Overview
The NYC AI Action Plan is a pioneering initiative by New York City to govern and leverage artificial intelligence technologies across its various governmental agencies. This comprehensive plan aims to establish robust frameworks for AI governance, ensuring that these technologies are implemented responsibly and ethically to enhance city operations. Key components of the plan include creating an Office of Algorithmic Data Integrity, and establishing guidelines to prevent biases and ensure transparency in AI deployments.
The plan also focuses on enhancing public engagement, building AI capabilities within city staff, and maintaining strict oversight through continuous monitoring and regular updates of AI applications. Anticipated outcomes include improved public service efficiency, enhanced public safety, and increased transparency in government operations. However, challenges such as technological complexity, resource allocation, and effective public engagement raise questions about the plan's implementation and overall efficacy.
Goals and Aspirations
Enhance Governance and Oversight of AI Technologies. The plan aims to establish a robust governance framework that ensures AI technologies used by city agencies are deployed responsibly and ethically.
Foster Public Engagement and Transparency. The initiative seeks to enhance public understanding and involvement in AI policy-making and its applications within the city. This involves holding public listening sessions, developing public-facing resources, and ensuring that the city's efforts in AI are transparent and accountable to New Yorkers.
Build AI Capability within City Government. The plan outlines steps to increase AI literacy and capability across city agencies. This includes identifying skills needs, fostering information sharing, and supporting the implementation of AI projects within the city government. The goal is to ensure that city personnel are equipped to utilize AI tools effectively and responsibly.
Key Characteristics
Development. The plan builds on extensive foundational research and previous initiatives, incorporating insights from a diverse range of stakeholders and experts. This includes the development of an AI Primer and the NYC AI Strategy, which helped establish a shared understanding of AI technologies and the issues they present. The other significant part of the plan's development phase involved establishing a comprehensive governance framework that outlines guiding principles, definitions, and a typology of AI projects. This framework serves to standardize the approach across city departments and agencies.
Engagement. The development of the action plan involved engaging over fifty city staff members from eighteen agencies, along with external experts from ten organizations. What's more, it emphasizes the importance of public engagement through introductory public listening sessions and ongoing communication channels. It aims to foster an environment where New Yorkers can provide input, ensuring the AI implementations consider the community's needs and concerns.
Implementation. The plan details a set of phased actions for implementation, focusing specifically on city government’s use of AI. These actions include establishing an AI Steering Committee, expanding public AI reporting, and developing an AI risk assessment and project review process. Implementation also involves supporting city agencies with the necessary policy guidance, support, and resources to advance their AI initiatives effectively. This includes building AI knowledge and skills within city government to prepare personnel for working with and managing AI tools.
Stakeholders
Eric Adams. As the Mayor of New York City, he plays a pivotal role in endorsing and promoting the AI Action Plan, aligning it with broader city governance and public service initiatives. New York City Office of Mayor
Matthew Fraser. The Chief Technology Officer of NYC, responsible for leading the OTI and overseeing the strategic direction of the city's technology initiatives, including AI. I was not able to conduct an interview with Fraser but I learned from his role and perpectives from his video: The AI Action Plan reflects NYC's ambition to lead by example in the governance of AI technologies at a municipal level. Fraser’s discussion emphasizes that the city is not waiting for federal regulations to catch up but is instead taking proactive steps to address the challenges and opportunities of AI now. A significant point made in the talk is the city's focus on ethical considerations in deploying AI. This includes ensuring that AI tools are used in a manner that does not discriminate or infringe on privacy and that they operate transparently. The plan is designed to build trust among citizens that AI technologies will enhance, not hinder, public services. Office of Technology and Innovation
Jennifer Gutiérrez. Gutiérrez introduced legislation to create an Office of Algorithmic Data Integrity in New York City. It would assess AI systems before they are deployed by the city to ensure they are free of bias and discrimination. City Council
Technology Interventions
AI Risk Assessment and Project Review Process.
- Use Case: This intervention is intended to evaluate AI projects for risks like bias, privacy breaches, and other potential negative impacts before their deployment.
- Value Proposition: Ensures AI tools used in city operations adhere to ethical standards, minimizing the risk of harm and enhancing trust in public sector AI use.
- Solution Architecture: Involves tools and frameworks for assessing AI algorithms against predefined criteria, possibly integrating automated testing and monitoring systems that continuously evaluate the performance and impact of AI applications.
