I founded Easie in 2018 to help organizations put AI to work on real problems, not theoretical ones.
Our platform, EasieOps, pulls documents and PDFs from hundreds of data sources, extracts structured data using fine-tuned AI models and pushes that data into integrated downstream systems.
So I have a clear view of what AI looks like in practice. It is not science fiction. It is a tool that automates the essential parts of work that keep businesses moving, and it should be allowed to do just that. But because it already plays a role in high-consequence decisions about lending, housing, criminal justice and employment, we need to treat it with the seriousness those areas deserve. That is why I support SB 2, Connecticut’s proposed legislation on artificial intelligence.
As someone who works in AI implementation every day, I see this bill as a practical framework. It creates space for responsible progress and positions Connecticut to lead on one of the most important technological shifts of this century.
SB 2 is designed for the U.S. business environment. It does not impose blanket rules. It focuses on high-risk AI systems, where harm is more than hypothetical. It includes phased implementation and flexibility for businesses to adapt.
The largest companies building AI already follow many of the best practices put forth by the bill. Microsoft, Google and Amazon conduct bias audits, risk analysis and impact assessments. These are not bureaucratic hurdles. They are common sense for any system making consequential decisions, and these measures can protect businesses from unforeseen complications.
Even OpenAI has called for clear standards. In their words, just as drivers are expected to follow common-sense rules to keep roads safe, developers have a responsibility to follow standards that keep the AI roads safe. That encourages investment, competition and greater freedom for everyone. Smart rules make it easier to build, test and launch better systems by setting clear boundaries.
EasieOps, for example, has helped clients reduce manual entry by up to 99 percent while improving accuracy and auditability. It is already reshaping how work gets done in compliance-heavy industries. AI tools are powerful and the future for them is bright, but as such, they must be handled with responsibility.
SB 2 also emphasizes transparency. When people engage with an AI system instead of a human, they should know. When a system helps decide whether someone gets a job, a loan or a criminal sentence, there should be a way to review what happened. This fosters trust and supports adoption and growth.
SB 2 rightly phases in requirements and supports education for small businesses. Small businesses typically automate internal tasks, not high-stakes decisions. Regulations should reflect that difference. A startup should not face the same reporting burdens as a multinational corporation.
Thresholds based on revenue or headcount, along with clearer definitions of high-risk AI, and carve-outs for low-impact systems could help refine the legislation to further meet its goals.
Free tools like IBM’s AI Fairness 360 toolkit, Canada’s Algorithmic Impact Assessment Framework and the Ontario Human Rights Commission’s AI assessment guide help small companies comply without legal teams. Policymakers should promote these resources.
The sponsors of SB 2 have built a framework grounded in how AI is actually used. Reasonable accountability that supports AI innovation is within reach. Artificial intelligence, combined with sensible guardrails, can foster the kind of innovation that solves problems, works across industries and builds trust along the way.
Smart regulation strikes the balance between social responsibility and innovation, and SB 2 does just that.
Rock Vitale is the Founder and CEO of Easie and serves as Chair of the Technology Council for the Greater New Haven Chamber of Commerce.