Who takes materials property and experimental data, and how

Name: Who takes materials property and experimental data, and how
Creator: Sell Material Data to AI
Published: 2026-10-28
License: https://sellmaterialdatatoai.com/terms

2 of the 6 organisations whose pages we read license or partner for materials data from the companies and labs that hold it, read 28 October 2026.

2 of 6 organisations Take data from the companies and labs that hold it of the organisations read from their own pages on 28 Oct 2026

4 of 6 Data it names: experimental or test data on materials the leading answer of 3

4 of 6 Its own words for the data: stated organisations whose pages state it, with the quote below

2 of 6 What a holder does first: stated organisations whose pages state it, with the quote below

7 Pages examined 19 named, every answer dated and hash-pinned

How it comes by the data

generates its own experimental or computed data3 organisations
licenses or partners for data from the companies and labs that hold it2 organisations
licenses a database it curates or owns1 organisation

6 organisations, each counted once; where a organisation's pages contradict each other it counts as "varies".

Counts

Row	Value
generates its own experimental or computed data	3 organisations
licenses or partners for data from the companies and labs that hold it	2 organisations
licenses a database it curates or owns	1 organisation

Data it names

experimental or test data on materials4 organisations
not stated1 organisation
crystal, molecular or structural data1 organisation

Each of 6 organisations answers once.

Counts

Row	Value
experimental or test data on materials	4 organisations
not stated	1 organisation
crystal, molecular or structural data	1 organisation

Its own words for the data

stated4 organisations
not stated2 organisations

Each of 6 organisations answers once.

Counts

Row	Value
stated	4 organisations
not stated	2 organisations

What a holder does first

not stated4 organisations
stated2 organisations

Each of 6 organisations answers once.

Counts

Row	Value
not stated	4 organisations
stated	2 organisations

How every organisation answered

One organisation to a row. Expand an answer for its source sentence, word for word, plus the page and the read date; "not stated" means that page does not cover the question.

Organisation	How it comes by the data	Data it names	Its own words for the data	What a holder does first
83 Sciences	licenses or partners for data from the companies and labs that hold it We train on that unpublished data through partnerships with labs across the country 83 Sciences company page Our ambition is to own or license the world’s experimental data and use it to build better models for science. Head of Data Partnerships job post	experimental or test data on materials The materials performance that matters to industry can only be measured in real experiments, but >85% of experimental research is never published. 83 Sciences company page Almost 90% of experimental data is discarded, losing over $100B in R&D value every year. Head of Data Partnerships job post	83 Sciences uses AI to mine discarded experimental data and drive scientific… 83 Sciences uses AI to mine discarded experimental data and drive scientific breakthroughs. 83 Sciences company page Our ambition is to own or license the world’s experimental data and use it to build better models for science. Head of Data Partnerships job post	From there, you'll expand into industry R&D and other major sources of proprietary… Not addressed in 83 Sciences company page From there, you'll expand into industry R&D and other major sources of proprietary experimental data. Head of Data Partnerships job post
CCDC	licenses a database it curates or owns Use the world's largest database of curated crystal structures to advance your research. CCDC home page	crystal, molecular or structural data Upload your data for inclusion in the Cambridge Structural Database or the Inorganic Crystal Structure Database. CCDC home page	Upload your data for inclusion in the Cambridge Structural Database or the Inorganic… Upload your data for inclusion in the Cambridge Structural Database or the Inorganic Crystal Structure Database. CCDC home page	Upload your data for inclusion in the Cambridge Structural Database or the Inorganic… Upload your data for inclusion in the Cambridge Structural Database or the Inorganic Crystal Structure Database. CCDC home page
CuspAI	licenses or partners for data from the companies and labs that hold it Through its AI Materials Foundry - a global network of over 45 partners, including NVIDIA and Meta - the company is accelerating breakthrough discoveries in semiconductors, energy, and climate. CuspAI home page	not stated Not addressed in CuspAI home page	not stated Not addressed in CuspAI home page	not stated Not addressed in CuspAI home page
Materials Data Foundry (NUS and University of Toronto)	generates its own experimental or computed data National University of Singapore and University of Toronto launch the Materials Data Foundry to fast-track discovery of functional materials with AI and robotics Materials Data Foundry launch release	experimental or test data on materials This data will power AI models to accelerate advances in electronics, energy and infrastructure . Materials Data Foundry launch release	This data will power AI models to accelerate advances in electronics, energy and… This data will power AI models to accelerate advances in electronics, energy and infrastructure . Materials Data Foundry launch release	not stated Not addressed in Materials Data Foundry launch release
Periodic Labs	generates its own experimental or computed data Periodic Labs is an AI research and deployment company creating models and autonomous labs to accelerate science. Periodic Labs home page	experimental or test data on materials They provide huge amounts of high-quality data (each experiment can produce GBs of data!) that exists nowhere else. Periodic Labs home page	They provide huge amounts of high-quality data (each experiment can produce GBs of data!)… They provide huge amounts of high-quality data (each experiment can produce GBs of data!) that exists nowhere else. Periodic Labs home page	not stated Not addressed in Periodic Labs home page
Radical AI	generates its own experimental or computed data Scientists input material performance goals, then AI-generated experiments go directly to the self-driving lab, which continues refining results until it meets those goals. Radical AI home page	experimental or test data on materials Scientists input material performance goals, then AI-generated experiments go directly to the self-driving lab Radical AI home page	not stated Not addressed in Radical AI home page	not stated Not addressed in Radical AI home page

