When it comes to AI, is New York City handing out treats while the rest of the world is stuck with tricks?
On October 16th, NYC announced the release of an AI Action Plan to promote responsible AI in the City government. The Plan contains initiatives like AI education, implementation, procurement, and stakeholder engagement. The 51-page document includes 37 actions to govern AI, with 29 set to be rolled out in 2024." - RAI Institute
Because most AI policies stifle innovation with restrictive guidelines and regulations, I lean towards a concept I call 'AI Δnαrchy.' I was pleasantly surprised that New York City's AI Action Plan didn't give off the restrictive undertones so often expressed. However, some of reactions to the policy did raise my antenna.
I've said it before, the fundamental challenge in regulating artificial intelligence is our inability to define 'general intelligence' itself.
Everything starts with a definition and one of my many views and queries about regulating artificial intelligence is how can we regulate artificial intelligence when we still cannot define 'general intelligence"? Fortunately, the Big Apple has put forth a definition of artificial intelligence.
The NYC Plan defines AI as:
“an umbrella term without precise boundaries, that encompasses a range of technologies and techniques of varying sophistication that are used to, among other tasks, make predictions, inferences, recommendations, rankings, or other decisions with data, and that includes topics such as machine learning, deep learning, supervised learning, unsupervised learning, reinforcement learning, statistical inference, statistical regression, statistical classification, ranking, clustering, and expert systems.”
Not bad, not bad at all.
A little long in the tooth, but the definition was created by committee. What's truly commendable is the way the definition focuses on delineating the technology behind artificial intelligence, rather than attempting to define the nebulous concept of AI itself. This approach represents a grounded and clarifying understanding serving as a solid foundation for governance and application.
The plan has seven 7 key city AI initiatives:
Having been a party to many and varied IT and Managed IT Services projects, I've found this deliverable easy to understand and actionable and would be pleased to work within the defined parameters. In general, I recommend reviewing anything that has to do with artificial intelligence in a business and work environment; from applications to AI implementation.
The NYC AI Policy is a good model to base your AI project. What follows is an in-depth look and reflections of the 55-page AI Policy.
The New York City Artificial Intelligence Action Plan, released in October 2023, is a comprehensive document that outlines the city's approach to harnessing the power of AI for public benefit while mitigating associated risks. The plan is produced by the Office of Technology and Innovation and represents contributions from 50 city employees across 18 agencies, as well as insights from industry, academia, and civil society. The plan introduces seven new initiatives, each with a set of phased actions, to provide more centralized policy, guidance, support, and resources for city agencies. These initiatives aim to ensure appropriate governance measures, equip agencies to advance their AI efforts, and work efficiently and accountably. The plan also emphasizes collaborative engagement with experts from academia, industry, civil society, community organizations, and organized labor. The document is organized into various sections, including letters from the Mayor and the Chief Technology Officer, an introduction to AI, foundational efforts, moving forward with action, and a summary of initiatives and actions.
The NYC AI Action Plan: A Deep Dive
The New York City AI Action Plan is a landmark document that sets the city apart as a leader in AI governance. Produced by the Office of Technology and Innovation, the plan outlines seven key initiatives aimed at providing centralized policy, guidance, support, and resources for city agencies. These initiatives are designed to ensure that New York City not only leverages AI for public benefit but also mitigates the associated risks.
Why This Plan Matters
The NYC AI Action Plan is not just another bureaucratic document. It represents the collective wisdom of 50 city employees from 18 agencies, as well as insights from industry, academia, and civil society. This multi-stakeholder approach ensures that the plan is both comprehensive and actionable.
The Seven Key Initiatives
The NYC AI Action Plan introduces seven new initiatives, each with a set of phased actions. These initiatives aim to provide more centralized policy, guidance, support, and resources for city agencies. They are designed to ensure that appropriate governance measures are in place across city government, that agencies are better equipped to advance their AI efforts, and that the city is working efficiently, collaboratively, and with accountability to the public.
Initiative 01: Design and Implement a Robust Governance Framework
The first initiative aims to establish a holistic, adaptable framework for AI governance. This framework acknowledges the risks of AI, including bias and disparate impact, and aims to ensure the responsible use of AI tools. The target outcomes include clear principles and definitions to guide city government's AI use and governance, engaging agencies to help define procedures, and supporting agencies with centralized AI policy, guidance, and processes.
Initiative 02: Establish a City AI Steering Committee
The second initiative aims to establish a City AI Steering Committee that will guide the city's AI governance. This committee will be responsible for setting guiding principles and definitions and providing preliminary use guidance on emerging tools. The objective is to ensure that the city's AI governance is aligned with its broader goals and values.
Initiative 03: Foster Public Engagement
This initiative aims to educate and empower the public, communicating openly about the city's progress and supporting a range of mechanisms for public input. The target outcomes are to make sure New Yorkers are well-informed about AI and the city's efforts and accomplishments, and to increase awareness among city stakeholders of residents' interests and concerns related to AI.
Initiative 04: Build AI Knowledge and Skills in City Government
The objective here is to prepare city personnel to effectively and responsibly work with and on AI. This initiative recognizes that AI literacy is critical not just for those in technical roles but also for the many public servants who use, manage, or make decisions about AI tools.
Initiative 05: Support AI Implementation
This initiative aims to support city agencies in their efforts to build and use AI, based on identified agency needs. It includes a process to provide comprehensive assistance throughout the lifecycle of AI projects and aims to develop a knowledge base to equip agencies with resources and guidance.
Initiative 06: Enable Streamlined and Responsible AI Acquisition
The objective of this initiative is to develop AI-specific procurement standards or guidance to support agency-level contracting. This ensures that procured products adhere to the city's AI principles and goals and takes steps to mitigate risks, including challenges with respect to transparency and explainability that can occur when procuring AI tools from third-party vendors.
Initiative 07: Maintain and Update Efforts
This initiative focuses on the need for an iterative approach to account for the changing technological, organizational, and social landscape, as well as emerging best practices in the field. It includes updates to a range of governance measures, public engagement efforts, and city knowledge-building resources.
New York City's AI Action Plan is a comprehensive document that outlines a structured approach to AI governance. It introduces seven key initiatives, each with specific actions and objectives, aimed at providing centralized policy, guidance, and resources for city agencies. These initiatives cover a range of topics, from establishing a robust governance framework to fostering public engagement and supporting AI implementation in city agencies.
The plan's multi-stakeholder approach is noteworthy. It incorporates insights from city employees across 18 agencies, as well as from academia, industry, and civil society. This collaborative effort aims to ensure that the governance model is both comprehensive and actionable, addressing real-world challenges and opportunities in AI.
One of the plan's strengths is its balanced focus on both the opportunities and risks associated with AI. It aims to leverage AI for public benefit while also putting in place measures to mitigate risks such as bias, discrimination, and data security concerns. This dual focus is essential for any governance model that aims to be both effective and responsible.
The NYC AI Action Plan could serve as a reference model for other cities and governmental bodies looking to navigate the complex landscape of AI governance. Its structured approach and focus on collaboration and balance offer valuable insights for how to manage the ethical and practical challenges posed by AI.