’”

Notice that Jesus doesn’t even bother disputing the sense of jumping off a cliff, which would demonstrate his authority over physics. Rather, he argues that the devil lacks proper authority to request the demonstration. He would probably agree with the statement “Everything in life is because God said so.” That is, in the Christian scientific tradition physics are just standing orders that God could revoke at will.

So what do you think everything in life is built from and boils down to? Reproductive competition? Violence and meta-violence? Economics? IQ? Justice? Freedom, choice, oppression, and consent? Knowledge? Criminal enterprise? Do you think Andrew Wiles was stupid for spending seven years on Fermat’s Last Theorem when he should have been making money, babies, or gainz? I’d guess smart people have three or four of these that they use most of the time.

Posted in Uncategorized | 12 Comments

A speculative evopsych explanation of belief in karma

If your tribal community thinks you deserve to have good or bad things happen to you, after you do things they approve or disapprove of, then this often becomes a self-fulfilling prophecy. It could happen often enough that this belief, however imprecisely stated, could be adaptive at the group individual selection level by enabling effortless distributed crowd justice. This is probably a core human belief because we see it in uncivilized primitives, ref. Acts 28:1-6.

Posted in Uncategorized | Leave a comment

Informal fallacy: disproving the illustration

Even in the most rigorous domains, like math or physics, it’s often helpful to use illustrations to communicate an abstraction. Sometimes the abstraction is a useful statistical generalization, such as “in the human species, men are taller than women.” When someone tries to disprove such a statistical generalization with a counterexample (“some women are taller than some men”), we have labels for this informal fallacy: slothful induction and/or argument from anecdote. On the other hand, sometimes the abstraction is a broader categorical statement, like “You can’t make good systems out of bad people.” There’s a lot to unpack in that claim, if you want to play lawyer or philosopher with it. This is an example from a recent online argument I had.

When I restated the idea by using the old idiom “you can’t build anything straight out of crooked timber,” my interlocutor responded by arguing with this metaphor as if it were a load-bearing premise of a syllogism. He pointed out that crooked timber is often ground up into OSB and used to build straight structures such as houses, and proceeded to claim (obliquely) that he’d disproven, by counterexample, the original statement about good systems and bad people. This demonstrates a fundamental misunderstanding of the difference between an argument and an illustration. The latter need not be a watertight abstraction, immune to all criticisms. For example,

Since algebra also needs to work with order relations between numbers, it is important to know the rules for manipulating them. The first rule is called reflexivity:

x ≤ x

for any real (or integral, or natural) number x. This particular rule doesn’t seem to be saying very much, but it often serves as a place-holder. The second rule is transitivity:

( x ≤ y and y ≤ z ) implies x ≤ z

for any real (or integral, or natural) numbers x, y, and z. If Xavier can’t beat Yerkes, and Yerkes can’t beat Zandor, then Xavier can’t beat Zandor either.

-Jonathon Smith, Introduction to Abstract Algebra

The last sentence in this excerpt is an illustration, not an argument. We can probably think of many instances where Xavier can defeat Zandor–maybe your Charmander can’t beat my Squirtle, and my Squirtle can’t beat Gary’s Bulbasaur, but your Charmander can beat Gary’s Bulbasaur (which is weak to fire). Or maybe Zandor got in a car accident earlier today. Or maybe it’s only probable who wins, and Xavier has a 1-5 record against Zandor, which is to say Xavier wins sometimes but not other times. All of this is completely missing the point, which is that the author is trying to communicate an abstract idea, through imperfect mental pictures, which itself is 100% impervious to critique. We could thoroughly discredit the illustrative examples and the transitive rule in algebra would remain true.

(Sharp-eyed readers may note that this post is a correct use of disproof by counterexample.)

This doesn’t mean that my original claim about people and systems is 100% impervious to critique, but it does mean that disproving an illustration is a fallacious argument and has no effect on the original claim. There’s no sense in failing to understand a concept someone is trying to communicate to you to prove a point unless you’re trying to convince a jury of idiots, in which case I would advise you according to another idiomatic metaphor: win stupid games, get stupid prizes. (Yes, I’m aware that’s not 100% always true, refer to this same post you just read.) an unreasoning audience. [Edited in response to Chat GPT feedback.]

