Contrary to Traditional machine learning approaches which focuses on identifying statistical patterns, causal AI focuses on understanding causal relationships between variables in a system, rather than associations or correlations, which are prone to data inaccuracies.
The field of causal AI is set to play an increasingly important role in developing more reliable and useful systems, and has demonstrated impressive capabilities in many domains. This methodology holds great promise for high-stakes applications, such as in healthcare, finance, and public policy, where “explainability” and accountability are paramount.
However, causal AI represents a paradigm shift in human-AI dynamics and can help address the major concern that AI systems will increasingly have limited or no human interactions.
Importantly, causal AI can equip other forms of AI with the ability to improve their decision-making and preserve human agency, with the AI system acting as a tool or executor, and humans as curators and decision-makers. This scientific methodology can significantly support regulatory and oversight mechanisms to ensure that AI systems explain their decision-making processes and outcomes in a consistent and transparent manner.
As regulators around the world struggle to understand the intricacies of governing these emerging technologies, the European Union recently established a legal and regulatory framework whose provisions will come into force over the next six to three months.
EU implementation Artificial Intelligence Act paves the way for increased international cooperation and harmonization of governance standards around AI. climate changeAI requires coordinated efforts and collaboration among stakeholders from different jurisdictions to ensure responsible development and implementation.
Like many other jurisdictions, Hong Kong does not have a comprehensive regulatory or statutory framework to govern AI. Guidelines on the ethical use of AI and the Monetary Authority has published “high level principles”while recognising the need to provide safeguards and guidance. But much more rigorous regulatory oversight is needed.
The Hong Kong Privacy Commissioner’s Office released its report, “Artificial Intelligence: A Model Framework for Personal Data Protection,” on June 11. Photo: May Tse
Hong Kong’s common law system, its role as international arbitration center and the depth of professional services, combined with the analytical capacity and transparency inherent in causal AI, could give the city a strong competitive advantage in establishing a robust, evidence-based framework for the responsible deployment of AI and its ethical governance.
A dedicated agency and regulatory structure is needed to set standards for AI development and deployment. This would enable Hong Kong to attracting the best talentsdrive innovation and become a global leader in ethical governance of innovation and responsible use of AI. Such proactive measures would benefit the city and allow it to play a greater role internationally as other jurisdictions look for models for AI policy development.
In this context, cybersecurity The issue of causality remains a key consideration. Traditional machine learning models, which often rely on statistical correlations in data, can be vulnerable to sophisticated cyberattacks that exploit these patterns. In contrast, causal AI’s focus on uncovering causal mechanisms provides a more robust defense against malicious actors seeking to manipulate or subvert AI systems.
By modeling the causes of complex phenomena, causal AI can better detect anomalies, identify sources of cyber threats, and respond with greater agility. This capability is critical as AI becomes increasingly integrated into critical infrastructure, financial systems, and other high-stakes domains.
02:15
Airports worldwide see operations disrupted due to Microsoft systems outage
Airports worldwide see operations disrupted due to Microsoft systems outage
Just look at the recent breakdown The losses suffered by millions of Microsoft users due to a botched software update provide some insight into the scale of the disruption that can occur when IT systems are disrupted on a global scale. Some sources estimate losses at more than $5 billion for Fortune 500 companies alone.
It is worrying that the annual cost of cybercrime is expected to reach $10.5 trillion by 2025, and consumers will ultimately bear the cost.
The discourse around advanced technologies reflects their profound transformative potential and the urgent need to thoughtfully address their complex ethical, social and economic challenges.
By seizing the opportunity to establish a comprehensive, evidence-based approach to AI governance and regulation, Hong Kong can become an important global player in the ethical and responsible use of these powerful technologies.
Dr Jane Lee is Chairperson of Our Hong Kong Foundation
We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.Ok
