--- title: "Voluntary AI Safety Standard for Marketers | Aemorph" url: https://aemorph.com/en-au/guides/voluntary-ai-safety-standard/ description: "Australia's Voluntary AI Safety Standard explained for marketers: the 10 guardrails, the Guidance for AI Adoption, AI content disclosure and the laws that apply." schema_type: Article publisher: Aemorph publisher_entity: https://aemorph.com/entitymap.json#org_aemorph language: en-AU updated: 2026-10-25 --- Guides # Australia's Voluntary AI Safety Standard for Marketers: the 10 Guardrails and the Guidance for AI Adoption **Australia's Voluntary AI Safety Standard is guidance, not law: the Department of Industry, Science and Resources published it on 5 September 2024 as 10 voluntary guardrails, and on 21 October 2025 it published the Guidance for AI Adoption, six essential practices that it describes as updated and simplified guidance that evolves the standard.** The department states that the standard creates no new legal duties, and the essential practices are the Australian Government's guidance for responsible AI use. The laws that already govern your marketing still apply to anything AI writes for it, the Australian Consumer Law and the Privacy Act 1988 among them. The National AI Plan of 2 December 2025 keeps those existing laws as the foundation of Australia's approach to AI regulation. [Book a strategic conversation](https://aemorph.com/en-au/contact/) Written by [Kevin Dam](https://aemorph.com/en-au/about/), founder of Aemorph, who has practised SEO since 2010 and sits on the [Forbes Agency Council](https://councils.forbes.com/profile/Kevin-Dam-Global-Search-Architect-%7C-SEO-GEO-AEO-IMO-LLM-Visibility-Aemorph/4c8c7a38-1285-470d-8ea8-e517a310a00b). Aemorph is a [Google Partner](https://www.google.com/partners/agency?id=4441861842). [LinkedIn](https://www.linkedin.com/in/kevindam) --- ## What the Voluntary AI Safety Standard Covers, and Who It Applies To **The standard's 10 guardrails apply to all organisations throughout the AI supply chain, and the first version focuses on deployers: the organisations that supply or use an AI system to provide a product or service.** That definition takes in a marketing team that drafts copy with a generative AI tool or runs a chatbot on its website. The National Artificial Intelligence Centre publishes the [Voluntary AI Safety Standard](https://www.industry.gov.au/publications/voluntary-ai-safety-standard) on the department's website, whose landing page was last updated on 2 December 2025. Each guardrail includes procurement guidance for working with suppliers, because most deployers rely on AI systems that third parties develop. ### Is the Voluntary AI Safety Standard law? **No: the department states that, being voluntary, the standard does not create new legal duties about AI systems or their use.** It sets a consistent practice for organisations and, in the department's words, expectations for what future legislation looks like as the government considers its options on mandatory guardrails for high-risk settings. The National AI Plan (Department of Industry, Science and Resources, 2 December 2025) states that the government's regulatory approach builds on Australia's existing legal and regulatory frameworks, and that agencies and regulators keep responsibility for AI-related harms in their own domains. The government is establishing an AI Safety Institute to support existing regulators with independent advice. ### What are the 10 guardrails? **The guardrails cover accountability, risk, data, testing, human oversight, disclosure, contestability, supply chain transparency, record keeping and stakeholder engagement.** In the department's wording, organisations: 1. establish, implement and publish an accountability process including governance, internal capability and a strategy for regulatory compliance 2. establish and implement a risk management process to identify and mitigate risks 3. protect AI systems, and implement data governance measures to manage data quality and provenance 4. test AI models and systems to evaluate model performance and monitor the system once deployed 5. enable human control or intervention in an AI system to achieve meaningful human oversight 6. inform end users regarding AI-enabled decisions, interactions with AI and AI-generated content 7. establish processes for people impacted by AI systems to challenge use or outcomes 8. are transparent with other organisations across the AI supply chain about data, models and systems 9. keep and maintain records to allow third parties to assess compliance with the guardrails 10. engage stakeholders and evaluate their needs and circumstances, with a focus on safety, diversity, inclusion and fairness. Full adoption of the standard means adopting all 10, starting with guardrail 1. The guardrails align with ISO/IEC 42001:2023 and the NIST AI Risk Management Framework, and the department describes them as ongoing activities, not one-off tasks. ### How does the Guidance for AI Adoption relate to the standard? **The Guidance for AI Adoption, published on 21 October 2025, is what the department calls updated and simplified guidance that evolves the standard, set out as six essential practices, and the department's own pages now point readers to it first while keeping the standard published.** The six practices are: decide who is accountable; understand impacts and plan accordingly; measure and manage risks; share essential information; test and monitor; and maintain human control. The National AI Centre publishes two versions on [ai.gov.au](https://www.ai.gov.au/staying-safe-and-responsible/essential-ai-practices): foundations, for organisations starting out or using AI in low-risk ways, and implementation guidance, for complex or higher-risk use. The National AI Plan states that the six practices underpin new tools and resources. The standard has not been withdrawn; the newer guidance is the department's current starting point. ### Which laws already apply to AI-generated marketing? **The National AI Centre lists the Australian Consumer Law's prohibitions on misleading and deceptive conduct among the laws relevant to AI that produces misleading representations, including misleading outputs, misleading representations or silence about when AI is in use, and misleading statements about an AI system's performance.** Its summary of [AI and Australian law](https://www.ai.gov.au/staying-safe-and-responsible/ai-and-australian-law) also lists privacy laws on the collection, use and disclosure of personal information, with specific provisions for some automated decision making applying from 10 December 2026, and a statutory tort for serious invasions of privacy that commenced on 10 June 2025. The centre notes the list is not comprehensive. The [Australian Consumer Law guide](https://aemorph.com/en-au/guides/accc-australian-consumer-law-marketing/) and the [Australian Privacy Principles guide](https://aemorph.com/en-au/guides/australian-privacy-principles-marketing/) cover those laws for marketers. *General guidance, not legal advice.* --- ## What the Standard Means for Your Marketing and SEO Copy **For a marketing team, the guardrails and the six practices turn into a short list of decisions: who owns AI use, which content tells people that AI made or changed it, who reviews AI output before it goes live, and what records show it.** The table lists common AI uses in marketing and search work against the guidance behind each one. | Claim type or AI use | What the guidance says | What to check | Source | | --- | --- | --- | --- | | AI-generated or AI-edited content | Disclose when content is AI-generated; the organisation chooses the mechanism to suit the use case, stakeholders and technology | A written rule on when your team labels AI content, and who decides | VAISS guardrail 6 | | Higher-impact content | Use clearer or more visible disclosure where AI content influences decisions or affects rights, safety or trust, such as health, recruitment, financial or legal information and public communications | Health, finance, careers and policy pages drafted with AI | Being clear about AI-generated content | | Disclosure methods | Labelling, watermarking and metadata recording, used alone or together; the guidance describes minimal or no disclosure for low-impact content with limited AI use | Image and video files, alt text and article footers | Being clear about AI-generated content | | Chatbots and AI assistants on your site | People know when they are interacting with AI; a customer talking to a chatbot knows it is not a human specialist | Chatbot greetings, names and avatars | Guidance for AI Adoption, practice 4 | | Claims about your own AI features | The Australian Consumer Law's prohibitions on misleading conduct are relevant to misleading statements about an AI system's performance | "AI-powered" and accuracy claims on product and service pages | AI and Australian law | | Facts and data fed into AI tools | Data governance measures manage data quality and provenance | The source of every fact an AI draft repeats | VAISS guardrail 3 | | Human review | Meaningful human oversight, matched to the system's autonomy and the stakes | A named reviewer approves AI drafts before publication | VAISS guardrail 5; Guidance for AI Adoption, practice 6 | | AI register | A register of every AI system, including AI embedded in customer engagement tools | Writing assistants, chatbots, ad platforms and CRM features | Guidance for AI Adoption, practice 4 | | Agencies and suppliers | Clarify who is responsible for each part of the AI supply chain, and ask suppliers for proof of testing | Agency and vendor contracts name responsibilities for AI use | Guidance for AI Adoption, practices 1 and 5 | *Sources: Department of Industry, Science and Resources, Voluntary AI Safety Standard, "The 10 guardrails" and "Introduction to the standard"; National AI Centre, "Guidance for AI adoption: foundations", "Guidance for AI adoption: implementation guidance", "Be clear about AI-generated content" and "AI and Australian law". All checked 24 October 2026.* ### Does AI-written marketing copy carry a label? **It depends on the content: the National AI Centre's guidance on [being clear about AI-generated content](https://www.ai.gov.au/staying-safe-and-responsible/essential-ai-practices/be-clear-about-ai-use) ties disclosure to the content's impact and to how much AI contributed.** It recommends transparency when AI created most of the content, changed existing content or changed the potential meaning of the information, and it notes that even a small edit changes meaning, such as "did" becoming "did not". The guidance describes minimal or no disclosure for low-impact content with limited AI use, and identifies combined disclosure methods as an option for higher-impact content with more AI involvement. The centre also states that laws on misleading information, privacy and online safety apply to AI-generated content in some cases. ### How does the guidance treat marketing emails drafted with AI? **As a use with different risks from assessing job applications: the foundations guidance compares the two to show that risk depends on how the AI tool is used.