Q&A: Global challenges surrounding AI deployment | MIT news

The Artificial Intelligence Policy Forum (AIPF) is an initiative of the MIT Schwarzman School of Computing to move the global conversation about the impact of artificial intelligence from principles to practical implementation of policies. Formed in late 2020, AIPF Bringing together leaders in government, business and academia to me Develop approaches to address societal challenges posed by rapid developments and increase the applicability of artificial intelligence.

The co-chairs of the Artificial Intelligence Policy Forum are Alexandre Madre, Professor of Rhythm Design Systems. Aso Ozdağlar, Vice Dean of Academics for the MIT Schwarzman School of Computing and Chair of the Department of Electrical Engineering and Computer Science; and Luis Videgaray, Senior Lecturer at MIT Sloan School of Management and MIT Director of Artificial Intelligence Policy for the Global Project. Here, they discuss talking about some of the key issues facing the AI ​​policy landscape today and the challenges surrounding AI deployment. The three are co-organizers of the next AI Policy Forum Summit on September 28which will further explore the issues discussed here.

s: Can you talk about the ongoing work of the AI ​​Policy Forum and the AI ​​policy landscape in general?

Ozdağlar: There is no shortage of discussion about AI in various places, but the conversations are often at a high level, focusing on questions of ethics and principles, or on policy problems alone. The approach that AIPF takes to its work is to target specific questions with actionable policy solutions and engage with stakeholders working directly in these areas. We work “behind the scenes” with smaller focus groups to address these challenges and aim to highlight some of the potential solutions along with the players working on them directly through the large gatherings.

s: AI affects many sectors, which naturally makes us worry about its credibility. Are there any best practices emerging from trustworthy AI development and deployment?

Madhuri: The most important thing to understand about trustworthy AI deployment is that AI technology is not a natural and pre-determined phenomenon. It’s something people built. People who make certain design decisions.

Thus we need to develop research that can guide these decisions as well as provide more desirable solutions. But we also need to be thoughtful and think carefully about the incentives that drive these decisions.

Now, these incentives largely stem from business considerations, but not exclusively. This means that we must also recognize that appropriate laws and regulations, as well as thoughtful industry standard setting have a big role to play here as well.

Indeed, governments can make rules that prioritize the value of AI deployment while being fully aware of the downsides, pitfalls, and corresponding impossibilities. Designing such rules will be an ongoing and evolving process as technology continues to improve and change, and we need to adapt to social and political realities as well.

s: The financial sector is perhaps one of the most developed areas in the deployment of artificial intelligence. From a policy perspective, how should governments, regulators, and legislators make AI work better for consumers in finance?

Videgaray: The financial sector is experiencing a number of trends that pose political challenges at the intersection of AI systems. First, there is the issue of interpretability. By law (in the United States and in many other countries), lenders need to provide explanations to customers when they take harmful actions in any way, such as refusing to take out a loan, in favor of the customer. However, as financial services increasingly rely on automated systems and machine learning models, the ability of banks to empty the machine learning “black box” to provide this level of mandatory interpretation is becoming weak. So how should the finance industry and regulators adapt to this technological advancement? We may need new standards and expectations, as well as tools to meet these legal requirements.

Meanwhile, economies of scale and data network effects are driving the proliferation of AI outsourcing, and on a larger scale, AI as a service is becoming increasingly popular in the finance industry. In particular, we see fintech companies providing the tools to underwrite for other financial institutions – whether they are large banks or small local credit unions. What does this segmentation mean for the industry’s supply chain? Who is responsible for potential problems in AI systems that are spread across multiple layers of outsourcing? How can regulators adapt to ensure their mandates of financial stability, equity, and other societal standards?

s: Social media is one of the most controversial sectors of the economy, resulting in many societal transformations and upheavals around the world. What policies or reforms might be required to ensure that social media is a force for the common good and not a public harm?

Ozdağlar: The role of social media in society is a growing concern for many, but the nature of these concerns can vary slightly – some argue that social media is not doing enough to prevent, for example, disinformation and extremism, while others see it as silencing certain views. without justification. This lack of a unifying view of what the problem is affects the ability to enact any change. All of this is coupled with the complexities of the US legal framework that includes the First Amendment, Section 230 of the Communications Decency Act, and commercial laws.

However, these difficulties in regulating social media do not mean that there is nothing that can be done. Indeed, regulators have begun to tighten their control over social media companies, both in the United States and abroad, whether through antitrust measures or other means. In particular, Ofcom in the UK and EU are already introducing new layers of oversight on platforms. Additionally, some have proposed taxes on online advertising to address the negative externalities caused by the current social media business model. So, the policy tools are in place, if the political will and proper direction to implement them are available.

Leave a Comment