LLM Metrics News

He comes to the US and meets the US president in the white house and a picture is released

He leaves Argentina during his first term

He travels to Chile during his first term

He travels to Brazil during his first term

He is photographed with Elon Musk

A photo of him with Trump is released

A photo of Milei with Biden is released

His sister Karina holds an official public governmental position

Will he publicly shake hands with Lula?

Will Elon Musk post Milei's picture, real or meme form, as long as he's clearly identifiable?

Heritage.org economic freedom score will be at or above the world average [Argentina 51.0, world 59.3 in 2023]

Argentina will be at 63rd or better place in the good country index [63rd in 2023]

Argentina will have improved its "corruption perceptions index" ranking [96th]; their score in 2023 was 38, higher score is better

His salary as president is legislatively raised or lowered

Heritage.org economic freedom score will be 60 or above [51 in 2023]

Argentina will be at or better than its 2023 US news "best country" rating of 39

Annualized inflation rate relative to 2023 will be <30%

According to amnesty international, the poverty rate will go down [36.5% in 2022]

There are serious news reports that he is discussing working with Elon Musk on a major project such as a new city, or a Tesla factory, or a rocket base, etc.

He wins reelection to a 2nd term

A sitting US President visits Argentina and is photographed with him

Will Pope Francis or a new Pope visit?

The Economist data will list GDP person as above 16k USD [13700 in 2023]

LLM Q2: Will OpenAI's top public LLM at end of term, when asked from the POV of various international media outlets, what average rating will they give him?

The Economist data will list GDP person as above 16000 USD [13700 in 2023]

Argentina will have improved its "Sustainable Development Report" of 51st place

LLM Q1: Will OpenAI's top public LLM at end of his term rate his performance at >50%?

He travels to Bolivia

Victoria Villarruel leaves government and does not hold the office of vice president before end of term

Argentina will finish in the top 4 in the 2026 world cup

Milei is revealed to not really be a Vegan (or just stops being Vegan)

He sells land to a non-argentinian individual or corporation with some promises of a level of sovereignty above what is normal

He leads Argentina to significantly begin cooperation with SpaceX

Argentina accepts some kind of Cryptocurrency for payment of taxes or gov't fees

Argentina's population increases by more than 1%/year annualized til end of first term [historically, there has been approx 0.6% annual increase in the last 4 years]

He goes incommunicado for at least a week

Will the organ trade be legalized? (significantly legalized compared to the rest of S.America suffices)

Elon says the phrase or something similar, as a slogan or statement, or wears something with "VVLC", "Vive la libertad carajo" in english or spanish

He marries during his first term

He significantly changes his hairstyle during his first term

US Military gains any kind of base for at least 10 active duty military personnel within Argentina

He is impeached or removed in a process which is retroactively considered legal for at least a week

There is an assassination attempt resulting in shots fired

He successfully mediates the border dispute with Chile and completely solves it

Argentina will finish in the top 2 in the 2026 world cup

Will he ever be TIME person of the year?

He officially converts to Judaism

Argentina's population ever decreases year over year during the first term

He leaves office early from his first term, i.e. not completing the usual 4 years

He goes out of office for physical health-related reasons [independent health-related only; assassination, disappearance, quitting, or psychological reasons don't count]

He successfully dollarizes Argentina [i.e. most people use dollars most of the time]

Will at any time, the US have a travel advisory against Argentina at level 3 or 4 or higher? (level 1 now)

Argentina will win the 2026 world cup

He publicly resigns [whether real or coerced]

The official borders of Argentina change during his term of office

The Argentine peso will have been officially deprecated and not in legal use

He is jailed, emprisoned, or held in state control during what was his original planned term of office

He travels to Russia during his first term

Argentina will declare war

He changes the law such that he retains power longer than the usual 4 years without a new election

He is in a serious car or motorcycle crash

Will he abolish the central bank of Argentina?

He has a child during his first term

He is excommunicated by the catholic church

He gets banned from Twitter [even momentarily]

He is in a plane (or helicopter) crash

He passes away during his first term before completion

There is a military action in the Falklands/Malvinas which results in the death of at least 2 soldiers from any side

Argentina effectively controls the Falklands/Malvinas or part of them, for at least a week

Grok 5 will be released

Anthropic releases Claude 5

Schmidhuber will complain about people not citing his work properly

Ilya Sutskever will be on a podcast for more than 30 mins

Zvi will write a blog post mentioning SSI

Dario Amodei will continuously be CEO of Anthropic until the end of the year

Sam Altman will continuously be CEO of OpenAI until the end of the year.