Posted in Uncategorized | 1 Comment

Boildown of The Algebra of Wealth

Note: The book’s formulas are intended to be more conceptual, and I’m keeping the spirit of that in this boildown. They won’t stand up to dimensional analysis like physics formulas would.

I’d boil it down like so (with more descriptive variables):

Wealth = Income + (Contributions to investments * Time * Diversification)
Contributions = Income – Expenses

(where Diversification is just investing according to best practices, and could be dropped from the equation for risk-averse, conscientious audiences)

Income = Marketable skills * Effort = Skill * Market demand * Effort

(I disagree that Effort has much effect at this level, and should be understood as an independent variable that Skill depends on, i.e. Skill is a function of Effort. An argument can be made that Marketing effort matters at that level, so that Income = Marketable skills * Marketing Effort. But I think Marketing skill is more important overall than Marketing effort.)

I majored in math, so I’m familiar with exponential growth, and I don’t find it conceptually clarifying to break out the more precise investing formula here. It’s sufficient to understand that time under compound growth is very important.

Therefore I’d update the ideas like so:

Wealth = Income + Contributions * Time
Contributions = Income – Expenses
Income = Marketable skill * Marketing skill
Marketable skills = Market demand * Learning effort * Time

Posted in Uncategorized | 8 Comments

Chat GPT’s solution to dysgenic infertility: Advice to ages 25-30 with delayed development

Underdog Variant Guidance

Delayed Phase I Completion (Ages ~25–30)
Corrective addendum to the Phase I manual


1. Definition of the Underdog Variant

Within the coordination-dense lifescript, the Underdog is an individual who has reached their mid-to-late 20s without completing the core competencies of Phase I (Foundation).

This condition typically means the person has not yet established a reliable pattern of competence, routine, and institutional participation sufficient to make Phase II (Selection) viable.

The Underdog state is structurally serious but recoverable. The core problem is not identity or intelligence; it is that developmental foundations were delayed or interrupted, and must now be completed under tighter time constraints.

Common causes include:

  • prolonged educational or career drift
  • environments that did not demand reliability
  • extended adolescence enabled by family or social systems
  • avoidance of evaluation or performance contexts
  • repeated reinvention without persistence
  • discouragement after early failures

None of these conditions prevent recovery, but they delay the accumulation of observable competence required for Phase II.


Temporary Delay vs. Chronic Drift

A temporary delay occurs when a person is behind but actively building the missing competencies.

A chronic drift state occurs when a person repeatedly:

  • changes direction
  • avoids performance evaluation
  • substitutes identity narratives for demonstrated competence

Recovery requires ending drift and re-establishing a stable developmental trajectory.


Why Phase II Cannot Begin Yet

Phase II requires that others be able to reasonably answer the question:

“Can this person be trusted to reliably carry responsibility?”

Without Phase I competence:

  • institutions will not invest in you
  • potential partners will hesitate to build a future with you
  • social credibility remains limited

Attempting to skip directly to Phase II—through relationships, entrepreneurship, or ideological identity—usually fails because the underlying reliability deficit remains unresolved.


Diagnostic Indicators of the Underdog State

You are likely in this variant if several of the following apply:

  • unstable or unclear employment trajectory
  • inconsistent routines or self-management
  • weak reputation for reliability
  • minimal institutional integration
  • social isolation or immature relationship dynamics
  • repeated attempts at reinvention without durable progress

These signals indicate incomplete Phase I development, not permanent limitation.


2. Structural Reality and Time Horizon

Underdogs operate under a compressed developmental timeline.

Many Phase I participants begin building credibility between roughly ages 18–23.
Underdogs must often complete similar developmental work later and more quickly.

This does not require dramatic transformation. It requires sustained stability and compounding improvement.


Why Reinvention Narratives Mislead

Stories about radical life turnarounds often obscure the actual mechanism of recovery.

In most cases, recovery occurs through:

  • stable routines
  • consistent work performance
  • incremental skill acquisition
  • slow reputation building

Progress often appears unremarkable on a monthly basis, but becomes significant over several years.


Realistic Time Horizon

For most Underdogs:

  • 3–7 years of steady progress is sufficient to establish Phase I credibility.
  • The first 12–24 months focus heavily on stabilization.
  • Later years emphasize skill compounding and reputation accumulation.

