How AI Is Helping Government Companies in Cleveland Cut Costs and Improve Efficiency

Ohio's early, practical use of AI to prune its Administrative Code - removing 2.2 million words and more than 900 outdated rules - shows why AI matters for Cleveland and the state: officials project roughly $44 million in savings and 58,000 manhours reclaimed over a decade, a concrete fiscal win for cities juggling an $810 million Cleveland budget; see the Ohio AI code-cleanup for details.

That model helped prompt Sen. Jon Husted's bill to streamline the Code of Federal Regulations, which would flag redundant language for agency review, not automated deletion.

Local teams can pair policy pilots with practical training - like the AI Essentials for Work bootcamp - to build prompt-writing and tool-governance skills that turn these statewide savings into everyday municipal efficiency.

“I wrote this bill to give government a tool that helps them reduce waste and save time - and to give job creators and taxpayers a look at just how much Washington could do to get out of their way and siphon less money from their pockets.”

Ohio's regulatory reforms paired a 2022 statutory target to cut rules by 30% with the Common Sense Initiative's practical use of the AI tool RegExplorer to surface redundant, outdated, or anachronistic language for human review; a small CSI team vets AI candidates and forwards decisions to agency experts so AI flags become actionable, not automatic.

The approach has already removed roughly 2.2 million words and more than 900 obsolete rules from the Administrative Code - including 600,000 words from the building code - putting the state on track to eliminate about 5 million unnecessary words (≈one‑third of a 17‑million‑word code) and reclaim an estimated $44 million and 58,000 man‑hours over the next decade.

Local Cleveland managers can follow the same playbook - tool-assisted discovery plus agency-led review - to cut paperwork, end needless in‑person requirements, and free staff for higher‑value public service (see the state rollout and Sen.

Husted's federal proposal for scaling the method).

“AI tools don't replace government workers but can empower them.”

In March 2025 Sen. Jon Husted (R‑Ohio) introduced S.1110, the "Leveraging Artificial Intelligence to Streamline the Code of Federal Regulations Act," which would use an AI tool to flag redundant or outdated federal rules for agency review rather than automatic deletion - an approach explicitly modeled on Ohio's cleanup that officials say will reclaim roughly $44 million and 58,000 man‑hours over a decade; see Sen.

Husted's press release for the bill and the GovTrack summary for legislative status (GovTrack summary for S.1110 - Leveraging AI to Streamline the CFR (2025)).

Because the CFR already exceeds 180,000 pages, inserting an annual AI‑assisted review process could sharply reduce the time Cleveland businesses and municipal staff spend parsing overlapping federal requirements, translating Ohio's state‑level productivity gains into tangible local savings if agencies adopt the flagged removals.

The bill's design keeps humans in control - identified language is returned to the originating agency for expert decision - so Cleveland teams can prepare governance, review workflows, and pilot similar tool‑plus‑expert processes ahead of possible federal adoption.

“I wrote this bill to give government a tool that helps them reduce waste and save time - and to give job creators and taxpayers a look at just how much Washington could do to get out of their way and siphon less money from their pockets.”

Cleveland governments can cut back-office burden and speed services today by following Ohio's practical bots-and-chatbot playbook: the Ohio Benefits Program family of five bots has reviewed and processed more than 500,000 cases - saving caseworkers “over five years” of working hours - and includes targeted tools that removed ~30,000 irrelevant long‑term‑care records, processed >60% of ~4,000 incarceration alerts within 24 hours, reviews about 6,000 SNAP recertification cases monthly, and delivered same‑day Medicaid access for well over 50,000 newborns; public‑facing chatbots already deployed at state offices (BMV, SNAP/Medicaid help) and translation tools reduce wait times and call volumes while keeping humans in the loop, consistent with Ohio's IT‑17 governance approach - so what: faster benefits, fewer manual entries, and reclaimed staff time that can be redeployed to complex casework rather than data cleanup.