- Data Created or Consumed: Consumes operational data from AI applications and produces reports on compliance and risk assessments.
- Key Platforms and Standards: Might utilize existing AI assessment platforms, integrate with the city’s IT infrastructure, and adhere to standards such as the AI Risk Management Framework by NIST.
- Business Models: Operates under a governance-focused model, with funding from city budgets allocated for technology oversight.
- Regulatory Issues: Aligns with local and federal regulations on data privacy and AI, such as NYC’s Local Law 49 concerning automated decision systems.
Public AI Reporting and Engagement Platform.
- Use Case: Designed to enhance transparency and public engagement through a comprehensive reporting and feedback mechanism on AI projects.
- Value Proposition: Provides a direct line of communication between the city’s AI initiatives and the public, fostering transparency, accountability, and trust.
- Solution Architecture: A web-based platform that allows users to access AI project details, feedback mechanisms, and educational resources about AI applications within city operations.
- Data Created or Consumed: Creates public engagement data and consumes feedback and AI performance data to inform continuous improvements and policy adjustments.
- Key Platforms and Standards: Could be built on open-source technologies with adherence to web accessibility standards to ensure broad public access.
- Business Models: Supported through city funding, focusing on public service and community engagement without direct revenue generation.
- Regulatory Issues: Must comply with digital accessibility and privacy regulations.
Centralized AI Knowledge Hub for City Employees.
- Use Case: Aims to build AI capability within city government by providing centralized resources and training for city employees.
- Value Proposition: Equips city employees with the skills necessary to implement and manage AI tools responsibly, enhancing the effectiveness and ethical deployment of AI technologies.
- Solution Architecture: An intranet portal featuring training modules, best practices, policy guidelines, and forums for collaboration among city employees.
- Data Created or Consumed: Consumes educational content and user interaction data; creates data on learning progress, skills assessment, and potentially anonymized case studies of AI usage.
- Key Platforms and Standards: Likely integrates with existing city IT infrastructure and learning management systems, adhering to standards for educational content and IT security.
- Business Models: Funded as a capacity-building initiative within city budgets, focusing on long-term value through enhanced employee performance and risk reduction.
- Regulatory Issues: Compliance with IT security and data protection standards is crucial, particularly concerning the handling of sensitive information.
Financing
The NYC AI Action Plan does not explicitly detail the specific financing mechanisms for each technology intervention, but it outlines the general approach to funding as part of the larger project framework. Typically, such initiatives are funded through a combination of city budget allocations, which are approved by the city council and mayor's office, potentially supplemented by state or federal grants, especially when the projects align with broader governmental goals of promoting innovative technologies and public safety.
Outcomes
- Establish a framework for AI governance that acknowledges the risks of AI, including bias and disparate impact; NYC could set a precedent for others.
- Enable responsible AI acquisition with AI-specific guidance to support agency-level contracting.
- Publish an annual AI progress report to communicate about the city’s progress and implementation.
Open Questions
-
How will the AI Action Plan handle data privacy concerns, particularly regarding the collection, use, and sharing of personal information by AI systems?
-
How does the plan propose to measure and report the success and impact of its AI implementations across different sectors of city governance?
-
Will the public engagement efforts sufficiently address the concerns and expectations of New Yorkers regarding AI use in city governance?
References
Primary Sources
- The New York City Artificial Intelligence Action Plan
- NYCOTI - OTI Announces Progress on Nation’s First Comprehensive Artificial Intelligence Action Plan
- Columbia University Data Science Institute- Data Science Institute Leadership Helps NYC Advance its AI Action Plan
Secondary Sources
- https://www.wired.com/story/us-failed-to-pass-ai-regulation-new-york-city-stepping-up/
- https://cities-today.com/new-york-sets-out-an-action-plan-on-artificial-intelligence/
- https://www.police1.com/artificial-intelligence/articles/nyc-unveils-ai-action-plan-detailing-the-planned-use-and-governance-of-evolving-tech-JC4rhVFglccuLHK8/
- https://www.cityandstateny.com/personality/2023/10/how-new-york-city-has-developed-its-ai-strategy/391394/
- https://www.bloomberg.com/news/articles/2023-10-16/how-mayor-eric-adams-plans-to-expand-nyc-use-of-ai
- https://www.bloomberg.com/news/articles/2024-03-13/ai-powered-urban-innovations-bring-promise-risk-to-future-cities
- https://statescoop.com/nyc-ai-plan-eric-adams/