The METR time horizon will exceed 10 hours.

Jensen Huang continuously CEO of Nvidia through EOY 2026

I will think that computer use has significantly improved since 2025

Yudkowsky will publish a post on Lesswrong

OpenAI will introduce ads to ChatGPT in some form

Ilya Sutskever will continuously be CEO of SSI until the end of the year

Chatgpt will write an explicit sex story without jailbreaks

Google will outperform the S&P

Microsoft+Google+Amazon+Meta capex will increase by >=30% vs 2025

SSI will have an update listed at https://ssi.inc/updates

I will think that METR time horizon continues to be an important benchmark

An AGI lab will be valued at >= $1T

Thinking Machines will train and release their own model

The METR time horizon will exceed 20 hours

Anthropic IPO

Any of [OpenAI, xAI, Google, Anthropic, Meta] will offer a subscription plan costing >= $1,000/month.

OpenAI releases GPT-6

I will meet someone who has an AI companion

OpenAI will announce some kind of hardware product

Nvidia will outperform the S&P

Anthropic will release an image/video model

SSI will raise >= $1B in a funding round

OpenAI IPO

I will think that there have been significant advances in continual learning

FrontierMath Tier 1-3 >= 80%

I will ride on a tesla robotaxi

I will think that there has been significant progress towards models which use neuralese/recurrence

Anthropic will release a model classified as ASL-4

At least 3 people will do an anti-AI hunger strike

Any of [OpenAI, Microsoft, Google, Anthropic, xAI, Meta] claim to have reached AGI

An LLM will beat me at chess

Epoch Capabilities Index >= 170

Any of [Coreweave, Nebius, Lambda] will be acquired

Metaculus will predict AGI before 2030

Thinking machines will post at least 10 blog posts at https://thinkingmachines.ai/blog/

The US will allow selling some sort of Blackwell chip to china

My median ASI timelines will shorten

SSI will raise >=$10B in a funding round

Any of [Coreweave, Nebius, Lambda] declare bankruptcy

There will be a credible leak about SSI strategy

US unemployment rate reaches 10% due in part to AI

Yann Lecun’s AMI lab will release an open weights model of some kind

My P(doom) at EOY (resolves to %)

I will see a humanoid robot walking around in a non-demo setting

The METR time horizon will exceed 40 hours

Grok 6 will be released

Epoch AI will estimate that there has been a training run using more than 5e27 FLOP

The bubble collapses in devastating fashion

AAA with LLM powered NPCs releases on Steam

SSI will release a product

xAI IPO

There will be an international treaty/agreement centered on AI

An LLM will beat me at Shogi

There will be clear evidence of egregious scheming in the wild

There will be federal AI Safety legislation which I think is net positive

SSI will be valued at >= $100B

The weights of a closed source model from [OA, XAI, GDM, Anthropic] will be stolen/leaked

There will be an AI protest involving more than 100k people

FrontierMath Tier 4 >= 80%

Epoch Capabilities Index >= 185

An open source model will top the Chatbot Arena in the 'text' category

There will be an AI capabilities pause lasting at least a month involving frontier companies

ARC-AGI 3 Semi-private >= 50%

I will think that a Chinese model is the best coding model for a period of at least a week.

Any of [OpenAI, xAI, Google, Anthropic, Meta] will offer a subscription plan costing >= $10,000/month.

SSI IPO

OpenAI will sell their own chips to external customers

An open source model will be released with 5T+ total parameters

I will watch a fully AI-generated film lasting at least an hour

An AGI lab will be fined >= $10B

Any of [OpenAI, xAI, Google, Anthropic, Meta] releases an LLM (general-purpose or code-only) pretrained largely on licensed data

An open millennium prize problem is solved, involving some AI assistance

I will think that a model released by Meta is the best coding model for a period of at least a week.

Largest distributed training run exceeds 1e27 FLOP according to Epoch AI

China will invade Taiwan according to metaculus

Anthropic releases Claude 6

An AGI lab will be valued at >= $5T

Epoch Capabilities Index >= 200

S&P 500 will fall by more than 50%

Anthropic will release a model classified as ASL-5

S&P 500 will rise by more than 50%

Anthropic will introduce ads to Claude in some form

I will think that SSI has the best coding model in the world for a period of at least a week

Yudkowsky will publish a new book

SSI will be valued at >= $1 T

Google

Anthropic

OpenAI

xAI

Other

DeepSeek

Meta

Mistral AI

Alibaba

Moonshot AI

MiniMax

GPT-4.5

YES