Massachusetts rolls out ChatGPT for state workers

On February 13, 2026, Governor Maura Healey and Lt. Governor Kim Driscoll announced that Massachusetts will roll out a ChatGPT-powered AI Assistant across the state’s executive branch.

What happened

The administration claims that the roll out's goal is to help government staff work faster and more efficiently, especially on routine paperwork, while still protecting sensitive information. Massachusetts plans to deploy the tool in phases, starting with employees in the Executive Office of Technology Services and Security, then expanding to other secretariats and agencies over the coming months.

State officials say the rollout follows a competitive procurement process and includes a contract with OpenAI to provide ChatGPT at scale. The system will run in a walled-off secure environment intended to protect state data, and the state says employee chat inputs will not be used to train public AI models. Use of the assistant will be governed through terms set by the TSS Privacy Office and policies that the state says will be updated over time.

What was said

According to Governor Healy in the press release, “AI has the potential to transform how government works, which is why we’re excited to partner with OpenAI on this AI Assistant, which will ensure a safe and secure environment for employees and improve their ability to deliver better service to the people of Massachusetts. I’m grateful to Secretary Snyder and Secretary Paley for their innovative leadership on this and to the state workers who are eager to embrace this new technology.”

The bigger picture

In the same state, House lawmakers advanced election-year legislation aimed at AI-generated audio and video that could mislead voters or harm candidates. The proposal would restrict the distribution of materially deceptive synthetic media close to elections and require disclosures on AI-generated political communications.

The bill, when viewed alongside the executive-branch ChatGPT rollout, shows the state pushing AI on two fronts. One adopts AI internally for productivity under privacy and security guardrails, and another is tightening rules for public-facing political uses where deception risks are higher.

Why it matters

Organizations like hospitals, clinics, Medicaid vendors, and government contractors will feel a spillover effect because state agencies run on high-volume communications and documentation. impact shows up in three places. Contracting and coordination pressure rises as public health agencies, Medicaid administrators, and state-facing vendors start expecting faster turnaround on narratives, audits, incident reports, and compliance documentation, pushing providers and business associates to adopt similar tools and governance just to keep pace. Risk then shifts to the seams between systems.

NYU Langone’s health-system GenAI rollout, discussed in JAMIA, describes staff using the tool for “drafting email responses,” which implies downstream use in email workflows rather than staying inside the AI environment.

A separate clinical LLM workflow study published in JMIR Medical Informatics describes the mechanics of using a web LLM as “web-based interface with manual copy-paste of clinical documents,” showing how day-to-day GenAI use often depends on manual text transfer between systems.

Taken together, those quotes support the point that a walled-off AI layer can still leave exposure at the handoff, because staff moves text between the AI tool and channels like email.

See also: HIPAA Compliant Email: The Definitive Guide (2026 Update)

FAQs

Does the U.S. have one comprehensive federal AI law?

There is no AI Act that covers everything nationwide; federal policy is split across sector laws, targeted bills, agency enforcement, and executive-branch directives.

What federal AI-related law is already on the books (not just a proposal)?

One clear example is the TAKE IT DOWN Act, which criminalizes certain nonconsensual intimate images and requires covered platforms to run a notice-and-removal process.

Who enforces the platform requirements under the TAKE IT DOWN Act?

The Congressional Research Service says the law creates new requirements for covered platforms that the Federal Trade Commission would enforce.