# Who takes materials property and experimental data, and how

> 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.

Sell Material Data to AI · https://sellmaterialdatatoai.com/datasets/material-data-buyers/ · computed 2026-10-28

**Take data from the companies and labs that hold it:** 2 of 6 organisations (7 pages read; 19 named).

## The answers, question by question

- **How it comes by the data:** generates its own experimental or computed data 3, licenses or partners for data from the companies and labs that hold it 2, licenses a database it curates or owns 1
- **Data it names:** experimental or test data on materials 4, crystal, molecular or structural data 1, not stated 1
- **Its own words for the data:** stated 4, not stated 2
- **What a holder does first:** not stated 4, stated 2

## Every organisation

| 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 | experimental or test data on materials | 83 Sciences uses AI to mine discarded experimental data and drive scientific… | From there, you'll expand into industry R&D and other major sources of proprietary… |
| CCDC | licenses a database it curates or owns | crystal, molecular or structural data | 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… |
| CuspAI | licenses or partners for data from the companies and labs that hold it | not stated | not stated | not stated |
| Materials Data Foundry (NUS and University of Toronto) | generates its own experimental or computed data | experimental or test data on materials | This data will power AI models to accelerate advances in electronics, energy and… | not stated |
| Periodic Labs | generates its own experimental or computed data | experimental or test data on materials | They provide huge amounts of high-quality data (each experiment can produce GBs of data!)… | not stated |
| Radical AI | generates its own experimental or computed data | experimental or test data on materials | not stated | not stated |

## The sentence each answer rests on (28)

- 83 Sciences, [83 Sciences company page](https://www.ycombinator.com/companies/83-sciences), 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](https://www.ycombinator.com/companies/83-sciences), 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](https://www.ycombinator.com/companies/83-sciences), 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](https://www.ycombinator.com/companies/83-sciences), 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](https://www.ycombinator.com/companies/83-sciences/jobs/NR8uCdC-head-of-data-partnerships), 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](https://www.ycombinator.com/companies/83-sciences/jobs/NR8uCdC-head-of-data-partnerships), 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](https://www.ycombinator.com/companies/83-sciences/jobs/NR8uCdC-head-of-data-partnerships), 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](https://www.ycombinator.com/companies/83-sciences/jobs/NR8uCdC-head-of-data-partnerships), 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](https://www.ccdc.cam.ac.uk/), 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](https://www.ccdc.cam.ac.uk/), 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](https://www.ccdc.cam.ac.uk/), 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](https://www.ccdc.cam.ac.uk/), 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](https://www.cusp.ai/), 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](https://www.cusp.ai/), 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](https://www.cusp.ai/), 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](https://www.cusp.ai/), 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](https://content.presspage.com/uploads/2580/6609582b-6231-47c2-9a06-ef76d471fcd6/materialsdatafoundry.pdf), 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](https://content.presspage.com/uploads/2580/6609582b-6231-47c2-9a06-ef76d471fcd6/materialsdatafoundry.pdf), 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](https://content.presspage.com/uploads/2580/6609582b-6231-47c2-9a06-ef76d471fcd6/materialsdatafoundry.pdf), 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](https://content.presspage.com/uploads/2580/6609582b-6231-47c2-9a06-ef76d471fcd6/materialsdatafoundry.pdf), 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](https://www.periodic.com/), 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](https://www.periodic.com/), 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](https://www.periodic.com/), 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](https://www.periodic.com/), 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](https://www.radical-ai.com/), 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](https://www.radical-ai.com/), 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](https://www.radical-ai.com/), 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](https://www.radical-ai.com/), 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 set aside

