Unsure when legal light ends where you’re hunting? Curious about the bag limit for grouse in a new area? Need to know the minimum caliber for deer hunting where you live? Those, plus a thousand other questions about hunting regulations, are sometimes so daunting and head-scratching that they keep beginning hunters out of the field. Or they throw up barriers to hunting a new place, out of fear that you might run afoul of an unfamiliar rule or regulation.
A new digital assistant can help hunters cut through the dense and sometimes confusing complexity of hunting regulations. It’s called Scout, an experimental project of the International Hunter Education Association that uses artificial intelligence to answer questions that hunters have about where, how, when, and how much they can legally hunt.
The idea behind the tool, which offers A.I.-derived answers for common — and uncommon — questions about hunting rules and regulations in all 50 states, is to help hunters be more knowledgeable about the myriad rules that govern hunting, says Jae Ellison, director of education for IHEA. Knowledge boosts confidence, and confidence boosts participation.
Read Next: Public Approval for Hunting and Recreational Shooting Continues to Decline in the U.S.
“For new hunters, or hunters who are exploring new types or areas of hunting, regulation complexity can be a daunting barrier,” says Ellison. “A lot of what the R3 (hunter recruitment, retention, and reactivation) movement does is remove barriers, and this is an effort to remove or at least reduce a known barrier.”
Many states have tried to simplify the complexity of their regulations or reduce their length, but it’s a hard issue to resolve, as a 2023 study from University of Montana researchers noted. Most regulations, also called season proclamations, annuals, or guidebooks, are required to have legally mandated information printed every year. Most also have granular hunting-district information, or detailed exceptions to statewide rules. Consequently, most regulation books run over 100 pages of extremely fine print and sometimes perplexing language.
In its testing phase, which is likely to continue through the summer, IHEA is asking users to try to stump Scout by asking hard-to-answer or obscure questions.
“We’re putting Scout out there, for free use with no strings attached or membership required, in order to be a widely used tool,” says Ellison. “In this testing phase, we want people to try to break it. The more people who utilize it and ask A.I. to interact with the data, the better it’s going to get.”
Photos by Andrew McKean
As an administrator of the program, Ellison can see questions come in — sometimes as many as 800 in a day — and says most relate to bag limits, weapons restrictions, and specific hunting area regulations. But he says there are some doozies in the mix.
“People try to fool Scout by using slang,” he says. “They’ll ask how many bucks I can shoot, or how many hawgs, or pigs, or they’ll use euphemisms for harvest or killing. Like how many deer can I slay, or blast, or whack. But those are all good inputs for Scout to learn about.”
Scout tries to cut through the lines and pages of tedious information to answer specific questions of users. To access its library, users first select a state they’re interested in hunting. Then they type in a query, and Scout combs the specific state’s guidebook for keywords, using what’s called a “large-language model” to deliver answers in conversational English that most users understand. If Scout is stumped on a question, it will ask users to rephrase it.
The tool relies on what’s called a closed database. Unlike many A.I. tools, commonly called chatbots, that scrape the entire internet for answers, Scout can only answer questions it derives from updated hunting regulations.
“It cannot provide an answer outside the data we give it,” says Ellison. “Every time it provides an answer, it follows with a statement to the effect ‘If you have more questions, follow up with your state agency,’ and we also provide the actual rules and regulations where Scout found the answer. The idea is to add validity to the answer, but to encourage users to dig deeper into the regulations and keep learning.”
The closed database is intended to avoid what’s called A.I. “hallucination,” that occurs when chatbots return information that satisfies users, rather than sometimes more difficult or nuanced answers that can be unsatisfying to users.
“We require every answer to exceed a confidence threshold prior to giving it to a user,” says Ellison. “If we can’t meet that confidence threshold, we tell the users that we don’t know the answer. It’s one of the guardrails we’ve built into the system. Our system is specifically designed to not hallucinate, and we default to a ‘we don’t know’ answer as a precaution against giving wrong or misleading information.”
Early concerns about Scout include those who suggest that an A.I. tool will provide speculative or interpretive information about regulations. But Ellison says the system is intentionally engineered to provide text-based answers, not interpretations.
Other critics have questioned liability if Scout provides users incorrect or misleading information that results in a wildlife infraction.
“People have asked who is responsible if someone goes out in the field and does something that’s outside the bounds of a regulation because they were given incorrect information,” says Ellison. “I’ve even had people ask if they can print out the answer that Scout gives them to take in the field in case they’re stopped by a game warden. We are in the testing phase precisely in order to minimize those possibilities.”
But Ellison says by far the most feedback has come from state agencies who are excited that someone is working on an issue that addresses a large and intractable problem.
“We’ve been talking for a long time in this industry about ways of reducing regulation complexity, but this is one of the first tools that leverages technology to potentially solve the problem, and I think that’s noteworthy.”
