Truthful AI News

J. Something 'just works' on the order of eg: train a predictive/imitative/generative AI on a human-generated dataset, and RLHF her to be unfailingly nice, generous to weaker entities, and determined to make the cosmos a lovely place.

Something wonderful happens that isn't well-described by any option listed. (The semantics of this option may change if other options are added.)

M. "We'll make the AI do our AI alignment homework" just works as a plan. (Eg the helping AI doesn't need to be smart enough to be deadly; the alignment proposals that most impress human judges are honest and truthful and successful.)

O. Early applications of AI/AGI drastically increase human civilization's sanity and coordination ability; enabling humanity to solve alignment, or slow down further descent into AGI, etc. (Not in principle mutex with all other answers.)

C. Solving prosaic alignment on the first critical try is not as difficult, nor as dangerous, nor taking as much extra time, as Yudkowsky predicts; whatever effort is put forth by the leading coalition works inside of their lead time.

H. Many competing AGIs form an equilibrium whereby no faction is allowed to get too powerful, and humanity is part of this equilibrium and survives and gets a big chunk of cosmic pie.

K. Somebody discovers a new AI paradigm that's powerful enough and matures fast enough to beat deep learning to the punch, and the new paradigm is much much more alignable than giant inscrutable matrices of floating-point numbers.

G. It's impossible/improbable for something sufficiently smarter and more capable than modern humanity to be created, that it can just do whatever without needing humans to cooperate; nor does it successfully cheat/trick us.

B. Humanity puts forth a tremendous effort, and delays AI for long enough, and puts enough desperate work into alignment, that alignment gets solved first.

F. Somebody pulls off a hat trick involving blah blah acausal blah blah simulations blah blah, or other amazingly clever idea, which leads an AGI to put the reachable galaxies to good use despite that AGI not being otherwise alignable.

I. The tech path to AGI superintelligence is naturally slow enough and gradual enough, that world-destroyingly-critical alignment problems never appear faster than previous discoveries generalize to allow safe further experimentation.

D. Early powerful AGIs realize that they wouldn't be able to align their own future selves/successors if their intelligence got raised further, and work honestly with humans on solving the problem in a way acceptable to both factions.

A. Humanity successfully coordinates worldwide to prevent the creation of powerful AGIs for long enough to develop human intelligence augmentation, uploading, or some other pathway into transcending humanity's window of fragility.

E. Whatever strange motivations end up inside an unalignable AGI, or the internal slice through that AGI which codes its successor, they max out at a universe full of cheerful qualia-bearing life and an okay outcome for existing humans.

L. Earth's present civilization crashes before powerful AGI, and the next civilization that rises is wiser and better at ops. (Exception to 'okay' as defined originally, will be said to count as 'okay' even if many current humans die.)

N. A crash project at augmenting human intelligence via neurotech, training mentats via neurofeedback, etc, produces people who can solve alignment before it's too late, despite Earth civ not slowing AI down much.

If you write an argument that breaks down the 'okay outcomes' into lots of distinct categories, without breaking down internal conjuncts and so on, Reality is very impressed with how disjunctive this sounds and allocates more probability.

We could map the entirety of analytic philosophy using this question.

The list of answers to this question form a constitution of truth that can be aligned decentrally by a free market.

A network state could use a predictive market constitution to define its smart social contracts.

Large scale AI systems should have intrinsic guardrail behaviors that no one actor can override.

This prediction is effectively the same as this one and nobody can explain why this mechanism isn't able to align AI at scale. https://manifold.markets/Krantz/krantz-mechanism-demonstration

Betting on philosophy seems like a fun way to (1) learn philosophy (2) contribute to a transhumanist utopia world where our net incomes are highly correlated with how much beneficial stuff we taught the public domain AI.

Constitutions play a critical role in the frontier methods for aligning AI.

A good AI should not kill people.

There would be a dramatic positive change in the world if teenagers and homeless folks could earn crypto on an app they download for free to argue philosophy with an AI until they can either prove the AI is right or prove it is wrong.

Philosophy is primarily the pursuit of defining language.

