The first International Report on AI Safety, led by Yoshua Bengio, is launched

The first international report on the safety of artificial intelligence, led by Université de Montréal computer-science professor Yoshua Bengio, was released today and promises to serve as a guide for policymakers worldwide. 

Announced in November 2023 at the AI Safety Summit at Bletchley Park, England, and inspired by the workings of the United Nations Intergovernmental Panel on Climate Change, the report consolidates leading international expertise on AI and its risks. 

Supported by the United Kingdom’s Department for Science, Innovation and Technology, Bengio, founder and scientific director of the UdeM-affiliated Mila – Quebec AI Institute, led a team of 96 international experts in drafting the report.

The experts were drawn from 30 countries, the U.N., the European Union and the OECD. Their report will help inform discussions next month at the AI Action Summit in Paris, France and serve as a global handbook on AI safety to help support policymakers.

Towards a common understanding

The most advanced AI systems in the world now have the ability to write increasingly sophisticated computer programs, identify cyber vulnerabilities, and perform on a par with human PhD-level experts on tests in biology, chemistry, and physics. 

In what is identified as a key development for policymakers to monitor, the AI Safety Report published today warns that AI systems are also increasingly capable of acting as AI agents, autonomously planning and acting in pursuit of a goal. 

As policymakers worldwide grapple with the rapid and unpredictable advancements in AI, the report contributes to bridging the gap by offering a scientific understanding of emerging risks to guide decision-making.  

The document sets out the first comprehensive, independent, and shared scientific understanding of advanced AI systems and their risks, highlighting how quickly the technology has evolved.  

Several areas require urgent research attention, according to the report, including how rapidly capabilities will advance, how general-purpose AI models work internally, and how they can be designed to behave reliably. 

Three distinct categories of AI risks are identified: 

• Malicious use risks: these include cyberattacks, the creation of AI-generated child-sexual-abuse material, and even the development of biological weapons; 

• System malfunctions: these include bias, reliability issues, and the potential loss of control over advanced general-purpose AI systems; 

• Systemic risks: these stem from the widespread adoption of AI, include workforce disruption, privacy concerns, and environmental impacts.  

The report places particular emphasis on the urgency of increasing transparency and understanding in AI decision-making as the systems become more sophisticated and the technology continues to develop at a rapid pace. 

While there are still many challenges in mitigating the risks of general-purpose AI, the report highlights promising areas for future research and concludes that progress can be made.   

Ultimately, it emphasizes that while AI capabilities could advance at varying speeds, their development and potential risks are not a foregone conclusion. The outcomes depend on the choices that societies and governments make today and in the future. 

"The capabilities of general-purpose AI have increased rapidly in recent years and months,” said Bengio. “While this holds great potential for society, AI also presents significant risks that must be carefully managed by governments worldwide.  

“This report by independent experts aims to facilitate constructive and evidence-based discussion around these risks and serves as a common basis for policymakers around the world to understand general-purpose AI capabilities, risks and possible mitigations.”