- [CuspAI, CuspAI launches global AI Materials Foundry with NVIDIA, Meta and 45 founding partners – Intelligent CIO Europe](https://www.intelligentcio.com/eu/2026/07/20/cuspai-launches-global-ai-materials-foundry-with-nvidia-meta-and-45-founding-partners/): 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](https://citrine.io/): 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](https://www.materials.zone/): 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](https://www.opendatabay.com/): 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](https://aws.amazon.com/data-exchange/providers/): 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](https://withprotege.ai/data-provider): the stored read is the cookie consent banner; the page's own text is not in it
- [Nyne.ai, Data partnerships \| Nyne.ai](https://nyne.ai/data-partnerships): 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](https://www.troveo.ai/business-data): 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](https://www.redpine.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
- [Springer Nature, Data Licensing for Research & AI \| Powering R&D with 150 Years of Expertise \| Springer Nature](https://www.springernature.com/gp/rd/research-intelligence): 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](https://www.springernature.com/gp/rd/pricing/data-solutions-intelligence): 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](https://www.lila.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

## Methodology

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.

## Sources

- [83 Sciences, 83 Sciences: AI-native materials discovery powered by unpublished experimental data \| Y Combinator](https://www.ycombinator.com/companies/83-sciences) (read)
- [83 Sciences, Head of Data Partnerships at 83 Sciences \| Y Combinator](https://www.ycombinator.com/companies/83-sciences/jobs/NR8uCdC-head-of-data-partnerships) (read)
- [CuspAI, CuspAI \| AI-powered materials discovery](https://www.cusp.ai/) (read)
- [CuspAI, CuspAI launches global AI Materials Foundry with NVIDIA, Meta and 45 founding partners – Intelligent CIO Europe](https://www.intelligentcio.com/eu/2026/07/20/cuspai-launches-global-ai-materials-foundry-with-nvidia-meta-and-45-founding-partners/) (not-terms)
- [Materials Data Foundry (NUS and University of Toronto), materialsdatafoundry.pdf](https://content.presspage.com/uploads/2580/6609582b-6231-47c2-9a06-ef76d471fcd6/materialsdatafoundry.pdf) (read)
- [Citrine Informatics, Chemicals & Materials Development Platform \| Citrine Informatics](https://citrine.io/) (not-terms)
- [Periodic Labs, Periodic Labs](https://www.periodic.com/) (read)
- [Radical AI, Radical AI](https://www.radical-ai.com/) (read)
- [CCDC, Structural Chemistry Data, Software, and Insights \| CCDC](https://www.ccdc.cam.ac.uk/) (read)
- [MaterialsZone, AI-Powered Materials Informatics \| Accelerate R&D and Innovation](https://www.materials.zone/) (not-terms)
- [Opendatabay, Licensed AI Training Data & LLM Fine-Tuning Datasets \| Opendatabay](https://www.opendatabay.com/) (not-terms)
- [AWS Data Exchange, Data Marketplace - AWS Data Exchange - AWS](https://aws.amazon.com/data-exchange/providers/) (not-terms)
- [Protege, For Data Providers - Protege](https://withprotege.ai/data-provider) (not-terms)
- [Nyne.ai, Data partnerships \| Nyne.ai](https://nyne.ai/data-partnerships) (not-terms)
- [Troveo, Turn Your Company Data Into Revenue \| Troveo](https://www.troveo.ai/business-data) (not-terms)
- [Redpine, Redpine \| Licensed data for AI agents and model training](https://www.redpine.ai/) (not-terms)
- [Springer Nature, Data Licensing for Research & AI \| Powering R&D with 150 Years of Expertise \| Springer Nature](https://www.springernature.com/gp/rd/research-intelligence) (not-terms)
- [Springer Nature, Licensing Options​ Data Solutions & Intelligence \| Powering R&D with 150 Years of Expertise \| Springer Nature](https://www.springernature.com/gp/rd/pricing/data-solutions-intelligence) (not-terms)
- [Lila Sciences, LILA \| Scientific Superintelligence](https://www.lila.ai/) (not-terms)

## Cite or embed

- 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).

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

CSV: https://sellmaterialdatatoai.com/datasets/material-data-buyers.csv
JSON: https://sellmaterialdatatoai.com/datasets/material-data-buyers.json
Embed (plain HTML, static link back): see https://sellmaterialdatatoai.com/datasets/material-data-buyers/#lead