{{Short description|Non-profit German artificial intelligence organization}}{{Infobox company | name = LAION | logo = LAION logo.svg | logo_size = 150px | type = Non-profit | industry = Artificial intelligence | founder = {{Unbulleted list| Christoph Schuhmann | Jenia Jitsev | Richard Vencu | Robert Kaczmarczyk | Theo Coombes | Mehdi Cherti | Aarush Katta | Jan Ebert}} }}
'''LAION''' (acronym for '''Large-scale Artificial Intelligence Open Network''') is a German non-profit organization which makes open-sourced artificial intelligence models and datasets.{{r|About}} It is best known for releasing a number of large datasets of images and captions scraped from the web which have been used to train a number of high-profile text-to-image models, including Stable Diffusion and Imagen.{{r|Ars-Trained}}{{r|BB teacher}}
In February 2023, LAION was named in the Getty Images lawsuit against Stable Diffusion as a non-party.<ref>{{Cite web |title=Getty Images (US), Inc. v. Stability AI, Inc., 1:23-cv-00135 |url=https://www.courtlistener.com/docket/66788385/getty-images-us-inc-v-stability-ai-inc/ |access-date=2023-02-08 |website=CourtListener |language=en-us}}</ref> In April 2023, LAION was directly sued by a German photographer who wanted to have his images removed from the training set.<ref>{{Cite web |title=A Photographer Tried to Get His Photos Removed from an AI Dataset. He Got an Invoice Instead. |url=https://www.vice.com/en/article/a-photographer-tried-to-get-his-photos-removed-from-an-ai-dataset-he-got-an-invoice-instead/ |access-date=2023-05-04 |website=Vice |date=28 April 2023 |language=en-us}}</ref> In September 2024, the Regional Court of Hamburg dismissed the lawsuit, in what was described as a "landmark ruling on TDM [<nowiki/>Text and data mining] exceptions for AI training data" in Germany and the EU more generally.<ref name=":2">{{Cite web |last1=Goldstein |first1=Paul |author-link1=Paul Goldstein (law professor) |last2=Stuetzle |first2=Christiane |last3=Bischoff |first3=Susan |date=2024-11-13 |title=Kneschke vs. LAION - Landmark Ruling on TDM exceptions for AI training data – Part 1 |url=https://copyrightblog.kluweriplaw.com/2024/11/13/kneschke-vs-laion-landmark-ruling-on-tdm-exceptions-for-ai-training-data-part-1/ |access-date=2024-11-25 |website=Kluwer Copyright Blog |language=en-US}}</ref>
On April 15, 2023, LAION and contributors publicly released an open source AI assistant chatbot called OpenAssistant.
==Image datasets== LAION has publicly released a number of large datasets of image-caption pairs which have been widely used by AI researchers.{{Citation needed|date=June 2025}} The data is derived from the Common Crawl, a dataset of scraped web pages. The developers searched the crawled html for <code><img></code> tags and treated their alt attributes as captions. They used CLIP to identify and discard images whose content did not appear to match their captions.{{r|Infoq-5b}} LAION does not host the content of scraped images themselves; rather, the dataset contains URLs pointing to images, which researchers must download themselves.{{r|Ars-medical}}
The first such dataset, LAION-400M, was released in August 2021 and consisted of 400 million image-caption pairs. The pairs were extracted from a random subset of webpages scraped by Common Crawl between 2014 and 2021.{{r|Laion-400m-blog}} It was an attempt to recreate the process used by OpenAI to collect the 400 million image-caption pairs they used to train the CLIP model - the company had chosen to open-source the model's code and weights, but not its training dataset.{{r|Infoq-5b}} Imagen, a text-to-image model announced by Google Brain in 2022, was trained on LAION-400M in combination with private internal datasets.{{r|imagen-paper}}
A successor of more than 5 billion pairs, LAION-5B, was released in March 2022.{{r|Laion-5b-blog}} As of its release, it was the largest freely available dataset of image-caption pairs in existence.{{r|Infoq-5b}} Its creation was funded by Doodlebot, Hugging Face and Stability AI, the AI company behind the funding of the Stable Diffusion text-to-image model, which was trained on it.{{r|tc-sai}}
