About this report

● This is the first international report on AI safety. Following the release of the interim report in May 2024, a diverse group of 96 experts in the field of artificial intelligence contributed to the preparation of this full report. Contributors included an international advisory panel nominated by 30 countries, as well as the Organisation for Economic Co-operation and Development (OECD), the European Union, and the United Nations. The report aims to provide scientific information that can support informed policymaking. However, it does not recommend specific policy actions.

● This report is the product of independent expert work. The independent experts responsible for drafting this report had full discretion over its content, under the leadership of the chair.

● While this report addresses the risks and safety of artificial intelligence, it also highlights many potential benefits for individuals, businesses, and society. There are various types of AI, each with distinct benefits and risks. In most cases and across most applications, AI helps individuals and organizations become more effective. However, the transformative benefits of AI cannot be safely and fully realized worldwide unless its risks are appropriately managed. This report focuses on identifying these risks and evaluating ways to mitigate them. It is not intended to be a comprehensive assessment of all societal impacts of AI, although it does reference many potential positive effects.

● The report focuses on general-purpose artificial intelligence. It specifically addresses a form of AI that has rapidly advanced in recent years and whose associated risks remain insufficiently studied or understood: general-purpose AI, which can perform a wide range of tasks. The analysis in this report concentrates on the most advanced general-purpose AI systems available at the time of writing, as well as potential future systems that may be even more capable.

● The report summarizes the scientific evidence related to three essential questions: What can general-purpose AI do? What are the risks associated with it? And what technologies are used to reduce those risks?

● The stakes and risks are significant. We, the contributing experts to this report, continue to hold differing views on many questions—both minor and major—regarding the capabilities, risks, and mitigation strategies for general-purpose AI. However, we believe this report is essential to improving our collective understanding of this technology and its potential risks. We hope it will help the international community move toward greater consensus on general-purpose AI and on mitigating its risks in a way that enables people to safely benefit from its many promising applications.

Source: Department for Science, Innovation and Technology, & AI Safety Institute. (2025, January 29).

(For more information about the report, please visit the original link)