arXiv.org

[2511.02824] Kosmos: An AI Scientist for Autonomous Discovery

Kosmos: An AI Scientist for Autonomous Discovery

andrewwhite01

TechCrunch

Experts don’t think AI is ready to be a ‘co-scientist’

FutureHouse

About

Metaculus

Next Millenium Prize Problem Solved by AI

Announcing Edison Scientific

Endpoints News

Exclusive: Research institute FutureHouse debuts AI scientist and for-profit spinoff

SGRodriques

Elon Musk

Rishi Sunak

Kamala Harris

Bill Gates

Andrew Ng

Yoshua Bengio

Fei-Fei Li

Demis Hassabis

Melanie Mitchell

Sam Altman

Dario Amodei

Geoffrey Hinton

Yann LeCun

Gary Marcus

Max Tegmark

Connor Leahy

Eliezer Yudkowsky

Jaan Tallinn

Marc Andreessen

Eric Schmidt

Norbert Wiener

Arthur Clarke

Irving John Good

Claude Shannon

Hans Moravec

John Smart

There's a "wall of fear-mongering and doomerism" in the AI world right now.

"Worrying about AI today is like worrying about overpopulation on Mars."

Concerns that AI could pose a threat to humanity is "preposterously ridiculous."

Claiming AI poses an existential threat is "such an extreme" and risks "wip[ing] out some of its potential benefits."

"I'm more concerned about... the risks that are here and now [than the existential threat of AI]."

Current AI is "not anywhere close" to posing an existential threat but it could in the future.

AI is the "biggest existential threat" to humanity.

AGI's worst-case scenario would be "lights-out for all of us."

Powerful AI systems "taking control" pose an "existential threat."

AI has a "10 to 25 per cent" chance of destroying humanity.

"The most likely result of building a superhumanly smart AI... is that literally everyone on Earth will die."

Does AI Pose an Existential Threat to Humanity?

Stuart Russell

Shane Legg

Society is more likely to get "cat-level" or "dog-level" AI years before human-level AI.

"[W]e will eventually reach AGI... and quite possibly before the end of this century."

"Superhuman AIs" will be achieved "between a few years and a couple of decades."

"Digital intelligence" could overtake us within "5 to 20 years."

We may achieve "something like AGI or AGI-like in the next decade."

"[W]e could get to real AGI in the next decade."

"Human-level" AI could occur in "two or three years."

"I'd be surprised if we don't have AGI by [2029]."

Will AI Surpass Human Intelligence, and When?

Andrew White, the co-founder of FutureHouse and its head of science, announced on Wednesday that the organization had succeeded in creating an AI scientist capable of making “genuine discoveries,” after two years of work.

Andrew White

Andrew White, the co-founder of FutureHouse and its head of science, announced on Wednesday that the organization had succeeded in creating an AI scientist capable of making “genuine discoveries,” after two years of work.

FutureHouse is an American nonprofit founded in 2023 to create AI scientists to “automate and accelerate fundamental science that improves human health.” The organization claims these systems can accomplish in one day what human scientists would take 4.4 months to achieve.

FutureHouse is an American nonprofit founded in 2023 to create AI scientists to “automate and accelerate fundamental science that improves human health.” The organization claims these systems can accomplish in one day what human scientists would take 4.4 months to achieve.

In a paper released on Wednesday, FutureHouse introduced Kosmos, an AI scientist that automates data-driven discovery by performing iterative cycles of parallel data analysis, literature searches and hypothesis generation over extended periods of up to 12 hours per run.

In a paper released on Wednesday, FutureHouse introduced Kosmos, an AI scientist that automates data-driven discovery by performing iterative cycles of parallel data analysis, literature searches and hypothesis generation over extended periods of up to 12 hours per run.

The system uses a structured world model to coordinate information sharing between two specialized agents, enabling it to “coherently pursue” research objectives across an average of 200 agent rollouts, collectively executing 42,000 lines of code and reading 1,500 papers per run.

The system uses a structured world model to coordinate information sharing between two specialized agents, enabling it to “coherently pursue” research objectives across an average of 200 agent rollouts, collectively executing 42,000 lines of code and reading 1,500 papers per run.

FutureHouse claimed Kosmos made seven scientific discoveries spanning metabolomics, materials science, neuroscience and statistical genetics, with an average factual accuracy rate of 79.4%. This rate varied from 85.5% for data analysis statements to 57.9% for interpretation statements.

FutureHouse claimed Kosmos made seven scientific discoveries spanning metabolomics, materials science, neuroscience and statistical genetics, with an average factual accuracy rate of 79.4%. This rate varied from 85.5% for data analysis statements to 57.9% for interpretation statements.

The researchers, however, found several limitations with the system, such as Kosmos inventing “unorthodox quantitative metrics” that were “conceptually obscure and difficult to interpret,” as well as the system’s technical specifications, which are only capable of managing datasets of up to 5GB.

The researchers, however, found several limitations with the system, such as Kosmos inventing “unorthodox quantitative metrics” that were “conceptually obscure and difficult to interpret,” as well as the system’s technical specifications, which are only capable of managing datasets of up to 5GB.

arXiv.org #

TechCrunch #7122

FutureHouse #

Metaculus #

Endpoints News #

Researchers developed Kosmos, an AI scientist system that automates data-driven discovery by performing iterative cycles of parallel data analysis, literature search and hypothesis generation over extended periods of up to 12 hours per run.</span

Researchers developed Kosmos, an AI scientist system that automates data-driven discovery by performing iterative cycles of parallel data analysis, literature search and hypothesis generation over extended periods of up to 12 hours per run.

