Title: Embedding AI Safely in Government Operations: Supporting Staff and Improving Citizen Outcomes

When I first read the Capita piece on building trustworthy AI-enabled services for the public sector, my immediate thought wasn’t about algorithms or procurement cycles—it was about the corner bodega on Flatbush Avenue in Brooklyn, where the owner still uses a paper ledger alongside a tablet for SNAP benefits. That tension between legacy systems and new technology isn’t abstract here; it’s the daily reality for thousands of New Yorkers interacting with city services. The source material rightly emphasizes embedding AI safely into live operations while supporting the people who deliver services—a goal that feels both urgent and deeply human in a place as layered as New York City.

New York’s approach to AI governance has been shaped by years of incremental steps, from the 2017 Automated Decision Systems Task Force to the more recent AI Strategy released under Mayor Adams’ administration. What’s different now, as highlighted in the GAO report on federal AI efforts, is the shift from principles to operational requirements—agencies aren’t just asked to consider fairness and transparency; they’re now guided by concrete frameworks for implementation and oversight. The OMB memorandum M-24-10 reinforces this, establishing new agency requirements for managing AI use that align closely with what New York City has been piloting through its Office of Technology and Innovation (OTI).

What makes this moment particularly salient for Brooklyn and the outer boroughs is how AI is being positioned not as a replacement for workers but as a tool to reduce administrative burden. Consider the Human Resources Administration (HRA), which processes over three million benefits applications annually. Pilot programs using natural language processing to triage incoming documents or chatbots to answer routine eligibility questions aren’t about cutting staff—they’re about freeing up caseworkers to spend more time in face-to-face conversations with residents navigating complex life challenges. This aligns directly with the source material’s focus on “supporting the people who deliver services” while improving outcomes for citizens.

Yet the path forward requires careful attention to local context. In neighborhoods like Bushwick or Sunset Park, where language access and digital literacy vary widely, any AI-enabled service must be designed with real-world usability in mind. The AI Guide for Government notes that successful implementation depends on understanding the types of problems best suited for AI—high-volume, rule-based tasks with clear outcomes—rather than forcing technology onto inherently human interactions. For New York City, this means prioritizing use cases like streamlining parking violation appeals or automating routine inspections for the Department of Buildings, where consistency and speed matter most, while keeping human judgment central to areas like child protective services or homeless outreach.

Historically, New York has struggled with technology rollouts that didn’t account for frontline worker input—remember the early struggles with NYC.gov’s redesign or the initial payroll system failures at the Department of Education. The current emphasis on oversight and advisory groups, as described in both the GAO report and the OMB memo, offers a chance to break that cycle. By involving DC 37 union representatives, OTI’s AI Advisory Council, and community board tech liaisons in the design phase, the city can build systems that workers actually wish to use, not just tolerate.

Given my background in urban policy and civic technology, if this trend impacts you in Brooklyn—whether you’re a caseworker at HRA, a compact business owner dealing with BIDS compliance, or a resident trying to access SNAP or Medicaid—here are the three types of local professionals you need to know:

• Public Interest Technologists: Look for individuals or small firms with proven experience working directly with NYC agencies like OTI or the Mayor’s Office of Data Analytics. They should understand not just AI/ML fundamentals but also the nuances of city procurement (PPP rules), FOIL considerations, and how to design tools that survive real-world field testing in places like a SNAP center in East New York or a HRA job center in the Bronx.
• Civic UX Researchers Specializing in Public Services: Seek professionals who’ve conducted usability studies with diverse populations—particularly older adults, limited-English speakers, or people with disabilities—in actual service environments. Their portfolios should include work with HRA, DSS, or NYCHA, demonstrating they know how to test AI interfaces where literacy levels vary and trust in government is often low.
• Algorithmic Impact Assessment Consultants: These experts help agencies evaluate whether an AI tool might create disparate impacts before deployment. In New York, this means familiarity with Local Law 89 of 2021 (which requires bias audits for automated employment decision tools) and the ability to adapt those principles to public-facing services like benefits determination or housing inspections. They should be able to reference specific frameworks used in OTI’s pilot reviews or the NYC Algorithmic Management and Oversight (AMO) policy.

Ready to find trusted professionals? Browse our complete directory of top-rated public interest technologists experts in the Brooklyn area today.