Hooked News

NO

NO

YES

No model picker

Personalized to you, the user

Trained on synthetic data

Very large context window, or another way to give the model long-term memory

Faster speed(smaller model, architecture improvements, mixture of experts keeping most parameters unused during inference, etc.)

Designed to inference cheaper than GPT-4(price must be lower)

Exact-query of training data or knowledge databases(similar to Google RETRO)

Online learning of specialized agents(finetuning, PEFTs, etc.)

Operates autonomously with instructions from you

capable of online learning

New Neural Network Architecture

Seed option to create the question

(Near) real-time update of training data (data delay is within 6 months and is kept updated)

Audio Modality

Acts autonomously

Has a catchier name

The site is never marketed well enough outside of the rationalist-adjacent community, and there are not enough rationalist-adjacent people to sustain the site.

It is never able to get a critical mass of cis women.

The basic premise of the site is too complicated for people, who prefer to use simpler dating apps.

The concept of gambling on romance is unappealing to most people.

Too few people are interested in trading on random people's matches

It just never ends up taking off, without any particular flaw being the obvious cause

It is never able to get a critical mass of active profiles.

People don't think strangers are more qualified to choose matches for them than themselves

Not enough rationalussy

Other

extreme "everything is public" approach alienates userbase

Manifold as a whole shuts down without it ever having taken off

Unpopular combo of excessive transparency imposed on users / lack of transparency from staff

James pivots to something else

Everyone on earth finds love

people leave the site after finding a match

lack of moderation / lots of harassment make the platform unpleasant to be on

Stripe threatens to stop processing payments

(threat of) legal action

another dating app captures our target audience

It is never able to get a critical mass of good SREs.

It allows minors to join and gets bad PR or legal issues as a result

Messaging failure: The unique value proposition is clear, predictive market match making, however the deliverable does not appear to be clear. Manifold.love could be calibrated to deliver hook-ups or long-term relationships, not both.

Markets are not predictive due to difficulty measuring or defining outcomes, or due to low probability of outcomes

The administration team proves incapable of writing or enforcing rules that effectively keep members safe, and this causes some form of scandal.

All the good dates join @Mira 's cult and the remainder are left to suffer alone - wailing and gnashing their teeth as incels.

YES

2025-2030

2035-2039

2030-2034

2024

2040-2044

2045-2049

2050-2054

2055-2059

2060 onwards

Dan Heng

Herta

Qingque

Xueyi

Guinaifen

Luka

Lynx

March 7th

Arian

Asta

Hanya

Hook

Misha

Natasha

Pela

Sampo

Serval

Sushang

Yukong

Gallagher

Single person

NO