Nambucca Valley Council formulates artificial intelligence policy documents


NAMBUCCA Valley Councillors voted to install a draft policy for artificial intelligence (AI) governance at last week’s general meeting on Thursday 31 October.

Council ICT manager Darren Moulds addressed what he sees as the need for council to address rapid changes in the field of AI which have the potential to improve efficiency and deliver improved outcomes in a range of ways, from taking-on mundane tasks to generating and gathering ideas.

Macksville Ex Servicemans ClubAdvertise with News of The Area today.
It’s worth it for your business.
Message us.
Phone us – (02) 4981 8882.
Email us – media@newsofthearea.com.au

He presented a draft version of the Artificial Intelligence Governance Policy that was subsequently adopted by Council.

“The emergence and rapid advancement of artificial intelligence provides opportunities for Local Government to operate more effectively and efficiently,” Mr Moulds reported to Council.

“However, there is the potential for AI to be poorly used or applied and there are also ethical questions when it comes to the use of AI for some tasks,” he said.

Mr Moulds reminded the councillors that while simpler forms of AI technology (e.g. spell checking, facial recognition, chatbots) have been around for decades, in recent years the level of sophistication and ‘generative learning’ capability of AI has taken a quantum leap.

While considering these benefits, governments in particular are advised to be cognisant of the risks associated with AI.

“These risks include the data risk posed by Council’s AI requests being stored in places and by companies about which little is known.

“Furthermore, councils need to consider the culpability for mistakes made by AI and responsibilities for its uses as well as the potential unknowing use of copyrighted, personal, confidential or commercially sensitive information,” he said.

To this end, Council’s AI Governance Policy limits the use of AI to “low-risk situations” such as brainstorming ideas or options, or to do initial drafting of content, emails, and reports.

The new policy states that AI must never be the final decision-maker on Council activities or services and users must ensure to comply with Council’s information security, records and privacy management policies.

The policy highlights the need for transparency from Council about which services and activities have been generated by AI tools and states no classified or personal information should ever be inputted into AI tools.

“We still need that ‘human element’ across the top to ensure that what (AI) it is spitting out is correct,” Council General Manager Bede Spannagle stated.

“Staff are using it already,” he said.

“(Staff currently) put the information in first, use the tool to do the ‘leg work’ and then the council officer will make the decision at the end of the day,” he said.

Councillor Jenvey expressed concerns that use of the technology might lead to staff redundancies.

“(For example) I’d hate to see (the council’s scribe) out of a job or something because AI can type better and write a better report,” she said.

Mr Spannagle responded jokingly that AI would not offer the same customer service skills as the staff member.

Cr Jenvey asked if a record could be kept of impacts on staff due to AI or impacts from decisions made by AI.

Mr Spannagle said that although he worried that such a register might quickly become burdensome, he would take the question on notice and report back with his comments at a later date.

Mr Moulds said he hopes the policy will address the use of AI tools by Council staff in a way that embraces the potential positive benefits of the technology, but also ensures that its utilisation aligns with Council’s legal, ethical, and community standards.

He predicted that future iterations of the policy would quickly become necessary in the rapidly changing field.

By Ned COWIE

Leave a Reply

Top