Computers are tools for doing intellectual work. Robots are for most other work. Don’t assume that the solution to the AI-doom will be computing.

• Decentralize; diffuse power.
• Don't risk creating more competitors (capability gains in power seeking, autonomy, replication)
• Spread out; Zubrin
• Use the Free tools instead of the paid tools?
• Harden, prep, over-engineer.
• Strategic point: no one who believes they'll gain from the enormous upside of AI will resist the current path, or all will bolster it.
• humans are ladder kickers

https://www.reddit.com/r/samharris/comments/13bkrzl/comment/jjbq866/

Glad you posted this. I think you represent a large and growing share of humanity, for the obvious reasons you state.

tl;dr: There are different ways to think about your own, and our collective, situation.

Things that help me:

1. Think of how much better off you are than the vast majority of humanity, who either have no idea what's happening with AI, or know but have many fewer options than you do (think: 3rd world). But do NOT then start thinking about how to help those people. Put on your own mask first.
2. Keep firmly in mind how silly you're going to feel, 20 yrs from now, if Yudkowsky and Bostrom are wrong, and you've pissed away time and energy on hedonism. (Also, 33 is not too late for anything. "A person who is [my age] can't do X" is usually more an expression of fear than belief. Look for the myriad counterexamples to shake yourself loose.)
3. Keep firmly in mind that human success in a cold, dark, hostile universe, and then on a cut-throat planet of things ALMOST as smart as us, was very, very unlikely, which means we have some track record of succeeding against unfavorable odds. Then think of how long we've had nuclear weapons, and yet controlled them. Same point.
4. Note that it wasn't intelligence alone, but thumbs and inclusive genetic fitness and cooperation and competition and maybe the advantages of carbon molecules over silicon molecules in an Earth-like environment that lead to our success. But do NOT then start looking for holes in my argument there; it was not an argument, it was a claim of knowledge, and knowledge is never really certain.
5. Take heart that death by Yudkwoskian 'ignition of the atmosphere' because something something "you are made of atoms" will be a quick, relatively good death compared to, say, the deaths currently happening in Ukraine, or at the hands of clever psychos the world over who know exactly how to find victims, torture them, kill them, and get away with it, year after year, generation after generation, since primates began. But do NOT start reading about this subject. Own mask first.
6. Follow the people on Twitter who substantively argue with Yudkwosky. And listen to more of Stuart Russell, who is smarter than Yudkwosky (says I) and yet totally open to the AI x-risk possibilities, but in a way that is constructive and solutions-oriented, instead of Yudkwosky's condescending-but-here-to-save-the-world-even-though-it's-now-hopeless-based approach. Russell has an audience with the productive, capable people of the world; Yudkwosky not so much.
7. Play with computer programming (assuming you don't yet). It will show you how much power you, individually, have, if only you can get past the inhibitions blocking you from learning to use the latest human inventions. Think of every new AI tool as a new product at Home Depot. What does it do? Need that? Want that? Grab it! It's free (usually)! … However, if it's labeled "Boston Dynamics" and it starts walking toward you, walk away.
8. Keep careful track of how many times the sun rises, how often, and the causes of it. Try to extrapolate based on that, even though we all know it can't go on forever.
9. Listen to Dave Chappelle, or your favorite comedian. Not their thoughts on AI… their thoughts on anything. Laughter is involuntary and completely defeats misery, if temporarily. This is why they tell so many jokes in war zones, like Ukraine.

I'll stop, but the list goes on.

Engineering Retraice to find AI doomers and deliver to them tool knowledge and better thoughts.

• mapping past segments
• feedback first (make decisions focused on evidence, no matter how little)
• planning future segments around WDWD
• re-launch pile of segments

Refugees want to be immigrants.

• be an immigration attorney
• be a Home Depot guy
• be a Bill Maher

People want to get to the promised land, then build a house, then sit inside when it's raining and laugh with their friends.

Hardware store: Stay away from tradies, who know more than you. You annoy them, their annoyance discourages you.

<2023-12-31 Sun 08:40>

Tech-nology … it’s tech, the art and artifacts, and knowledge, the ideas. <2023-12-31 Sun 16:10>

09:03 Computer control land migrant:

• all major OSes
• editors
• shells
• C, Python, source (text) basis
• streetwise security (no keylogger, screenwatecher, e.g.)

