Extra 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.

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.

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.

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.

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.)

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.

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.

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.

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.

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.

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.)

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.

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.

Trump will be the POTUS

FED rates will be below 5% on 1st January 2025

MrBeast Channel Subscribers will be between 300M to 350M on 1st Jan 2025

At Least 1 earthquake with magnitude 8.0 or higher

1 or more Jan 6 rioters get pardoned

Israel attacks Iranian nuclear facilities

Elon removed as doge head

Pope dies

Yoon Suk Yeol impeachment upheld by supreme court

Its revealed that Elon Musk had more children then 13 already known

Famous global brand goes through a rebranding

GPT 5 releases

Russia retains de facto control over Crimea

Avatar: Fire and Ash released

Hollow Knight Silk song released

King Charles will be alive for the whole year of 2025

This whole market gets more than 300 traders

2 or more hurricanes make landfall in the US

LEMMiNO releases a new video

Any incumbent world leader diagnosed with cancer after 28th January 2025

Biden gets hospitalised

FED rates will be below 4% on 31st December 2025

Yoon Suk Yeol convicted of insurrection

In one of the monthly polls, >= 80.0% of Manifold users who provide an opinion agree that weak AGI has been achieved

MKBHD marries his girlfriend Nikki Hair

After the release of GPT-5, every monthly Manifold poll results in a plurality of respondents agreeing that weak AGI has been achieved

Early general elections in Pakistan

At least one world leader (prime minister, president, or monarch) gets assassinated while in office

Alan's Conservative Countdown to AGI >= 97%

Ali Khamenei assassisnated

Attempt to assassinate trump in 2025

Department of Education ended by executive order (even if temporarily)

T1 three-peat as League of Legends world champions

Mike Johnson is not Speaker

Bitcoin reaches 150k sometimes during 2025

Ali Khamenei dies a natural death

In one of the monthly polls, >= 33.3% of Manifold users who provide an opinion agree that non-human intelligence has influenced Earth

Russia-Ukraine war will end in 2025

There will be less MAU at manifold in December 2025 then there would be in January 2025

Marco Rubio fired or "resigns" upon rumors of his impending termination

Californian independence gains enough signatures

Attempted assassination on Zelensky in which he gets hurt but survives

Dodgers win World Series

Los Ratones win against T1, or lose 2-3 (either case should resolve YES, if they don't play against each other this should be NO)

Elections in Bangladesh come to pass (currently planned for late 2025 to early 2026)

Saudi Arabia - Israel normalize relations

At least one Premier of a Canadian province proposes joining the USA

A country leaves NATO

Phillies win the World Series

Harry and Meghan divorce

2025 hotter then 2024

Putin dies

Biden dies

Taylor Swift will be "single" anytime during 2025

In one of the monthly polls, a plurality of Manifold users declares the achievement of weak superintelligence

Congress repeals any tariff

Apple will release first foldable smartphone during 2025

Chuck Schumer resigns

United States government acknowledges the existence of life anywhere other than on or close to Earth

Humanity's Last Exam passed with a grade of 80%

Elon Musk diagnosed with COVID-19

Bethesda releases a new game in the Elder Scroll series

China annexes Taiwan

Filibuster abolished

Zelensky Assassinated

Article 5 of NATO invoked

OpenAI o6 or any o6 variant released

Ukraine internal coup attempt

FrontierMath solved (>= 80%) and model is available to any person in the United States willing to pay for it

Kate and William divorce

Sam Altman gets fired again

At least one Premier of a Canadian province takes a concrete step, such as holding a referendum, to join the USA

Humanity's Last Exam passed with a grade of 90%

Putin assassinated

NATO at war with Russia

Trump assassinated

Any newspaper reports a US Default on any debt

US-China war

A country leaves the EU

A nuclear bomb explodes

2025 is the 15th "Year of Three Popes"

US debt ceiling will be scrapped during 2025

United States government acknowledges the existence of non-human intelligence on or close to Earth

One of the Millenium prize problems falls to a model

The winds of winter by GRR Martin will release

USA claims intelligences of extra-terrestrial origin have credibly visited or sent technology to a location within our solar system in an official government statement

