New York sets out an action plan on artificial intelligence

New York has released an Artificial Intelligence Action Plan, which it claims is the first of its kind for a major US city.

The document sets out how the city will develop a framework for agencies to evaluate AI tools and risks, help city government employees build AI knowledge and skills, and support the implementation of these technologies.

The city is also piloting an AI chatbot through the MyCity Business site, which is part of the MyCity initiative to streamline digital services. The use of the chatbot aims to help business-owners to access information from more than 2,000 city web pages.

“While artificial intelligence presents a once-in-a-generation opportunity to more effectively deliver for New Yorkers, we must be clear-eyed about the potential pitfalls and associated risks these technologies present,” said Mayor Eric Adams. “I am proud to introduce a plan that will strike a critical balance in the global AI conversation – one that will empower city agencies to deploy technologies that can improve lives while protecting against those that can do harm.”

An audit released in February by the New York state comptroller found that “ad hoc and incomplete approaches to AI governance” in New York City were not adequate to ensure that the technology’s use was transparent, accurate, and unbiased.

Key actions

The city’s new AI plan outlines 37 key actions – 29 of which are set to be started or completed within the next year. These include establishing a framework for AI governance that acknowledges risks such as bias and disparate impact. The city will create an external advisory network to consult with stakeholders across sectors around the opportunities and challenges posed by AI. It will also establish AI-specific procurement standards or guidance. An annual AI progress report will be published.

“With the creation of the Office of Technology and Innovation, Mayor Adams has set a new standard for how to enable smarter tech adoption across city government,” said Chief Technology Officer Matthew Fraser. “I am proud of my colleagues at OTI who have put forward a thoughtful approach that will guide how the City of New York supports its agencies through responsible adoption and shape the way cities around the world approach AI.”

Last week, Jennifer Gutiérrez, chair of New York City Council’s technology committee, also introduced a bill that would create an Office of Algorithmic Data Integrity to oversee AI in New York. If established, the office could act as an ombudsman for automated decision-making systems and would be responsible for evaluating AI systems prior to their implementation by the city.

The new plan is not New York’s first attempt to govern AI. The former administration under Mayor Bill de Blasio published an AI strategy in 2021.

In 2018, the city assembled an automated decision systems taskforce which was hailed as the first of its kind in the US. This resulted in a report of recommendations, including to create a centralised resource, but it also received some criticism related to transparency and ultimate impact.

The city also created the position of Algorithms Management and Policy Officer in 2019 but it was discontinued when Mayor Adams took office in 2020 and consolidated various agencies under the Office of Technology and Innovation.

In January this year, OTI advertised for a new position of Director of Artificial Intelligence and Machine Learning and the recruitment process is ongoing.

‘Deep fake’

Concerns have previously been raised about New York’s use of AI such as the police department’s facial recognition programme and robot dog deployment.

During the announcement of the new AI plan, Mayor Adams disclosed that artificial intelligence is employed in robocalls that feature his voice to convey messages in languages he doesn’t actually speak, including Mandarin and Yiddish. The calls are about issues such as recruitment or community events and do not disclose that AI is being used.

“The mayor is making deep fakes of himself,” said Albert Fox Cahn, executive director of the Surveillance Technology Oversight Project. “This is deeply unethical, especially on the taxpayer’s dime. Using AI to convince New Yorkers that he speaks languages that he doesn’t is outright Orwellian. Yes, we need announcements in all of New Yorkers’ native languages, but the deep fakes are just a creepy vanity project.”

Responding to potential concerns, Mayor Adams said: “These are part of the broader conversations that the philosophical people have to sit down and figure out, you know: is this ethically right or wrong? I got one thing: I’ve got to run the city, and I have to be able to speak to people in the languages that they understand, and I’m happy to do so. And so to all, all I can say is ‘ni hao’.”

According to the AI plan, the city will use its existing public reporting of algorithmic tools  as well as additional research to create a typology of AI projects. This is scheduled to be complete within six months.

Within the next 12 months, it will initiate the development of an AI Risk Assessment and Project Review Process to enable the analysis of existing and proposed AI projects.