The 28 sentences the answers come from

83 Sciences, 83 Sciences company page, read 2026-10-28, How it comes by the data: We train on that unpublished data through partnerships with labs across the country
83 Sciences, 83 Sciences company page, read 2026-10-28, Data it names: The materials performance that matters to industry can only be measured in real experiments, but >85% of experimental research is never published.
83 Sciences, 83 Sciences company page, read 2026-10-28, Its own words for the data: 83 Sciences uses AI to mine discarded experimental data and drive scientific breakthroughs.
83 Sciences, 83 Sciences company page, read 2026-10-28, What a holder does first: The page is silent here; it was read in full on 2026-10-28, 4,837 characters long.
83 Sciences, Head of Data Partnerships job post, read 2026-10-28, How it comes by the data: Our ambition is to own or license the world’s experimental data and use it to build better models for science.
83 Sciences, Head of Data Partnerships job post, read 2026-10-28, Data it names: Almost 90% of experimental data is discarded, losing over $100B in R&D value every year.
83 Sciences, Head of Data Partnerships job post, read 2026-10-28, Its own words for the data: Our ambition is to own or license the world’s experimental data and use it to build better models for science.
83 Sciences, Head of Data Partnerships job post, read 2026-10-28, What a holder does first: From there, you'll expand into industry R&D and other major sources of proprietary experimental data.
CCDC, CCDC home page, read 2026-10-28, How it comes by the data: Use the world's largest database of curated crystal structures to advance your research.
CCDC, CCDC home page, read 2026-10-28, Data it names: Upload your data for inclusion in the Cambridge Structural Database or the Inorganic Crystal Structure Database.
CCDC, CCDC home page, read 2026-10-28, Its own words for the data: Upload your data for inclusion in the Cambridge Structural Database or the Inorganic Crystal Structure Database.
CCDC, CCDC home page, read 2026-10-28, What a holder does first: Upload your data for inclusion in the Cambridge Structural Database or the Inorganic Crystal Structure Database.
CuspAI, CuspAI home page, read 2026-10-28, How it comes by the data: Through its AI Materials Foundry - a global network of over 45 partners, including NVIDIA and Meta - the company is accelerating breakthrough discoveries in semiconductors, energy, and climate.
CuspAI, CuspAI home page, read 2026-10-28, Data it names: The page is silent here; it was read in full on 2026-10-28, 2,773 characters long.
CuspAI, CuspAI home page, read 2026-10-28, Its own words for the data: The page is silent here; it was read in full on 2026-10-28, 2,773 characters long.
CuspAI, CuspAI home page, read 2026-10-28, What a holder does first: The page is silent here; it was read in full on 2026-10-28, 2,773 characters long.
Materials Data Foundry (NUS and University of Toronto), Materials Data Foundry launch release, read 2026-10-28, How it comes by the data: National University of Singapore and University of Toronto launch the Materials Data Foundry to fast-track discovery of functional materials with AI and robotics
Materials Data Foundry (NUS and University of Toronto), Materials Data Foundry launch release, read 2026-10-28, Data it names: This data will power AI models to accelerate advances in electronics, energy and infrastructure .
Materials Data Foundry (NUS and University of Toronto), Materials Data Foundry launch release, read 2026-10-28, Its own words for the data: This data will power AI models to accelerate advances in electronics, energy and infrastructure .
Materials Data Foundry (NUS and University of Toronto), Materials Data Foundry launch release, read 2026-10-28, What a holder does first: The page is silent here; it was read in full on 2026-10-28, 10,004 characters long.
Periodic Labs, Periodic Labs home page, read 2026-10-28, How it comes by the data: Periodic Labs is an AI research and deployment company creating models and autonomous labs to accelerate science.
Periodic Labs, Periodic Labs home page, read 2026-10-28, Data it names: They provide huge amounts of high-quality data (each experiment can produce GBs of data!) that exists nowhere else.
Periodic Labs, Periodic Labs home page, read 2026-10-28, Its own words for the data: They provide huge amounts of high-quality data (each experiment can produce GBs of data!) that exists nowhere else.
Periodic Labs, Periodic Labs home page, read 2026-10-28, What a holder does first: The page is silent here; it was read in full on 2026-10-28, 4,196 characters long.
Radical AI, Radical AI home page, read 2026-10-28, How it comes by the data: Scientists input material performance goals, then AI-generated experiments go directly to the self-driving lab, which continues refining results until it meets those goals.
Radical AI, Radical AI home page, read 2026-10-28, Data it names: Scientists input material performance goals, then AI-generated experiments go directly to the self-driving lab
Radical AI, Radical AI home page, read 2026-10-28, Its own words for the data: The page is silent here; it was read in full on 2026-10-28, 1,907 characters long.
Radical AI, Radical AI home page, read 2026-10-28, What a holder does first: The page is silent here; it was read in full on 2026-10-28, 1,907 characters long.