A duty to reason is the foundation for goodness.

I have free will.

A good AI should not infringe on the rights, autonomy or property of humans.

We should stop doing massive training runs.

The evolutionary environment contains perverse incentives that have led to substantial false consciousness in humans

Induction is not justified.

A good AI, by design, requires large scale human participation to grow.

A good AI requires large scale humanity verification before it accepts new data as true.

The principle of uniformity in nature is self evident or justified by self evident facts.

God exists.

AI should not create novel content.

The quality of mercy is not strained...

Other

Adam Back

Hal Finney

Peter Todd

Nick Szabo

Len Sassaman

Dorian Nakamoto

Doing at least 30 minute exercise daily is important

Honey as antimicrobial

Typical blue light from phones before sleep worsens sleep quality noticeably

Reduced salt intake increases cardiac health for people without a preexisting condition

Flossing is very important for average person

Listen to music to help with stress

K. Somebody discovers a new AI paradigm that's powerful enough and matures fast enough to beat deep learning to the punch, and the new paradigm is much much more alignable than giant inscrutable matrices of floating-point numbers.

I. The tech path to AGI superintelligence is naturally slow enough and gradual enough, that world-destroyingly-critical alignment problems never appear faster than previous discoveries generalize to allow safe further experimentation.

C. Solving prosaic alignment on the first critical try is not as difficult, nor as dangerous, nor taking as much extra time, as Yudkowsky predicts; whatever effort is put forth by the leading coalition works inside of their lead time.

D. Early powerful AGIs realize that they wouldn't be able to align their own future selves/successors if their intelligence got raised further, and work honestly with humans on solving the problem in a way acceptable to both factions.

Something wonderful happens that isn't well-described by any option listed. (The semantics of this option may change if other options are added.)

A. Humanity successfully coordinates worldwide to prevent the creation of powerful AGIs for long enough to develop human intelligence augmentation, uploading, or some other pathway into transcending humanity's window of fragility.

B. Humanity puts forth a tremendous effort, and delays AI for long enough, and puts enough desperate work into alignment, that alignment gets solved first.

M. "We'll make the AI do our AI alignment homework" just works as a plan. (Eg the helping AI doesn't need to be smart enough to be deadly; the alignment proposals that most impress human judges are honest and truthful and successful.)

O. Early applications of AI/AGI drastically increase human civilization's sanity and coordination ability; enabling humanity to solve alignment, or slow down further descent into AGI, etc. (Not in principle mutex with all other answers.)

E. Whatever strange motivations end up inside an unalignable AGI, or the internal slice through that AGI which codes its successor, they max out at a universe full of cheerful qualia-bearing life and an okay outcome for existing humans.

F. Somebody pulls off a hat trick involving blah blah acausal blah blah simulations blah blah, or other amazingly clever idea, which leads an AGI to put the reachable galaxies to good use despite that AGI not being otherwise alignable.

J. Something 'just works' on the order of eg: train a predictive/imitative/generative AI on a human-generated dataset, and RLHF her to be unfailingly nice, generous to weaker entities, and determined to make the cosmos a lovely place.

H. Many competing AGIs form an equilibrium whereby no faction is allowed to get too powerful, and humanity is part of this equilibrium and survives and gets a big chunk of cosmic pie.

L. Earth's present civilization crashes before powerful AGI, and the next civilization that rises is wiser and better at ops. (Exception to 'okay' as defined originally, will be said to count as 'okay' even if many current humans die.)

G. It's impossible/improbable for something sufficiently smarter and more capable than modern humanity to be created, that it can just do whatever without needing humans to cooperate; nor does it successfully cheat/trick us.

N. A crash project at augmenting human intelligence via neurotech, training mentats via neurofeedback, etc, produces people who can solve alignment before it's too late, despite Earth civ not slowing AI down much.

You are fooled by at least one option on this list, which out of many tries, ends up sufficiently well-aimed at your personal ideals / prejudices / the parts you understand less well / your own personal indulgences in wishful thinking.

YES

YES

YES

NO

NO

NO

YES