Behavioral Use Licensing for Responsible AI

By Danish Contractor, Daniel J. McDuff, J. Haines, Jenny Lee, Christopher Hines, Brent J. Hecht. 2020

95 citations recorded by the source corpus.

With the growing reliance on artificial intelligence (AI) for many different applications, the sharing of code, data, and models is important to ensure the replicability and democratization of scientific knowledge. Many high-profile academic publishing venues expect code and models to be submitted and released with papers. Furthermore, developers often want to release these assets to encourage development of technology that leverages their frameworks and services. A number of organizations have expressed concerns about the inappropriate or irresponsible use of AI and have proposed ethical guidelines around the application of such systems. While such guidelines can help set norms and shape policy, they are not easily enforceable. In this paper, we advocate the use of licensing to enable legally enforceable behavioral use conditions on software and code and provide several case studies that demonstrate the feasibility of behavioral use licensing. We envision how licensing may be implemented in accordance with existing responsible AI guidelines.

Subfields: Artificial Intelligence, Safety Research, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Safety Research

In Safety, Robustness & Interpretability

Citations in the corpus

BLOOM: A 176B-Parameter Open-Access Multilingual Language Model (2022, background): “…Being mindful of the potentially harmful use-cases that BLOOM could enable, we chose to strike a balance between unrestricted open-access and responsible-use by including behavioral-use clauses (Contractor et al., 2022) to limit the application of the model towards potentially harmful use-cases.”
The BigScience ROOTS Corpus: A 1.6TB Composite Multilingual Dataset (2023, methodology): “, 2022) to uphold the rights and responsibilities of the many stakeholders in NLP data generation and collection, and provide assurances to downstream users as to how they are and are not authorized to use the dataset (Contractor et al., 2020).”
Auditing large language models: a three-layered approach (2023, background): “By assessing the adequacy of the governance structures surrounding LLMs, including licencing agreements [176] and structured access protocols [76], governance audits help reduce the risk of malicious use.”