=== Criticism === Several studies show that the images in LAION-5B contain problematic images and text pairs of rape, pornography, malign stereotypes, racist and ethnic slurs, and other extremely problematic content.<ref>{{Cite journal |last1=Birhane |first1=Abeba |last2=Prabhu |first2=Vinay Uday |last3=Kahembwe |first3=Emmanuel |date=2021 |title=Multimodal datasets: misogyny, pornography, and malignant stereotypes |arxiv=2110.01963}}</ref><ref>{{Citation |last1=Birhane |first1=Abeba |title=Into the LAIONs Den: Investigating Hate in Multimodal Datasets |date=2023-11-06 |arxiv=2311.03449 |last2=Prabhu |first2=Vinay |last3=Han |first3=Sang |last4=Boddeti |first4=Vishnu Naresh |last5=Luccioni |first5=Alexandra Sasha|author-link5=Sasha Luccioni}}</ref>
An investigation by Bayerischer Rundfunk showed that LAION's datasets, hosted on Hugging Face, contain large amounts of private and sensitive data harvested from public websites.<ref>{{Cite web |last1=Brunner |first1=Katharina |last2=Harlan |first2=Elisa |date=2023-06-07 |title=We Are All Raw Material for AI |url=https://interaktiv.br.de/ki-trainingsdaten/en/index.html |website=Bayerischer Rundfunk}}</ref>
In December 2023, the Stanford Internet Observatory released a report on LAION-5B that found 3,226 suspected instances of links to child sexual abuse material with 1,008 of these being externally validated. In response, LAION temporarily removed LAION-5B and LAION-400M citing its "zero tolerance policy for illegal content" and "an abundance of caution".<ref>{{cite web |last1=Cole |first1=Samantha |title=Largest Dataset Powering AI Images Removed After Discovery of Child Sexual Abuse Material |url=https://www.404media.co/laion-datasets-removed-stanford-csam-child-abuse/ |website=404 Media |language=en |date=20 December 2023 |accessdate=22 December 2023}}</ref> In August 2024, LAION released a cleaned dataset called Re-LAION-5B.<ref>{{Cite web |last=Belanger |first=Ashley |date=2024-08-30 |title=Nonprofit scrubs illegal content from controversial AI training dataset |url=https://arstechnica.com/tech-policy/2024/08/nonprofit-scrubs-illegal-content-from-controversial-ai-training-dataset/ |access-date=2024-08-31 |website=Ars Technica |language=en-us}}</ref>
== OpenAssistant == {{Infobox software | title = OpenAssistant | screenshot = Open Assistant Dashboard.png | logo = Open-Assistant-logo.svg | logo caption = | caption = Screenshot of the data collection web portal | developer = LAION and contributors | released = {{Start date and age|2023|04|15|df=yes}} | genre = {{ indented plainlist | *Large Language Model *Generative pre-trained transformer *Chatbot }} | license = Apache License 2.0 | website = {{URL|https://open-assistant.io}} }}
OpenAssistant was an artificial intelligence (AI) open source chat-based assistant that could understand tasks, interact with third-party systems and retrieve information dynamically to do so. The project was developed by a group of volunteers in collaboration with LAION. One of the goals for development included free access to large language models that can be run locally on consumer hardware.<ref>{{Citation |title=Open-Assistant |date=2023-03-09 |url=https://github.com/LAION-AI/Open-Assistant |access-date=2023-03-09 |publisher=LAION AI}}</ref><ref name=":0">{{Cite arXiv |eprint=2304.07327 |class=cs.CL |first1=Andreas |last1=Köpf |first2=Yannic |last2=Kilcher |title=OpenAssistant Conversations -- Democratizing Large Language Model Alignment |date=2023-04-14 |last3=von Rütte |first3=Dimitri |last4=Anagnostidis |first4=Sotiris |last5=Tam |first5=Zhi-Rui |last6=Stevens |first6=Keith |last7=Barhoum |first7=Abdullah |last8=Duc |first8=Nguyen Minh |last9=Stanley |first9=Oliver |last10=Nagyfi |first10=Richárd |last11=ES |first11=Shahul |last12=Suri |first12=Sameer |last13=Glushkov |first13=David |last14=Dantuluri |first14=Arnav |last15=Maguire |first15=Andrew}}</ref> The project was backed by a worldwide crowdsourcing effort involving over 13,500 volunteers who have created 600k human-generated data points.<ref name=":0" /><ref>{{Cite web |title=Open Assistant: Explore the Possibilities of Open and Collaborative Chatbot Development |url=https://www.kdnuggets.com/open-assistant-explore-the-possibilities-of-open-and-collaborative-chatbot-development.html |access-date=2023-05-05 |website=KDnuggets |language=en-US }}{{Dead link|date=July 2025 |bot=InternetArchiveBot |fix-attempted=yes }}</ref> The project has since been shut down; however, the datasets and models remain available on Hugging Face.