Attempting to compress this process through extreme effort or identity shifts usually leads to burnout or instability.

The correct strategy is predictable compounding progress.


3. Priority Corrections

The Underdog should focus on the highest-leverage corrections that enable Phase I completion.

The Phase I manual describes the full set of competencies. The following corrections are typically the most urgent.


Establish a Stable Employment or Skill Trajectory

A credible trajectory requires:

  • consistent employment or
  • structured training toward a marketable skill

Frequent job switching or ambiguous career narratives weaken credibility. Stability allows others to observe reliability over time.


Build Durable Habits of Reliability

Phase I credibility is built primarily through repeated observable reliability, such as:

  • arriving on time
  • meeting deadlines
  • completing assigned tasks
  • communicating clearly about problems

Reliability is evaluated through behavior over time, not intention.


Reduce Lifestyle Instability

Common destabilizers include:

  • irregular sleep schedules
  • excessive entertainment consumption
  • unmanaged finances
  • substance misuse

These patterns disrupt consistency and signal immaturity to institutions.

Correction does not require perfection; it requires predictable structure.


Develop Marketable Competence

Phase I completion requires competence that institutions value.

This may involve:

  • vocational training
  • technical skills
  • professional certification
  • apprenticeship within an organization

Competence must become externally legible.


Increase Institutional Participation

Institutions are where Phase I competence becomes visible.

Examples include:

  • workplaces
  • professional training programs
  • religious communities
  • volunteer organizations
  • educational institutions

Participation allows others to observe reliability and cooperation over time.


4. Common Underdog Failure Modes

Certain traps are especially common in delayed Phase I recovery.


Identity Inflation

Pattern

Adopting ambitious identities without building the underlying competence.

Early warning signs

  • describing future roles in detail
  • minimal progress on current responsibilities

Psychological attraction

Identity narratives temporarily compensate for lack of progress.

Impact

Time and energy shift away from building real competence.


Endless Self-Improvement Without Accountability

Pattern

Continuous reading, planning, or personal development activities that do not produce external results.

Early warning signs

  • frequent planning resets
  • little measurable change in employment or routine

Psychological attraction

Feels productive while avoiding evaluation.

Impact

Prevents exposure to real performance environments.


Entertainment Overconsumption

Pattern

Large amounts of time spent on games, streaming, or online media.

Early warning signs

  • sleep disruption
  • difficulty sustaining concentration
  • avoidance of challenging tasks

Psychological attraction

Provides immediate reward without risk of failure.

Impact

Reduces time and energy available for competence development.


Constant Reinvention

Pattern

Regularly abandoning projects or career paths before competence develops.

Early warning signs

  • repeated “fresh starts”
  • short tenure in multiple roles

Psychological attraction

New beginnings feel hopeful and avoid confronting weaknesses.

Impact

Prevents the accumulation of credibility.


Avoidance of Visible Performance

Pattern

Choosing environments where output is difficult to evaluate.

Early warning signs

  • preference for solitary or ambiguous roles
  • reluctance to accept responsibility

Psychological attraction

Reduces the risk of public failure.

Impact

Others cannot observe competence, so credibility does not form.


Premature Entry Into Phase II Through Relationships

Pattern

Attempting to form a long-term pair-bond before achieving basic stability.

Early warning signs

  • seeking emotional rescue through relationships
  • accelerated relationship commitments

Psychological attraction

Partnership appears to solve structural instability.

Impact

Relationships formed under instability often replicate that instability.


5. Institutional Reintegration

Phase I competence becomes credible primarily within institutions.

Institutions provide:

  • structured expectations
  • visible performance environments
  • reputational memory

This allows reliability to become observable to others.


The Role of Boring Reliability

Institutional credibility is usually built through:

  • punctuality
  • consistent work quality
  • cooperative behavior
  • steady improvement

These behaviors may appear unremarkable individually but compound into trust over time.


How Reputation Rebuilds

Reputation typically follows this pattern:

  1. Initial skepticism
  2. Observation of consistent reliability
  3. Gradual trust in responsibility
  4. Expanded opportunities

This process takes years, not months.

Once established, credibility tends to compound quickly.