For planners, pilot a generative‑AI sandbox for citizen‑facing services and pair it with a human-review workflow to scale safely and transparently.

“The use cases we have today are public-facing chatbots at the BMV or for people using SNAP or Medicaid,” said Flory.

Cleveland can draw directly on Ohio's practical learning stack to train both school-to-work pipelines and municipal teams: InnovateOhio's AI Toolkit - built with aiEDU and designed as a five‑step policy and curriculum roadmap - has already logged more than 30,000 visits as a ready set of templates and classroom resources, and the State's AI in Education Strategy (Dec.

3, 2024) pushes districts to teach foundational AI and data literacy while pointing districts and employers to TechCred reimbursements for job-focused professional development; combine those resources with regional supports (ESCs and the OESCA implementation supplement) and Cleveland can run short, paid cohorts that certify prompt-writing, data‑ethics, and tool‑governance skills and recover PD costs through TechCred - so what: a visible demand (30k+ visits) plus state funding routes means city HR and department leaders can upskill 50–100 frontline workers with little net cost and redeploy saved hours into higher‑value citizen services.

See the state toolkit and strategy for downloads and templates.

“The more resources we place in the hands of school leaders, educators, families, and students, the better positioned we will be to use AI tools thoughtfully and responsibly.”

Ohio's AI-assisted code cleanup delivers concrete savings Cleveland can measure: the state has already removed roughly 2.2 million words and more than 900 obsolete rules and projects a total cut of about 5 million words, with an estimated $44 million in taxpayer savings and 58,000 man‑hours reclaimed over a decade - numbers that translate into real staffing capacity and budget relief Cleveland can repurpose to frontline services.

Policymakers should note the scale: the federal Code of Federal Regulations tops more than 180,000 pages, which is why Sen. Jon Husted's bill to apply a similar AI review to the CFR frames the tool as a way to flag redundancies for agency review rather than automate deletions (Sen. Jon Husted's bill to streamline the Code of Federal Regulations).

Practical pilots that couple AI discovery with clear human review and governance - recommended in industry guidance - are the low-risk path to convert those projected savings into faster permitting, fewer manual reviews, and more hours for complex casework (Deloitte report on reducing government red tape).

“I wrote this bill to give government a tool that helps them reduce waste and save time - and to give job creators and taxpayers a look at just how much Washington could do to get out of their way and siphon less money from their pockets.”

Responsible Cleveland adoption depends on the guardrails Ohio already wrote into law: the State's IT‑17 policy authorizes AI use while requiring statewide planning, procurement, security, privacy, and governance resources - from an AI Governance Framework and Council charter to procurement checklists - so municipal leaders don't have to invent oversight from scratch (Ohio IT‑17 policy: Use of Artificial Intelligence in State of Ohio Solutions).

At the same time, national experts urge a risk‑based, human‑centered posture - balancing innovation with bias mitigation, workforce impacts, and privacy - and the U.S. Chamber's Commission recommends proportional rules, cross‑agency collaboration, and human-in-the-loop validation before automated action (U.S. Chamber AI Commission report on governance and human-in-the-loop validation).

Real litigation risks reinforce those cautions: a 2025 class action alleges a transcription service recorded meetings without participant consent, highlighting why Cleveland pilots must log AI inputs, require affirmative consent for automated notetakers, and pair every citizen‑facing system with clear human review and procurement checklists to prevent trust‑eroding incidents (NPR coverage of the Otter.ai transcription lawsuit and consent concerns).

The takeaway: adopt IT‑17 templates, require explicit consent and audit trails, and make human sign‑offs a non‑negotiable step so saved staff hours aren't lost to legal and ethical remediation.

“There was a person with judgment that was looking what the AI algorithms were doing and making the final decision. But that human's job was so much easier because of what the AI algorithms provided upfront...” - Dr. Lara Jehi, Cleveland Clinic

Start small, govern first: adopt Ohio's IT‑17 templates for procurement, privacy, and lifecycle governance, then mirror the Multi‑Agency AI Council model by assigning legal, privacy, and IT leads to approve use cases and run a centralized “sandbox” for citizen‑facing pilots; see the Ohio IT‑17 summary for the policy framework and the Nucamp sandbox design for a tested pilot approach.