Los Ratones go to LoL Worlds

LK99 replicates at last

Greenland Joins USA

End of the world by celery

Ukraine joins NATO

Humanity's Last Exam passed with a grade of 80% on or before June 30

A US state secedes from the union

51st US State

Peter Turkson elected Pope

GTA 6 releases

Federal Reserve cuts rates to zero

Harris will be the POTUS

Bitcoin will be > 100k on 1st January 2025

MrBeast Channel Subscribers will be below 300M on 1st Jan 2025

MrBeast Channel Subscribers will be above 350M on 1st Jan 2025

Jimmy Carter will be alive on 1st January 2025

Jimmy Carter will be alive on 31st December 2025

Humanity's Last Exam passed with a grade of 80% on or before March 31

Francis is always Pope

New pope takes name "Francis"

First papal conclave >= 10 ballots

NO

Make the bread taste good

Don't eat anything for at least 48 hours before eating the bread

Stretch-and-fold after mixing, 3x every 30 min

Create indentation, fill with melted cheese and butter

Bake on upside-down sheet pan, covered with Dutch oven

Resolve this option YES while eating the bread

Donate the bread to a food pantry, homeless person, or someone else in need

Use sourdough instead of yeast

Sprinkle 3 grams of flaky sea salt on top of each loaf before the second bake

Watch the video

Autolyse 20 minutes

3 iterations of stretch-and-fold, at any time during the 14h waiting period. Minimum wait time between iterations 1 hour

Make a poolish 12 h ahead: 100 g flour + 100 g water + 0.8 g yeast (0.1 %). After it ferments, use this poolish in place of 100 g flour and 100 g water in the final dough.

Bake it with your best friend.

Add 50g honey

Swap 200ml water for milk

Incorporate a whole grain flour (buckwheat for example)

More steam! Either spritz with more water (preferably hot) or actually pour some boiling water in just before closing the lid.

Bake for an amount of minutes equal to the percent this market answer is at when it comes time to begin baking. (Maintain the ±3 minute tolerances and the 2:1 ratio of time before:after the water spritz.)

Use King Arthur Bread Flour instead of All-Purpose

Decompose it into infinite spheres, then a few parts per sphere, rotate the spheres by arccos(1/3), unite them and you will find 2 chilis (Banach-Tarski)

Let dough rise on counter only until double volume or 2h max, any time longer in fridge

Add lots of butter (0.2 ml per gram)

Use 50% whole grain flour

Ditch current process, do everything the same as the video

Toast the bread

Eat the bread while punching @realDonaldTrump in the face

Eat the bread while watching your mana balance steadily tick to (M)0

Throw the bread at a telescope

Add 50g sugar

Put a baking rack in the Dutch oven before putting the loaf in, raising the loaf off the floor and lofting it over a layer of air.

Replace all water spritz steps with a basting of extra virgin olive oil.

Use flour made from an unconventional grain e.g. barley, millet, oats, rye, sorghum, maize etc.

Assume the chili is not in the interval [0,1], square it for more chili, if it is in (0,1), take the square root, else (equals 0 or 1) add 1 to it.

Assume the chili is in the interval (0,1), square it for less chili, if it is in (1,infinity) take the square root, if it is in (-infinity,0) take the negative of the square of the of the chile, else (equals 0 or 1) subtract 1 from it.

Get your friends to help you make a batch ten times the size, but add a Pepper X (2.7M Scoville heat units) to the mixture

Add 1tsp of diastatic malt powder per 3cps of flour

replace 10% of flour with farina bona

Bake the bread into a fun shape, like a fish, or an octagon

While the bread is baking, tip every user who voted "Yes" on this option 25 Mana

Add 50g vital wheat gluten

Give ChatGPT your current recipe as well your take on what optimal bread tastes like, then take that advice for your next bake

Bread flour, 3x yeast, cut rise to ~3h

Use whole wheat to improve the nutrition of the bread

Add an amount of MSG equivalent to half the current salt content

Place small ice cubes between parchment and pot instead of water

Cook the bread with a rod/puck of aluminum foil (or similar) in the core in an attempt to conduct heat through the center of the bread, cooking it evenly like a doughnut.

Make all of the ingredients from scratch.