The system uses a structured world model to coordinate information sharing between specialized agents, enabling coherent pursuit of research objectives across an average of 200 agent rollouts that collectively execute 42,000 lines of code.</span

The system uses a structured world model to coordinate information sharing between specialized agents, enabling coherent pursuit of research objectives across an average of 200 agent rollouts that collectively execute 42,000 lines of code.

Independent scientist evaluators found 79.4% of statements in Kosmos reports to be accurate, with higher accuracy rates for data analysis statements at 85.5% compared to interpretation statements which achieved 57.9% accuracy.</span

Independent scientist evaluators found 79.4% of statements in Kosmos reports to be accurate, with higher accuracy rates for data analysis statements at 85.5% compared to interpretation statements which achieved 57.9% accuracy.

Academic collaborators estimated that a single 20-cycle Kosmos run performed work equivalent to approximately 6 months of human research time, with the number of valuable scientific findings scaling linearly with the number of cycles.</span

Academic collaborators estimated that a single 20-cycle Kosmos run performed work equivalent to approximately 6 months of human research time, with the number of valuable scientific findings scaling linearly with the number of cycles.

The researchers report seven validated discoveries made by Kosmos spanning multiple scientific fields including metabolomics, materials science, neuroscience and statistical genetics, with three reproducing unpublished findings.</span

The researchers report seven validated discoveries made by Kosmos spanning multiple scientific fields including metabolomics, materials science, neuroscience and statistical genetics, with three reproducing unpublished findings.

Each statement in Kosmos scientific reports includes direct citations to either data analysis code or primary literature sources, ensuring complete traceability of the system's reasoning and enabling independent verification of findings.</span

Each statement in Kosmos scientific reports includes direct citations to either data analysis code or primary literature sources, ensuring complete traceability of the system's reasoning and enabling independent verification of findings.

Google's AI co-scientist is pure hype without substance, delivering vague results that no legitimate scientist would take seriously. The tech giant has a track record of trumpeting supposed AI breakthroughs without providing reproducible methods, from falsely claiming better breast tumor detection than radiologists to promoting zero genuinely new materials despite bold claims. Scientists don't want AI automating hypothesis generation — the most intellectually fulfilling part of research — while leaving them stuck with tedious implementation work.

Google's AI co-scientist is pure hype without substance, delivering vague results that no legitimate scientist would take seriously. The tech giant has a track record of trumpeting supposed AI breakthroughs without providing reproducible methods, from falsely claiming better breast tumor detection than radiologists to promoting zero genuinely new materials despite bold claims. Scientists don't want AI automating hypothesis generation — the most intellectually fulfilling part of research — while leaving them stuck with tedious implementation work.</span

Kosmos represents a genuine breakthrough in AI-driven scientific discovery, delivering seven externally validated discoveries across multiple fields after two years of rigorous development. The system achieves 80% agreement with human analysis and operates coherently across millions of tokens by mapping discovery processes to world models. Unlike vaporware announcements, this technology is immediately available for educational users and backed by real collaborative research spanning materials science, biochemistry and biology.

Kosmos represents a genuine breakthrough in AI-driven scientific discovery, delivering seven externally validated discoveries across multiple fields after two years of rigorous development. The system achieves 80% agreement with human analysis and operates coherently across millions of tokens by mapping discovery processes to world models. Unlike vaporware announcements, this technology is immediately available for educational users and backed by real collaborative research spanning materials science, biochemistry and biology.</span

Artificial Intelligence

AI Scientist 'Kosmos' Automates Research, Makes 7 Discoveries

Headlines

Kosmos is a genuine breakthrough in AI-driven scientific discovery, having made seven externally validated discoveries across multiple fields following years of rigorous development. With its high factual accuracy, Kosmos could quickly advance scientific research and technological development, becoming transformative in the field.

Kosmos is a genuine breakthrough in AI-driven scientific discovery, having made seven externally validated discoveries across multiple fields following years of rigorous development. With its high factual accuracy, Kosmos could quickly advance scientific research and technological development, becoming transformative in the field.

So-called AI scientists are hype without substance, delivering vague results that no legitimate scientist would take seriously. These systems fundamentally lack the empirical rigor, transparency and contextual understanding necessary for credible research. By entrusting AI to spearhead scientific discovery, its supporters risk undermining the integrity of science itself.

So-called AI scientists are hype without substance, delivering vague results that no legitimate scientist would take seriously. These systems fundamentally lack the empirical rigor, transparency and contextual understanding necessary for credible research. By entrusting AI to spearhead scientific discovery, its supporters risk undermining the integrity of science itself.

There is a 40% chance that the next Millennium Prize Problem will be solved by AI, according to the Metaculus prediction community.

There is a 40% chance that the next Millennium Prize Problem will be solved by AI, according to the Metaculus prediction community.

AI Scientist 'Kosmos' Automates Research, Makes 7 Discoveries

AI Scientist 'Kosmos' Automates Research, Makes 7 Discoveries

The Spin

Metaculus Prediction

The Controversies

Andreessen

Ng

LeCun

Mitchell

Li

Bengio

Musk

Altman

Hinton

Amodei

Yudkowsky

LeCun

Marcus

Bengio

Hinton

Hassabis

Altman

Amodei

Musk

Go Deeper

Articles on this story

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters

AI Scientist 'Kosmos' Automates Research, Makes 7 Discoveries

AI Scientist 'Kosmos' Automates Research, Makes 7 Discoveries

The Spin

Metaculus Prediction

The Controversies

Andreessen

Ng

LeCun

Mitchell

Li

Bengio

Musk

Altman

Hinton

Amodei

Yudkowsky

LeCun

Marcus

Bengio

Hinton

Hassabis

Altman

Amodei

Musk

Go Deeper

Articles on this story

Sign Up for Our Free NewslettersSign Up for Our Free Newsletters

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters