Using artificial intelligence (AI) as a NSW Government public servant comes with certain responsibilities.
To know how to safely use AI in your systems, you must first identify if you’re using AI. Once you’re clear on this, you will be better able to identify risks and ensure compliance with assurance frameworks in NSW Government, including the NSW AI Assessment Framework (AIAF).
This article complements the learning module for NSW Government public servants, called Identifying AI. Check it out for more detailed information on this topic.
What is AI?
It can be difficult to identify and agree that AI is being used within a system, especially because there is not yet a universally accepted technical or legal definition of AI. Where there is the possibility of AI use, it’s suggested that the mandatory AIAF be applied.
The NSW AI Assessment Framework says,
“AI is the ability of a computer system to perform tasks that would normally require human intelligence, such as learning, reasoning, and making decisions.”
How to know if your systems use AI
Different AI systems have different characteristics, but it’s common to see the following exhibited:
Probabilistic vs Deterministic: AI systems make decisions with uncertainty and are data-driven, whereas traditional systems follow predetermined rules and produce the same result for identical inputs.
Learning vs Preprogrammed Behaviour: AI systems learn and improve over time, while traditional systems need manual updates to change behaviour.
Data-Driven vs Rule-Based: AI systems rely on data to make predictions or decisions, while traditional systems execute predefined rules and workflows without adapting.
Complex Pattern Recognition vs Simple Logic: AI systems recognise complex patterns within data, while traditional systems perform tasks based on straightforward logic.
Dynamic Decision-Making vs Static Workflows: AI systems adapt their decisions based on new data and changing conditions, while traditional systems follow fixed workflows that do not evolve.
The Identifying AI module has a helpful questionnaire that helps differentiate Non-AI use cases from AI use cases.
What’s next?
If you think your systems may use AI, the AIAF must be applied to ensure safe and responsible use of AI.
If you’re not sure if you are using AI or what your responsibilities may be, seek advice from your Agency Chief Information Officer, Digital Officer or Data Officer.
Learn more about the mandatory AI Assurance Framework.