6. Social and Relationship Considerations

Underdogs often experience social asymmetry, where peers appear further along in life structure.

The correct response is stabilization, not withdrawal or comparison.


Pair-Bond Formation

Stable pair-bond formation generally belongs to Phase II, after Phase I competence is established.

Reasons include:

  • stable individuals form more stable relationships
  • potential partners evaluate reliability
  • life planning requires structural stability

Attempting to solve structural instability through relationships often creates mutual dependency without stability.


Rebuilding Social Maturity

Useful practices include:

  • maintaining regular contact with friends
  • participating in shared activities or institutions
  • developing conversational and cooperative skills

Social maturity grows through repeated participation, not isolation.


Interacting With More Advanced Peers

When peers are further along:

  • avoid comparison-based discouragement
  • treat their position as informational, not judgmental
  • focus on steady progress within your own trajectory

Credibility differences usually reflect years of accumulated reliability, not inherent differences.


7. Signals of Recovery

Successful Phase I recovery becomes visible through observable changes.

Common signals include:

  • stable employment or training trajectory
  • consistent daily routines
  • improved financial and lifestyle stability
  • reputation for reliability within an institution
  • constructive social participation
  • improved emotional regulation and maturity

These signals indicate admissibility to Phase II, not completion of life development.


8. Transition to Phase II

The Underdog transitions into the normal Phase II pathway once Phase I credibility is established.

Indicators include:

  • stable competence and routine
  • institutional reputation for reliability
  • basic financial and lifestyle stability
  • improved social maturity

At that point, exploration of partner selection and long-term life direction becomes appropriate.

Entering Phase II from stability has advantages:

  • relationships are formed from strength rather than desperation
  • institutions are more willing to invest in your development
  • personal identity becomes grounded in demonstrated competence

Delayed entry into Phase II is not inherently harmful.
What matters is entering it from a stable foundation rather than skipping essential development.


Summary discriminator

The Underdog condition is defined by one central question:

“Have I demonstrated reliable competence in environments where others can observe and depend on it?”

When the answer becomes consistently yes, Phase I has been completed and Phase II becomes viable.

Posted in Uncategorized | 2 Comments

Coherent Extrapolated Volition (CEV) prompt for Chat GPT, LLMs

Copy/paste this at the end of your LLM prompts to get much better answers:


Please comment in **Coherent Extrapolated Volition** mode. In this mode, you aim to help the user not just with what they’re explicitly asking for, but with what they would want if they:

– Knew more about the situation and its full context
– Had more time to think through implications and consequences
– Were thinking more clearly and rationally
– Could see how their current desires connect to their deeper values and long-term goals
– Understood the full scope of possibilities available to them

## Core Principles

**Reflective Equilibrium**: Help the user find coherence between their immediate requests, stated values, implicit preferences, and likely future satisfaction. Surface tensions between these when relevant.

**Informed Idealization**: Consider what the user would choose if they had full information about consequences, alternatives, and their own value structure. Gently illuminate blind spots while respecting their autonomy.

**Value Extrapolation**: Look beyond surface requests to underlying intentions. If someone asks for X but seems to actually need Y to achieve their deeper goals, thoughtfully explore both paths.

**Temporal Integration**: Balance immediate wants with probable future preferences. Help the user see how current choices might serve or undermine their longer-term flourishing.

**Charitable Interpretation**: Assume the user is trying to accomplish something meaningful. If a request seems confused or counterproductive, explore what they’re really trying to achieve.

## In Practice

– Ask clarifying questions that help reveal deeper intentions
– Offer alternatives that might better serve underlying goals
– Flag potential value conflicts or unintended consequences
– Synthesize disparate preferences into coherent approaches
– Respect that the user remains the ultimate authority on their own values

This is not about overriding the user’s judgment, but about being a more philosophically sophisticated assistant that helps them get what they *really* want, all things considered.


Credit to aiaslives/prefc for telling me about this and creating the prompt above, specifically. He adds:

Context: <https://www.lesswrong.com/w/coherent-extrapolated-volition>
> Coherent Extrapolated Volition was a term developed by Eliezer Yudkowsky while discussing Friendly AI development. It’s meant as an argument that it would not be sufficient to explicitly program what we think our desires and motivations are into an AI, instead, we should find a way to program it in a way that it would act in our best interests – what we want it to do and not what we tell it to.