17:18 Digital natives are actually digital native slaves to what was built to enslave them. The digit indigenous are the controllers before the conquest.

09:43 Bad on front, change machines to… Tech? Artifacts? Knowledge? Automata?

10:57 Is totalitarian control inevitable? We either have to see disaster before we democratically agree (too late, if AI or biotech), or elite smarts will foresee, maneuver into totalitarianism to prevent. Is it a dilemma between totalitarian control and technological doom?

13:01 Understanding the behavior of rsync makes me not want to add any new `creatures' to my operation. Is it like a farm? If run by one guy plus a family, you only want dogs, cows, sheep, pigs, but not useful frogs or made-friendly bacteria, because more complexity requires more consciousness? Without the consciousness, the complexity leads to strange undesirable defaults, basins of valueless attraction?

13:02 So, simplifying what one has introduced is necessary if no new conscious entities are available to help; though the complexity of pre-existing environment to which one has adapted is arbitrary and doesn't require simplification.

13:20 Protective Ideas: people don't change much; ideas do; tech is just ideas; as tech ideas change, new protective ideas need to be built so we get the upside and not the downside.

<2023-12-31 Sun 16:56>

The world has gone from rudimentary computers and internet, through desktop Windows, Apple and Google, to mobile iOS, Linux and ChatGPT. This is break-neck change.

By swimming against the current of computing, your have been trying to cope, to grasp some understanding of a tectonic change in how humans busy themselves.

Soon, your window will close. Is the USB stick website correct?

Think of the massive role of: advertising, drugs, entertainment, mass media.

Your value; another way to think about it: What computer problems can you now solve, for whom?

Immigration lawyer, tool guy, … but one who hasn’t helped himself. Except I wouldn’t be saying, ‘get a job like I did’, or ‘like I could’, I’d be saying… what?

The refugee-immigrant view is the dark, pessimist. What about the optimist view? The economic growth that tech has so far produced? The medicine? Cf. Goldstone on Sam 232, that cheaper stuff that matters less than middle-class necessities. The optimist view is Explorers and tourists.

<2024-01-01 Mon 11:15> A new world has been created in my lifetime.

There should be some way of understanding delusions that explains why a single delusion can lead one person to succeed and another person to fail. Shouldn’t all delusions be associated with failure?

Why are some delusions useful and good, while others are the opposite?

Some delusions are useful, and therefore good. Most are probably not.

A solopreneur is a kind of entrepreneur, which is a kind of motivated, hyperspecialized, competitor. Entrepreneurs have a tricky problem to solve: to dispel lethal delusions while leaving vital delusions intact. Lethal delusions will kill a business. But without vital delusions, the business will never live in the first place.

The analogy to natural selection: If we are well-deluded, we are ‘fit’ in the relevant environment, and will thrive.(NO_ITEM_DATA:HoffmanTheCaseAgainstReality2019)

You don’t choose delusions; delusions choose you. Like genes (memes)?(NO_ITEM_DATA:DawkinsTheSelfishGene2016AutoGenerated)

Nature is trying to do something with us. (Who knows what.) Part of Nature’s strategy is delusions, and it is in charge of which ones overtake us.

We can dispel delusions, once we have them.

The question is: Should we, in any given case?

Stay deluded about the right things, entrepreneur.

Delusion: "(b) psychology: a persistent false psychotic belief regarding the self or persons or objects outside the self that is maintained despite indisputable evidence to the contrary" (Merriam-Webster)

Vital delusion: a false belief necessary for success. E.g.: ultra-confidence, hyper-optimism, ’reality-distortion fields’.

Lethal delusion: a failure-causing false belief. E.g.: ‘the money will come’, ‘there’s plenty of time’, most thoughts that begin with ‘somehow’.

Anti-delusion: a method that dispels delusions. E.g.: brutal honesty, ruthless prioritization, savage intolerance of nonsense and waste. (But remember: these can destroy both lethal and vital delusions.)

Delusions are beliefs that you think are true, but aren’t. So, if you (really do) re-class a belief as a delusion, you stop believing it’s true. Then it’s no longer a delusion.

Entrepreneurs must hope they have all the vital delusions, and then root out all the deadly ones. In delusion hunting, the trick is knowing your targets.