Add a pinch of sugar

Make the bread edible then throw it in

Buy bread from a michelin star restaurant.

Increase water by 50 g

Drink vodka while eating the bread

Cover bread with damp paper towel instead of initial water spritz. Rehydrate paper towel during 2nd spritz. Remove paper towel before placing on cooling rack.

Do FOLDED

Quit Manifold into the bread.

Kill the bread into Manifold.

Improve the bread

Start at 500F, drop to 450F and uncover half way through

Grind/powderize all salt used into a fine powder (with pestle & mortar or similar device)

it needs more salt

Add 1/2 cup yogurt to the bread and name the bread “gurt” while addressing it with “yo, gurt”.

Half yeast

Ship a piece of the bread to a random person.

Encourage people to participate in the market in good faith while making the bread

Add 2g? of baking soda

Let dough sit 48 hrs

Resolve this option NO while eating the bread

put butter into it

Mix half sodium/potassium chloride

Add a tablespoon of sugar

Bake for 5 fewer minutes

Bake one more minute

Mail the bread to 1600 Pennsylvania Ave. Washington D.C.

Use tap water instead of fancy RO water

Frost it and put sprinkles on it to make it a birthday cake.

Add sawdust to increase the volume of the bread (but only like 10% sawdust by volume max. maybe 20% if it's good sawdust)

Add as many Jack Daniel's whiskey barrel smoking chips as feasible to the Dutch oven before baking, physically separating them from the bread as necessary while baking.

Eat the bread while sending all your mana to @realDonaldTrump

Bake the Manifold Crane into the Bread

Don't eat anything for at least 24 hours before eating the bread

Quadruple salt

Do all the changes in the top 5 open options by probability, excluding this option

Have someone sell the bread to you at an expensive price

Use lemonade instead of water.

Bake one fewer minute

Bake the cake while wearing a onesie.

Bake vegimite into it.

Bake for 5 more minutes

Replace salt with sugar

Eat the bread in front of the White House.

Bake vodka into it

Implement all options that resolved NO

Make the bread inedible then throw it out.

Replace flour with flowers

Throw the bread at @realDonaldTrump

Force Feed it to @realDonaldTrump

Make the bread great again

Cut the bread into the number of traders in the market slices.

Make naan bread, an easy-to-make bread

Only buy ingredients from 7/11.

Implementing every element listed below.

Put a non-lethal dose of any rat poison.

Just make donuts instead

Bake it in an easy bake kids oven

Think positive thoughts before tasting

Use a plastic baking sheet.

Eat the bread while betting yes on Cuomo on Manifold

Ditch all the steps. Just buy the bread from the supermarket

Double oven temperature

Halve oven temperature

Play classical music while baking

Light it on fire with birthday candles.

Bake it with a microwave

Eat the bread while betting yes on Mamdani on Manifold

Wear a suit while baking the cake.

Bake your social security number into it.

Bring it to Yemen and put a bomb in it

Bake America Great Again

Sacrifice a lamb

Add MAGA and a splash of Trump juice

Bake in a cat and a dog

Explode it:

Take a fat dump in the dough

Sit in dough 24 hrs

Let dough sit 24 hrs

Bake in rectangular tin

double yeast

halve salt

Double salt

Add 2tsp olive oil

Refrigerate dough instead of room temp wait

Do not mix salt and yeast in water together

Put fork in microwave

Don't eat anything for at least 12 hours before eating the bread

Add 2tbsp vanilla extract

Eat the bread with friends

Bake it in the country you were born in.

Eat the bread over the course of a week.

Brakes hit

Capability limit in AI

Huge alignment effort

New AI paradigm

Slow and gradual capability gains in AI

Enhancing human minds and/or society

Major capability limit in AI

Non-AI tech

Alignment relatively easy

Brakes not hit

Alignment unnecessary

Well-behaved AI with bad ends

YES

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.

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.

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.)

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.

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.

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.)

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.

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.

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.

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.

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.

NO

Prediction markets

Signalling

Ems

Criminal law vouching

Tax Career Agents

Healthcare

Reversible agents

Other

Extra-terrestrial civilizations (Grabby aliens, Great Filter...)

NO

YES