Run one tightly scoped pilot (customer‑service chatbot or benefits‑triage) that preserves human review at every decision point, measure throughput and accuracy (HHS's chatbot first resolved ~20% of calls with a 65% full‑implementation projection), and use those metrics to justify scale‑up and TechCred‑eligible staff training drawn from state toolkits.

Build a simple approval workflow: request → risk review (privacy/bias) → sandbox test → human‑in‑the‑loop pilot → audit and repository of approved use cases; this sequence turns abstract policy into a repeatable municipal process that protects residents while freeing staff hours for complex work - so what: a single, well‑governed pilot can prove value fast and create a reusable playbook for other departments.

“Keeping ‘humans in the loop' is a key feature of AI solution development in Ohio.”

Cleveland's best AI case studies come from health care where practical pilots already improve speed and diagnostic power: the Cleveland Clinic is leading regional efforts to “accelerate and enhance the ethical use of AI” across diagnostics, triage, chatbots, and research (Cleveland Clinic AI initiatives in healthcare), while the Center for Diagnostics and Artificial Intelligence (CDAI) - founded in 2022 - centralizes image analytics and R&D to move tools into clinical workflows (CDAI center image analytics and R&D).

Concrete wins include AI-assisted stroke triage that reviews scans faster than humans and radiology tools that act as a “second pair of eyes,” and a new NIH R37 MERIT award ($2.6M) to test a federated‑learning gastric‑cancer screening model across institutions - an approach designed to identify high‑risk patients for targeted endoscopy and speed early treatment (NIH MERIT award for federated gastric cancer screening AI).

So what: Cleveland's integrated research-to-clinic pipeline proves AI can shave critical minutes in urgent care and create validated, shareable models that city health and government teams can pilot with clear governance to lower cost and improve outcomes.

“The AI is cutting down precious minutes by being the first and fastest agent in this process to review those images.” - Po‑Hao Chen, MD

Looking ahead, Cleveland can move from isolated pilots to city‑wide impact by running controlled generative AI sandboxes, pairing them with cross‑sector partnerships, and protecting institutional know‑how: a tightly scoped sandbox lets teams safely test citizen‑facing services before full launch (generative AI sandbox pilot design for municipal services in Cleveland), while collaborations - like Cleveland Clinic–municipal partnerships - can accelerate responsible pilots by bringing clinical governance and research rigor to public workflows (Cleveland Clinic municipal partnership for AI-driven public health workflows).

Adopted tools could streamline grant writing and service descriptions, but teams must guard against knowledge loss: generative drafting can boost output yet risk eroding institutional memory unless final edits and provenance are preserved (AI drafting tools for municipal grant writers in Cleveland).

So what: a single, well‑governed sandbox pilot - backed by a health‑sector partner - can prove efficiency gains while keeping humans firmly in charge of institutional expertise and public trust.

Cleveland leaders should move from strategy to action with three tightly linked steps: (1) adopt Ohio's IT‑17 governance templates and run one short, human‑in‑the‑loop pilot (customer‑service chatbot or benefits‑triage) inside a controlled generative‑AI sandbox so flagged outputs get agency expert review; (2) require bias mitigation and audit trails from day one - use the published bias‑mitigation checklist as a standard for model development and monitoring (Published bias-mitigation checklist (NCBI)); and (3) pair pilots with workforce training so reclaimed hours translate to better services, not layoffs - start with a cohort using practical prompt‑writing and governance training such as the AI Essentials for Work syllabus to certify staff who will manage and audit systems (AI Essentials for Work syllabus).

The payoff is concrete: Ohio's cleanup projects an estimated $44M saved and 58,000 man‑hours reclaimed over a decade - track the same throughput and error metrics locally to justify scale‑up and preserve public trust.