The world’s largest association of information technology professionals publishes Principles for Generative AI Technologies

In response to the great advances in AI technologies, as well as the significant questions these technologies pose in areas such as intellectual property, the future of work and even human security, the Global Technology Policy Council (ACM TPC) of the Association for Computing Machinery has issued “Principles for the development, deployment and use of AI Generative Intelligence Technologies”.

Drawing on the deep technical expertise of computer scientists in the United States and Europe, the ACM TPC Statement outlines eight principles intended to promote fair, accurate and beneficial decision-making regarding generative technologies and all other artificial intelligence technologies. Four of the principles are specific to Generative AI, and four other principles are adapted from the TPC’s 2022 “Statement of Principles for Responsible Algorithmic Systems.”

The introduction to the new principles advances the central argument that “the growing power of generative AI systems, the speed of their evolution, broad application, and potential to cause significant or even catastrophic damage, means that great attention must be paid attention in researching, designing, developing, distributing and using them. Existing mechanisms and ways to avoid such harms will probably not be sufficient”.

The document then sets out these eight instrumental principles, outlined here in abbreviated form:

Specific principles of generative AI

• Limitations and guidance on distribution and use: In consultation with all stakeholders, the law and regulation should be reviewed and enforced as written or revised to limit the implementation and use of AI technologies when necessary to minimize harm. No high-risk AI system should be allowed to operate without clear and adequate safeguards, including a “human being in the loop” and a clear consensus among stakeholders that the benefits of the system will substantially outweigh its potential negative impacts . One approach is to define a hierarchy of risk levels, with unacceptable risk at the highest level and minimal risk at the lowest level.
• Property: The intrinsic aspects of how generative AI systems are structured and function are not yet adequately addressed in intellectual property (IP) laws and regulations.
• Personal data control: Generative AI systems should allow a person to forego using their data to train a system or facilitate its generation of information.
• Correctability: Vendors of Generative AI systems should create and maintain public archives where errors made by the system can be noted and, optionally, made corrections.

Previous principles adapted

• Transparency: Any application or system that uses Generative AI should clearly state that it does so to the appropriate stakeholders.
• Verifiability and contestability: Generative AI system providers should ensure that system models, algorithms, data and outputs can be recorded where possible (with due consideration of privacy), so that they can be verified and/or challenged in appropriate cases .
• Limit the environmental impact: Given the large environmental impact of generative AI models, we recommend developing a consensus on methodologies to actively measure, attribute, and reduce that impact.
• Greater security and privacy: Generative AI systems are susceptible to a wide range of new security and privacy risks, including new attack vectors and malicious data leaks, among others.

“Our field needs to tread carefully in the development of Generative AI because it is a new paradigm that goes significantly beyond previous AI technology and applications,” explained Ravi Jain, Chair of the Generative AI Working Group of the ACM Technology Policy Council and lead author of the Principles. “Whether we celebrate generative AI as a wonderful scientific advance or fear it, everyone agrees that we need to develop this technology responsibly. In outlining these eight instrumental principles, we have tried to consider a broad range of areas where Generative AI could play a role. These include aspects that haven’t been covered so much in the media, including environmental considerations and the idea of ​​creating public repositories where errors in a system can be noticed and corrected.

“These are guidelines, but we also need to build a community of scientists, policymakers and industry leaders who will work together in the public interest to understand the limitations and risks of generative AI, as well as its benefits. ACM’s position as the world’s largest association of information technology professionals makes us well suited to promote that consensus, and we look forward to working with policy makers to shape the regulations under which generative AI should be developed, implemented, but also controlled,” added James Hendler, Professor at Rensselaer Polytechnic Institute and chair of ACM’s Technology Policy Council.

“Principles for the Development, Deployment, and Use of Generative AI Technologies” was jointly produced and adopted by the United States Technology Policy Committee (USTPC) and ACM’s European Technology Policy Committee (Europe TPC).

The lead authors of this document for USTPC were Ravi Jain, Jeanna Matthews and Alejandro Saucedo. Important contributions were made by Harish Arunachalam, Brian Dean, Advait Deshpande, Simson Garfinkel, Andrew Grosso, Jim Hendler, Lorraine Kisselburgh, Srivatsa Kundurthy, Marc Rotenberg, Stuart Shapiro and Ben Shneiderman. Assistance was also provided by Ricardo Baeza-Yates, Michel Beaudouin-Lafon, Vint Cerf, Charalampos Chelmis, Paul DeMarinis, Nicholas Diakopoulos, Janet Haven, Ravi Iyer, Carlos E. Jimenez-Gomez, Mark Pastin, Neeti Pokhriyal, Jason Schmitt and Darryl Scriven.

About the ACM Technology Policy Council
ACM’s Global Technology Policy Council sets the agenda for global initiatives to address evolving technology policy issues and coordinates the activities of ACM’s regional technology policy committees in the United States and Europe. It serves as the central meeting point for ACM’s interactions with government organizations, the computing community, and the public in all public policy matters related to computing and information technology. Board members are drawn from the global membership of the ACM.

About the ACM
ACM, the Association for Computing Machinery, is the world’s largest educational and scientific computing society, bringing together computer science educators, researchers, and practitioners to inspire dialogue, share resources, and address industry challenges. ACM strengthens the collective voice of the IT profession through strong leadership, the promotion of the highest standards and the recognition of technical excellence. ACM supports the professional growth of its members by providing opportunities for lifelong learning, career development and professional networking.

###