Read and excluded

12 named documents were read and left out because they are not what this board covers; each reason is below. They are counted, never answered.

CuspAI, CuspAI launches global AI Materials Foundry with NVIDIA, Meta and 45 founding partners – Intelligent CIO Europe: the charter admits it for: a trade report naming CuspAI's Foundry data partners (name discovery only, never CuspAI's own terms)
Citrine Informatics, Chemicals & Materials Development Platform | Citrine Informatics: a 1,715-character product home page about the customer's own data on its platform, not a route for selling it
MaterialsZone, AI-Powered Materials Informatics | Accelerate R&D and Innovation: a product home page for the customer's own R&D data on its platform, not a route for selling it
Opendatabay, Licensed AI Training Data & LLM Fine-Tuning Datasets | Opendatabay: a general data buyer, venue or publisher that names no materials data, read on sellscientificdatatoai.com's buyer board; this board keeps to organisations whose pages name materials data
AWS Data Exchange, Data Marketplace - AWS Data Exchange - AWS: a general data buyer, venue or publisher that names no materials data, read on sellscientificdatatoai.com's buyer board; this board keeps to organisations whose pages name materials data
Protege, For Data Providers - Protege: the stored read is the cookie consent banner; the page's own text is not in it
Nyne.ai, Data partnerships | Nyne.ai: a general data buyer, venue or publisher that names no materials data, read on sellscientificdatatoai.com's buyer board; this board keeps to organisations whose pages name materials data
Troveo, Turn Your Company Data Into Revenue | Troveo: a general data buyer, venue or publisher that names no materials data, read on sellscientificdatatoai.com's buyer board; this board keeps to organisations whose pages name materials data
Redpine, Redpine | Licensed data for AI agents and model training: a general data buyer, venue or publisher that names no materials data, read on sellscientificdatatoai.com's buyer board; this board keeps to organisations whose pages name materials data
Springer Nature, Data Licensing for Research & AI | Powering R&D with 150 Years of Expertise | Springer Nature: a general data buyer, venue or publisher that names no materials data, read on sellscientificdatatoai.com's buyer board; this board keeps to organisations whose pages name materials data
Springer Nature, Licensing Options Data Solutions & Intelligence | Powering R&D with 150 Years of Expertise | Springer Nature: a general data buyer, venue or publisher that names no materials data, read on sellscientificdatatoai.com's buyer board; this board keeps to organisations whose pages name materials data
Lila Sciences, LILA | Scientific Superintelligence: a general data buyer, venue or publisher that names no materials data, read on sellscientificdatatoai.com's buyer board; this board keeps to organisations whose pages name materials data

How this was built

Each organisation's own partner, data provider, licensing and company pages were read through estate-fetch and stored whole (sha256 on every row).