== See also ==
* Artificial intelligence and copyright
== References == <references>
<ref name="About">{{cite web |work=LAION.ai |title=About |access-date=26 September 2022 |url=https://laion.ai/about/ }}</ref> <ref name="Ars-Trained">{{cite news |work=Ars Technica |date=15 September 2022 |last=Edwards|first=Benj |title=Have AI image generators assimilated your art? New tool lets you check |url=https://arstechnica.com/information-technology/2022/09/have-ai-image-generators-assimilated-your-art-new-tool-lets-you-check/ }}</ref> <ref name="BB teacher">{{cite news |last1=Newman |first1=Marissa |last2=Cantrill |first2=Aggi |title=The Future of AI Relies on a High School Teacher's Free Database |url=https://www.bloomberg.com/news/features/2023-04-24/a-high-school-teacher-s-free-image-database-powers-ai-unicorns |access-date=24 April 2023 |work=Bloomberg News |date=24 April 2023 |language=en}}</ref> <ref name="Infoq-5b">{{cite web |work=InfoQ |last=Alford|first=Anthony |date=17 May 2022 |title=LAION Releases Five Billion Image-Text Pair Dataset LAION-5B |url=https://www.infoq.com/news/2022/05/laion-5b-image-text-dataset/ }}</ref> <ref name="Laion-400m-blog">{{cite web |work=LAION blog |last=Schuhmann|first=Christoph |title=LAION-400-Million Open Dataset |date=8 August 2021 |access-date=26 September 2022 |url=https://laion.ai/blog/laion-400-open-dataset/ }}</ref> <ref name="Laion-5b-blog">{{cite web |work=LAION blog |last=Beaumont|first=Romain |date=3 March 2022 |title=LAION-5B: A New Era of Open Large-Scale Multi-Modal Datasets |url=https://laion.ai/blog/laion-5b/ }}</ref> <ref name="Ars-medical">{{cite news |work=Ars Technica |last=Edwards|first=Benj |date=21 September 2022 |title=Artist finds private medical record photos in popular AI training data set |url=https://arstechnica.com/information-technology/2022/09/artist-finds-private-medical-record-photos-in-popular-ai-training-data-set/ }}</ref> <ref name="imagen-paper">{{cite arXiv |title=Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding |date=23 May 2022|last1=Saharia|first1=Chitwan |last2=Chan|first2=William |last3=Saxena|first3=Saurabh |last4=Li|first4=Lala |last5=Whang|first5=Jay |last6=Denton|first6=Emily |last7=Kamyar Seyed Ghasemipour|first7=Seyed |last8=Karagol Ayan|first8=Burcu |last9=Sara Mahdavi|first9=S. |last10=Gontijo Lopes|first10=Rapha |last11=Salimans|first11=Tim |last12=Ho|first12=Jonathan |last13=J Fleet|first13=David |last14=Norouzi|first14=Mohammad |class=cs.CV |eprint=2205.11487 }}</ref> <ref name="tc-sai">{{cite web |work=TechCrunch |last=Wiggers|first=Kyle |date=12 August 2022 |title=This startup is setting a DALL-E 2-like AI free, consequences be damned |url=https://techcrunch.com/2022/08/12/a-startup-wants-to-democratize-the-tech-behind-dall-e-2-consequences-be-damned/ }}</ref>
</references>
Category:Applications of artificial intelligence Category:Open-source artificial intelligence Category:Artificial intelligence laboratories Category:Non-profit organisations based in Germany