The City of Fort St. John is outlining how transparent and verified outputs produced by artificial intelligence may, and may not, be used by staff under a new policy
As the technology continues to develop and become integrated into software used by the city, staff said it was important for council to determine what appropriate use will look like.
“The city (must) establish clear policy direction to ensure that any use of AI aligns with public sector values, legislative requirements, and council’s expectations regarding accountability, transparency, and trust. Implementing a policy that provides guidance on how staff can ethically and responsibly use AI as a support tool is timely,” wrote Fort St John city staff in their report for the city council.
The policy allows the use of AI as long as it operates within a closed program. Because data entered into the system may be confidential or related to city matters, the policy states the program must keep the information private and limits the capabilities allowed.
Systems used by staffers must be ethical and transparent and will be “governed and monitored through documentation and risk review,” according to the AI policy. Its use must also better the delivery of a message without replacing human judgement.
Final calls will never be made by computers in Fort St John, and any outputs produced by them should be reviewed by a “qualified individual. AI generated content shall be treated as a draft or preliminary resource, not a final authority,” the newly approved policy reads.
“Each AI use case must have clearly defined objectives and rely on approved, accurate and relevant data sets. Staff must ensure the validity and quality of data used, recognizing that inaccurate data produces unreliable outputs,” it continues.
Annual reviews of the AI policy will take place under the document’s continuous improvement section. Reviews may occur more often if needed to keep up with legislation from higher levels of government or other guidelines.