For every page, four questions were answered from its own words: how it comes by the data, data it names, its own words for the data, what a holder does first.

Every stated answer carries the sentence it rests on, checked against the stored text; 'not stated' means the page read does not address the question.

A organisation with several pages reads 'varies' where they disagree.

6 organisations, 7 pages.

Where we looked (19)

83 Sciences, 83 Sciences: AI-native materials discovery powered by unpublished experimental data | Y Combinator (cited for reporting; the publisher's own partner, data provider, licensing and company pages)
83 Sciences, Head of Data Partnerships at 83 Sciences | Y Combinator (cited for reporting; the publisher's own partner, data provider, licensing and company pages)
CuspAI, CuspAI | AI-powered materials discovery (cited for reporting; the publisher's own partner, data provider, licensing and company pages)
CuspAI, CuspAI launches global AI Materials Foundry with NVIDIA, Meta and 45 founding partners – Intelligent CIO Europe (read, set aside)
Materials Data Foundry (NUS and University of Toronto), materialsdatafoundry.pdf (cited for reporting; the publisher's own partner, data provider, licensing and company pages)
Citrine Informatics, Chemicals & Materials Development Platform | Citrine Informatics (read, set aside)
Periodic Labs, Periodic Labs (cited for reporting; the publisher's own partner, data provider, licensing and company pages)
Radical AI, Radical AI (cited for reporting; the publisher's own partner, data provider, licensing and company pages)
CCDC, Structural Chemistry Data, Software, and Insights | CCDC (cited for reporting; the publisher's own partner, data provider, licensing and company pages)
MaterialsZone, AI-Powered Materials Informatics | Accelerate R&D and Innovation (read, set aside)
Opendatabay, Licensed AI Training Data & LLM Fine-Tuning Datasets | Opendatabay (read, set aside)
AWS Data Exchange, Data Marketplace - AWS Data Exchange - AWS (read, set aside)
Protege, For Data Providers - Protege (read, set aside)
Nyne.ai, Data partnerships | Nyne.ai (read, set aside)
Troveo, Turn Your Company Data Into Revenue | Troveo (read, set aside)
Redpine, Redpine | Licensed data for AI agents and model training (read, set aside)
Springer Nature, Data Licensing for Research & AI | Powering R&D with 150 Years of Expertise | Springer Nature (read, set aside)
Springer Nature, Licensing Options Data Solutions & Intelligence | Powering R&D with 150 Years of Expertise | Springer Nature (read, set aside)
Lila Sciences, LILA | Scientific Superintelligence (read, set aside)

Take this figure with you

2 of the 6 organisations whose pages we read license or partner for materials data from the companies and labs that hold it, read 28 October 2026.

How it comes by the data, by organisation: generates its own experimental or computed data 3 of 6, licenses or partners for data from the companies and labs that hold it 2 of 6, licenses a database it curates or owns 1 of 6 (6 organisations, read 28 October 2026).

Data it names, by organisation: experimental or test data on materials 4 of 6, not stated 1 of 6, crystal, molecular or structural data 1 of 6 (6 organisations, read 28 October 2026).

Its own words for the data, by organisation: stated 4 of 6, not stated 2 of 6 (6 organisations, read 28 October 2026).

What a holder does first, by organisation: not stated 4 of 6, stated 2 of 6 (6 organisations, read 28 October 2026).

To cite: "Who takes materials property and experimental data, and how", updated 2026-10-28, https://sellmaterialdatatoai.com/datasets/material-data-buyers/.

Copy the embed (static HTML)

2 of the 6 organisations whose pages we read license or partner for materials data from the companies and labs that hold it, read 28 October 2026.

Who takes materials property and experimental data, and how, sellmaterialdatatoai.com

<figure style="margin:0;padding:16px;border:1px solid #d0d0d0;background:#ffffff"><blockquote style="margin:0 0 8px;background:#ffffff">2 of the 6 organisations whose pages we read license or partner for materials data from the companies and labs that hold it, read 28 October 2026.</blockquote><figcaption style="background:#ffffff"><a href="https://sellmaterialdatatoai.com/datasets/material-data-buyers/" style="background:#ffffff">Who takes materials property and experimental data, and how, sellmaterialdatatoai.com</a></figcaption></figure>

Also as data: JSON, CSV, Markdown.

The data as a CSV (28 answers, compiled 2026-10-28)