Presenter tells Montana task force 'AI is already here,' outlines benefits and risks
Loading...
Summary
At a Feb. 12 task force meeting, a speaker summarized AI history, practical uses for Montana (agriculture, rural health, fraud detection) and risks including bias, energy use and legal questions about liability and 'AI personhood.' Members emphasized human-in-the-loop safeguards and statutory clarifications.
Mr. Pittman told the Blockchain and Digital Innovation Task Force on Feb. 12 that artificial intelligence is no longer hypothetical: "AI is not coming. It's already here," he said, describing rapid recent adoption and a 70‑year research arc that culminated in the current proliferation of generative models and specialized systems.
Pittman framed the technology by category — chatbots and assistants, image and voice generators, specialized medical and legal systems, and "invisible" infrastructure that shapes what people see and what decisions are made — and offered concrete Montana-relevant examples. He said precision‑agriculture systems can boost yields while cutting inputs, that AI helps radiologists read images, and that fraud detection systems helped the IRS identify what Pittman cited as $9.1 billion in fraud in fiscal 2024.
The presenter also outlined notable risks: workplace disruption, legal uncertainty about liability when AI agents hire or transact, documented bias in some systems, privacy concerns from large training datasets, deep fakes and high energy use from model training. He used standard thought experiments — including Nick Bostrom's "paperclip maximizer" — to explain specification and incentive‑alignment problems, and raised recent debates about whether autonomous AI agents that earn income create new tax and legal questions.
Task force members responded by discussing Montana's existing "right to compute" framework and when the state should act. Co‑chair Lammers and others urged balancing innovation and protection, noting prior legislation that requires human oversight for critical infrastructure and citing examples such as child‑safety or fraud as compelling government interests. Members stressed stakeholder engagement, recommended clarifying liability standards, and flagged the need to avoid overbroad state rules that could be preempted by federal action.
Pittman and members also compared international and state approaches: he cited the EU's AI Act taking effect in August 2024 and the patchwork of state laws in the U.S., and he encouraged the task force to consider practical, targeted statutes that preserve rights while addressing concrete harms.
The task force did not adopt formal policy at the meeting; chairs directed follow‑up discussion into workgroups and the July 1 report to the Economic Affairs Interim Committee.