International cooperation needed to tackle uncertain future of AI, says expert report

The future of general-purpose AI is highly uncertain and effectively addressing its risks to safety requires international collaboration, robust regulation, and improved research, according to the interim publication of the first International Scientific Report on the Safety of Advanced AI. Released on Friday, the report highlights AI’s potential to drive significant societal benefits, as well as the serious risks the technology can pose if not properly developed and regulated. It also notes a lack of consensus among the world’s AI experts in these areas, calling for societies to urgently prioritize improving their understanding of its capabilities. A group of 75 AI experts contributed to the report, including an international expert advisory panel nominated by 30 countries and representatives from the European Union and the United Nations. It was commissioned at the world’s first AI Safety Summit last November and is chaired by Yoshua Bengio, University of Montreal professor, scientific director of the Mila AI research institute, and known as one of the ‘godfathers of AI.’ In February, Bengio urged the House industry committee to adopt Bill C-27’s Artificial Intelligence and Data Act as soon as possible, warning of serious risks if the technology continues to go unregulated. While AI can drive progress in healthcare, education, and scientific research, it also poses risks such as biased decision-making, disinformation, privacy violations, and labour market disruptions. Society’s responses to these dangers will significantly shape the future trajectory of AI development, says the report. "Our mission is clear: to drive a shared, science-based, up-to-date understanding of the safety of advanced AI, and to continue to develop that understanding over time," said Bengio in the report’s foreword. “Governments, academia, and the wider society need to continue to advance the AI safety agenda to ensure we can all harness AI safely, responsibly, and successfully,” he added in an accompanying press release from the United Kingdom’s Department for Science, Innovation and Technology. The report outlines the current capabilities of general-purpose AI, noting that while these systems have made significant strides in natural language processing, image generation, and other tasks, they still face limitations. Current AI systems often struggle with tasks requiring complex reasoning and can exhibit biases or fabricate content through ‘hallucinations.’ It calls for further research to improve the understanding of AI models' inner workings, as many systems operate in ways that are not fully clear, even to their developers. Systemic risks posed by the widespread adoption of AI were examined. These include the potential for significant economic disruptions, with many jobs potentially being automated, and the risk of increasing inequality as AI capabilities are concentrated in a few countries, limiting who can reap the benefits. Developing state-of-the-art, general-purpose AI models requires substantial up-front investment, creating barriers to entry and disproportionately benefiting large technology companies. The report explains that this concentration of market power among a few companies could lead to systemic risks, as any flaws in dominant AI systems could cause widespread disruptions. It also notes some environmental concerns, such as the energy consumption associated with training large AI models and how it could exacerbate carbon emissions and strain resources. The report stresses the need for regulatory frameworks and measures to ensure fairness, transparency, privacy, and accountability in AI systems. The importance of including diverse perspectives in the development and evaluation of AI is also highlighted as a way to ensure that its benefits are broadly shared. “Building on the momentum we created with our historic talks at Bletchley Park, this report will ensure we can capture AI’s incredible opportunities safely and responsibly for decades to come,” said the UK’s Secretary of State for Science, Innovation, and Technology Michelle Donelan. The report, she added, will serve as a critical foundation for discussions at next week’s AI Seoul Summit, scheduled for May 21. The interim report is seen as a starting point for the upcoming discussions. A final report, incorporating additional evidence and feedback from a wider range of stakeholders, is expected to be released later this year ahead of the next AI Safety Summit, hosted by France. – Reporting by Phalen Tynes-MacDonald at pcreed@thewirereport.ca

International cooperation needed to tackle uncertain future of AI, says expert report

News | May 17, 2024

Unlock all the Canadian telecom, broadcasting and digital media news you need.

FREE TRIAL

SUBSCRIBE

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us

International cooperation needed to tackle uncertain future of AI, says expert report

News | May 17, 2024

Unlock all the Canadian telecom, broadcasting and digital media news you need.

FREE TRIAL

SUBSCRIBE

Related Stories

OPINION: Government agencies using AI must implement governance programs, or risk losing trust

Champagne announces launch of Canadian AI Safety Institute

Ericsson invests $634 million in R&D in Canada

Public broadcasters formally commit to addressing disinformation, news availability

Reuse & Permissions

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us