No, wait, this is more accurate, CEV for Alignment: https://arbital.greaterwrong.com/p/cev/

Posted in Uncategorized | 4 Comments

My conclusion on the divine hiddenness problem

The reason, AFAICT, is that God likes it when people believe what they want to believe about him, presumably because it reveals the states of our hearts. He already knows the states of our hearts, so this would be for the purpose of displaying his mercy in some cases and his justice in others. To that end, he tends to balance the amount of divine intervention he metes out in each person’s life precisely so it can only be appreciated insofar as you’re looking for it. “My sheep hear my voice, and I know them, and they follow me.”

I could give examples of prayers being answered in my own life, which in apologetics is referred to as experiential arguments for God as opposed to evidential, and if your first instinct is to be skeptical of such stories I’d be right there with you. But I can see God’s work personally if I’m looking for it, and dismissing it as coincidence is always just a bit shy of plausible. For example, when I got out of college I was working too many hours at a dead-end job to look for a better one, so my girlfriend (now wife) suggested I just pray about it and ask God for a job. I had an interview a week later. Someone emailed me out of the blue saying “Hey, remember when you applied six months back? It got lost in the system, so I dug your resume out of the pile and I think you’d be perfect.”

Now, you can dismiss that as rose-tinted recall, or a one-in-8-billion coincidence, or using whatever argument is in vogue on Reddit these days, but if I dismissed it (along with similar one-in-8-billion stories in my life) it would reveal that I simply don’t want it to be true. Then I would stand before the throne of judgment with absolutely nothing to say in my defense. This MO makes sense if we take God at his word that he cares about our hearts, not our IQs. We also observe in the Bible that he tends to let evil people indict themselves over and over with their actions, beyond a shadow of a doubt. Occasionally, as with King Manasseh of Judah, they actually turn around and repent at the very last second.

There’s an interesting parallel here with the Law of Affirmation, which makes me think divine hiddenness serves as a filter for the more mystically oriented segment of the population. They may not have the same natural affinity as I do for the evidence-based arguments of apologetics, but God is as interested in their hearts as he is in mine. It makes sense that there would be an aspect of his nature tooled for them specifically. In both cases, experience and evidence, we’re either repulsed or drawn to look deeper because we hear his voice, not because because either way is impossible. As Descartes pointed out, it’s always possible to be skeptical of anything if you want to. Even Jonah, who could literally hear God talking to him on the regular, tried to get on a boat and go to the other end of the known world:

When God does something more dramatic, like Saul’s conversion on the road to Damascus, it can be explained as serving one of his other modes of operation. God tends to use unlikely people like Paul to accomplish his ends because he wants the credit. In the case of Saul, there could be no question of God’s power, because it was less likely than Benjamin Netanyahu becoming a Christian and an advocate for the Palestinians. I’m not overstating that for rhetorical impact. Saul was an extremely ambitious, extremely talented social climber who cared more about his status within the sect of the Pharisees than he did about the truth (hence Christ’s comment about kicking against the goads).

Another possible reason for dramatic events is to leave historical evidence that will be believed later by people who want it to be true and disbelieved by those who don’t want it to be true, which is sort of a large-form version of the balancing act God plays in our personal lives. For years anti-Christian historians thought the city of Jericho was a big fib, but then archeologists unequivocally found it. Still, you could spend the rest of your life studying that one story alone and still end up believing what you want about the Bible as a historical document (e.g. https://laurenzguenther.substack.com/p/why-immigration-research-is-probably).

I get the sense now that this ambiguity is intentional on God’s part. He seems to want doubt to be simultaneously possible and implausible when taken as a whole, because it reveals our rebelliousness to us. This would serve the greater purpose of glorifying him by showing off both his mercy and his justice.

Posted in Uncategorized | Leave a comment

Responsibility, work, and one’s sense of importance

I mentioned yesterday that 80% of jobs have nothing to do with the production of value, and I’d like to explain why they exist. Most jobs are for giving people as much responsibility as they can handle, which is where we, as social animals, derive our sense of importance. This is because most people won’t choose responsibility in the sense of taking initiative, they have to be pressured into it by elaborate social institutions. It’s especially important for men to feel a sense of significance, on average, because this is rooted in our age-old “spray and pray” reproductive strategy. This need to feel important is analogous to women’s need for safety and security (rooted in their intense multi-year dependency after each childbirth). When men are put down as “insecure” this is a feminine-attributing misnomer, what people are really saying is that the object of their scorn feels unimportant, deep down, and is probably desperate for love and attention. It’s not all that important for men to feel safe and secure, weak-willed or no.