** It asks organisations to check each specific use, consider how AI tools get misused and keep clear records of the actions taken under each practice. For a marketing team, that means one entry per use in the AI register, a named owner and a review step scaled to the stakes, for example a closer review for a health claim than for a subject line. *General guidance, not legal advice.* --- ## What the Standard Means for AI Answers and AI-Generated Content **The standard guides organisations that develop or deploy AI; it gives no business control over what ChatGPT, Claude, Gemini or Perplexity say about it, and those answers rest with the systems' developers and the sources the systems draw on.** What your business controls is its own use of AI and the public facts it publishes. ### Does the standard control what AI assistants say about your business? **No: the voluntary standard creates no right to dictate a third-party assistant's description of your business; its guardrails address the organisations that develop and deploy those systems.** In Aemorph's working model, ChatGPT, Claude, Gemini and Perplexity describe a business from the public sources they find, your own pages and profiles among them, so an outdated or unsupported claim on your site becomes material for their answers. The fix sits at the source, on the pages and profiles you control. The free [Hallucination Report](https://tools.aemorph.com/hallucination) compares what the AI engines say about your business with what your website currently says, and flags where they conflict. ### How do provenance and human oversight apply to the facts you publish? **Guardrail 3 asks for data governance that manages data quality and provenance, and guardrail 5 asks for meaningful human oversight; a record of client-approved facts supports source provenance and a documented human review of what you publish.** Oversight of your own AI systems, and disclosure to the people who use them, remain separate controls under guardrails 5 and 6. The implementation guidance adds that organisations keep clear records of governance decisions, testing, incidents and monitoring, because good records support audits and reviews. *General guidance, not legal advice.* --- ## How Aemorph Checks Claims Before Publication **Every fact Aemorph publishes for a client traces to a source the client supplied and approved, and the client's nominated point of contact signs off each regulated claim and each showstopper answer before Aemorph publishes it.** Aemorph is not a law firm. It does not assess your organisation's conformance with the Voluntary AI Safety Standard or the Guidance for AI Adoption, and adopting either stays your organisation's decision: Aemorph's sign-off step supports your own governance and does not replace it. | Check | Applies to | Standard | Checked against | When | | --- | --- | --- | --- | --- | | AI use and AI-generated content | Organisations applying Australia's Voluntary AI Safety Standard | Each published fact traces to an approved source and a named approver | Your point of contact's sign-off | Before Aemorph publishes | | Every regulated claim | Every client in a regulated industry | Approved by the person your business nominates | Your point of contact's sign-off | Before Aemorph publishes | | Showstopper answers | Every Showstopper Sensor (Sprint 1) client | Your business answers each question and your point of contact approves the answer | Your point of contact's sign-off | Before the answers go live | *Source: Aemorph compliance practice, October 2026.* ### Who signs off the claims on your pages? **You decide who signs off: the business owner, the chief marketing officer or the risk team.** The facts in your verified record are the facts your business supplies and approves. Inside the [Showstopper Sensor (Sprint 1)](https://aemorph.com/en-au/services/30-day-showstopper-sensor-sprint/), a 30-day showstopper-answer sprint for ChatGPT, Claude, Gemini and Perplexity, the same sign-off covers the showstopper answers before they go live. *General guidance, not legal advice.* --- ## Australian Businesses: Start With a Free Strategic Conversation **Australian businesses that want AI visibility built on facts their own reviewers have approved start with a free strategic conversation with Aemorph.** Aemorph is headquartered in Singapore, serves Australian clients remotely in Australian business hours and replies to every message within 24 hours. The [LLM Influencer System page](https://aemorph.com/en-au/llm-influencer-system/) sets out how Aemorph runs SEO, AEO and GEO as one programme, with GEO leading: earning citations and mentions for your business inside ChatGPT, Claude, Gemini and Perplexity. [Book a strategic conversation](https://aemorph.com/en-au/contact/) **Disclaimer:** This article reflects Aemorph's reading and understanding of the laws, regulations and official guidance cited, based on the sources and review dates stated. It provides general information, not legal advice, and does not assess compliance for your specific circumstances. THE STRATEGIC CONVERSATION ## Start with a free *strategic conversation*. Every Aemorph engagement starts with a free strategic conversation. It is the fastest way to see whether Aemorph fits your business before you commit budget. [Book a strategic conversation →](https://aemorph.com/en-au/contact/) STARTING POINT Free You talk through your goals with Aemorph, and Aemorph quotes your engagement after the conversation.