Artificial Intelligence Use Policy

Our approach to artificial intelligence

Southern Foodservice recognises that artificial intelligence (AI) can improve the way we work, communicate, analyse information and serve our customers.

We are committed to using AI responsibly, securely and transparently. AI is used as a tool to support our people and our business — not as a replacement for appropriate human judgement, accountability or decision-making.

Our approach is guided by Australian Government recommendations for the safe and responsible adoption of AI and by recognised industry practices relating to privacy, information security, risk management and human oversight.

Our principles

When Southern Foodservice uses AI, we aim to follow these principles:

• Responsible use. AI must be used for legitimate business purposes and in a manner consistent with our values, policies and legal obligations.
• Human accountability. People remain responsible for decisions and outcomes involving AI. We do not treat AI-generated information as automatically accurate or authoritative.
• Privacy and data protection. Personal, confidential, commercially sensitive and customer information must be handled appropriately when AI systems are used.
• Security. We consider the security and data-handling practices of AI services before approving them for business use and apply appropriate controls based on the information and activities involved.
• Accuracy and reliability. AI-generated content can contain errors, incomplete information or misleading results. Important AI-generated information must be appropriately reviewed and verified before it is relied upon.
• Fairness. We seek to use AI in ways that avoid unfair discrimination, inappropriate bias or unreasonable impacts on individuals.
• Transparency. Where appropriate, we will be open about the use of AI, particularly where AI materially contributes to content, services or processes that may affect our customers, suppliers, employees or other stakeholders.

How we use AI

Southern Foodservice may use approved AI technologies to assist with activities including:

• drafting and improving business communications;
• creating or assisting with marketing and promotional content;
• summarising and organising information;
• researching and analysing information;
• assisting with reports, presentations and documentation;
• improving internal processes and productivity;
• supporting data analysis and business insights;
• assisting with technology, software and automation;
• supporting customer and supplier communications; and
• other approved business activities.

The use of AI does not remove the responsibility of our employees to exercise professional judgement and review their work.

Human oversight

Southern Foodservice maintains human oversight of AI use.

The level of human review required will depend on the potential impact of the activity. AI-generated information relating to significant business, financial, legal, employment, safety, customer or supplier decisions should not be relied upon without appropriate human assessment and, where necessary, independent verification or professional advice.

Where an AI system produces an unexpected, inappropriate or potentially harmful result, our people are expected to stop relying on the output and escalate the matter where appropriate.

Accuracy of AI-generated information

AI systems are probabilistic technologies and may occasionally produce information that is inaccurate, outdated, incomplete or fabricated.

For this reason, Southern Foodservice requires appropriate verification of AI-generated information before it is used for important business purposes or published externally. AI-generated content should not be presented as established fact where its accuracy has not been reasonably verified.

Privacy, confidential information and data

Southern Foodservice takes the protection of information seriously.

Our employees must use approved AI systems in accordance with our privacy, confidentiality, cybersecurity and information-handling requirements. Information entered into an AI system must be appropriate for that system and its approved business purpose.

Employees must not enter information into an unapproved AI service where doing so could expose:

• personal or sensitive information;
• customer or supplier confidential information;
• passwords, authentication credentials or security information;
• confidential financial information;
• commercially sensitive information;
• confidential contracts or legal material;
• proprietary business information; or
• information that Southern Foodservice does not have authority to disclose.

Where AI services process business information, we consider factors such as security controls, access controls, data handling, retention and privacy protections as part of determining whether their use is appropriate.

Approved AI services

Southern Foodservice may establish approved AI platforms for business use.

AI services must be assessed according to the nature of their intended use and the information they may process. Employees should not adopt AI tools for business purposes simply because they are publicly available or free to use. Higher-risk AI applications may require additional assessment and approval before they are introduced.

Security

We apply a risk-based approach to AI security.

Depending on the system and its intended use, this may include consideration of:

• access and authentication controls;
• data encryption;
• information retention and deletion;
• privacy protections;
• vendor security practices;
• system monitoring;
• data access;
• incident response;
• integration security; and
• third-party and supply-chain risk.

Where third-party AI providers are used, we seek to understand their relevant security, privacy and data-management practices before using their services for appropriate business information.

AI and decision-making

AI may assist our employees by providing information, analysis or recommendations.

However, AI should not independently make significant decisions about individuals where appropriate human review is required. Decisions with significant consequences for employees, customers, suppliers or other individuals should include meaningful human oversight appropriate to the circumstances.

Fairness and responsible outcomes

Southern Foodservice recognises that AI systems can reflect limitations or biases contained in their training data, design or operation.

Where AI contributes to decisions or processes that could materially affect people, reasonable steps should be taken to identify and reduce the risk of unfair, discriminatory or inappropriate outcomes.

Intellectual property

Employees must respect copyright, trademarks, confidential information and other intellectual property when using AI. AI must not knowingly be used to reproduce, misuse or disclose intellectual property where Southern Foodservice does not have the appropriate rights or authority.

AI-generated material intended for significant commercial use should be reviewed where intellectual property ownership, licensing or attribution may be relevant.

AI-generated content

Southern Foodservice may use AI to assist in creating written, visual, analytical or other content. AI-assisted content remains subject to our normal standards for accuracy, quality, confidentiality, intellectual property and approval.

Where AI has played a material role and disclosure would reasonably be expected or is required by law, regulation, contractual obligation or industry practice, Southern Foodservice will consider appropriate disclosure. Minor uses of AI, such as proofreading, formatting, brainstorming or improving wording, will not ordinarily require specific disclosure.

Prohibited uses

Southern Foodservice does not permit AI to be knowingly used to:

• deceive, defraud or deliberately mislead;
• impersonate another person without legitimate authority;
• create unlawful, discriminatory, abusive or harmful material;
• compromise cybersecurity or bypass security controls;
• improperly disclose personal, confidential or commercially sensitive information;
• infringe intellectual property rights;
• manipulate business records or fabricate evidence;
• conceal misconduct;
• make unauthorised high-impact decisions about individuals; or
• undertake activities that breach applicable laws, contractual obligations or Southern Foodservice policies.

Governance and accountability

Southern Foodservice is responsible for how AI is adopted and used within our organisation. We maintain appropriate oversight of AI technologies used for business purposes and may establish different approval, monitoring and review requirements depending on the potential risk of a particular AI system or use case.

Responsibility for AI-generated work remains with the person or business area using the technology. AI does not replace individual or organisational accountability.

Reporting concerns

Employees are encouraged to report suspected inappropriate AI use, inaccurate or harmful AI outcomes, privacy or security incidents, or other concerns relating to AI. Potential incidents should be raised through the appropriate Southern Foodservice manager or internal reporting process so that the issue can be assessed and corrective action taken where necessary.

We encourage employees to raise concerns rather than conceal mistakes involving AI.

Continuous improvement

AI technology is developing rapidly.

Southern Foodservice will continue to review developments in AI technology, Australian regulation, government guidance, cybersecurity and industry best practice. Our approach to AI governance, approved technologies and internal controls may be updated as technologies and risks evolve.

Policy review

This policy will be reviewed periodically and when significant changes occur in our use of AI, applicable legislation, regulatory guidance or relevant technology. Southern Foodservice reserves the right to update this policy as required.

Contact

Questions about Southern Foodservice's use of artificial intelligence or this policy can be directed to Southern Foodservice through our usual contact channels.