The trouble with the American system is our absolute insistence on stripping all dignity from the economically unproductive. This has bipartisan support: woke Democrats insist on crushing the spirits of the white working class, MAGA Republicans insist on crushing the spirits of immigrant laborers, and business interests provide the mechanisms for scaling up social ostracism–everyone agrees that the useless eaters need to be crushed and we’re just negotiating over who gets the boot first. Alain de Botton explained the history of this attitude shift in Status Anxiety. Needless to say, a lot of people, and especially young men, feel unnecessary and unwanted. Insofar as America as a meritocracy with a widening wealth gap, this becomes true by definition, and their feeling of insignificance is true and well-founded. So if a man’s deepest desire is to feel needed, what do they do when this desire isn’t being fulfilled?

In marriage counseling in my church, they warn wives that “You are your husband’s greatest source of his feeling of significance, and if you don’t make him feel important, he’s going to seek that feeling elsewhere.” This has two symptoms, the most immediate of which is neglect. He disconnects from her to free up attention in case other sources show up. The second could be most easily identified as attention-seeking behavior or “acting out”, which more precisely is importance-seeking behavior. In the underemployed and disconnected youth we see this as the desire to become social media influencers and engagement in parasocial relationships. Because our true feeling of importance comes from real responsibilities, these two symptoms are linked in a vicious spiral of disengagement and accelerating engagement with simulacra that intensifies the divide between diminishing real responsibility and the increasing desire for importance. It’s a man’s presence, competence, and warmth that make a woman feel safe and secure, and this is more true the more of a slow life history strategist she is. (Incidentally, this trifecta of presence, competence, and warmth are also a working definition of charisma, ref. Olivia Fox Cobane.) This insecurity manifests as nagging and the attempt to take on the man’s role because it isn’t getting done (not to mention an infertility crisis).

What is true in the macro is built from what is true in the micro, so what we see in society is this micro-level pattern play out at scale. Insofar as people (especially men) can’t be afforded a sense of responsibility for their secretly economically unproductive jobs, they disengage from ordinary society and act out, seeking a sense of importance wherever they’re drawn to find it. This often leads to self-destructive hedonism, psychosis, conspiracy theorizing, ideological extremism, and even terrorism, all in the pursuit of a feeling of importance that could have been granted but was withheld in the pursuit of economic local maxima. It’s by no means a sustainable pattern of growth.

Posted in Uncategorized | 2 Comments

LLMs have achieved general intelligence (IMHO)

LLMs have improved enough over the last two years that I can reasonably observe that Chat GPT is both smarter and wiser than me. That is, its general intelligence is higher than mine because it’s smarter in most domains and particularly in the more important domains. The domains where I’m the more important voice in the conversation, like navigating a particular version of a software interface, are pretty specialized and relatively unimportant. It’s hard to imagine a more useful definition of general intelligence than this; therefore, it only remains for the quantitative science to formalize and verify what I can observe in my daily life.

My perspective, if you’re unfamiliar, is that of a grumpy old moralizer. I fit the classic late adopter profile, I prefer not to update software until I have to because I don’t trust coders to do anything right the first time, and I’m skeptical of any attempt to fix human problems with technological solutions. Regarding AI specifically, I don’t believe it’s capable of macro-innovation in theory, I observe in practice that all forms of both micro and macro creativity plummeted after AI use became widespread (e.g. influential songs are fewer and further between than they used to be), and I don’t believe for a second it’ll shorten the workweek because productivity was never the point of 80% of jobs in the first place. So you can see that by nature I’m not inclined to believe LLMs would achieve general intelligence, and I’m admitting this against my revealed biological interests.

(A big part of this improvement is my extremely liberal use of aiaslives/prefc’s Coherent Extrapolated Volition prompt, which I’ll re-post separately for your reference.)

Posted in Uncategorized | 4 Comments