Modern Slavery Reform: AI Due Diligence
From a Statement to a Standard of Care
For most of the businesses captured by it, the Modern Slavery Act 2018 has felt like an annual writing task. An entity with annual consolidated revenue of at least 100 million dollars prepares a modern slavery statement each financial year, lodges it on the government's public register, and moves on. The statement describes the risks of modern slavery in the entity's operations and supply chains and the steps taken to assess and address them, measured against the mandatory reporting criteria set out in the Act. Around 3,000 entities report this way. For many, the process has settled into a document produced by the compliance or legal team, reviewed once, and largely forgotten until the next cycle.
That settled state is ending. In July 2026 the Australian Government announced its intention to strengthen the framework, signalling a move that had been building since the statutory review of the Act and the recommendations of Australia's Anti-Slavery Commissioner. The proposed direction includes civil penalties for non-compliance and a new offence, framed around a large organisation's failure to prevent modern slavery in its supply chains, with a defence available to entities that can show they took reasonable steps. The Commissioner has recommended a mandatory, risk-based due diligence obligation and a mechanism to declare that a particular product, service or industry carries a high risk of modern slavery. The clear direction of travel is from a disclosure framework to an action framework, backed by penalties and oversight.
For the operations, procurement or compliance lead at an importer, manufacturer or wholesaler, this is a material change. A statement can be written once a year by a couple of people. Genuine, risk-based due diligence across a supply chain cannot. It is a continuous data exercise, mapping suppliers, assessing risk, gathering evidence and monitoring for change, sustained across an entire supplier base and provable to a regulator on demand. This guide is for the person who now has to build that capability without standing up a large new team to do it.
Are You In Scope, and How Deep Does It Go
The reporting threshold under the Act is clear: entities with annual consolidated revenue of at least 100 million dollars that carry on business in Australia must report. A business can easily sit above that line, particularly an importer, distributor or manufacturer where revenue is high even when headcount is not. But the reform conversation reaches further than the reporting entities themselves, because a due diligence obligation pushes responsibility down the supply chain. A large customer conducting genuine due diligence will ask its suppliers for evidence, and so businesses below the threshold increasingly feel the obligation second-hand, through the demands of the customers they supply.
How Modern Slavery Reform Reaches You
The practical significance is that the population affected by this reform is much larger than the 3,000 reporting entities. Every business that sells into a large customer's supply chain is drawn in, because that customer's due diligence becomes the supplier's compliance requirement. Getting the scope question right is therefore about more than whether you must lodge a statement. It is about understanding how far down and across your own supply chain a credible due diligence process has to reach, and how much of that same evidence your customers will start to demand from you.
Why the Reform Breaks the Old Approach
The reason a due diligence obligation is so much heavier than a reporting obligation comes down to the difference between describing risk and managing it. A modern slavery statement can be written from the top down: the team gathers what it knows, describes the supply chain in general terms, lists the policies in place, and produces a document. Due diligence works from the bottom up. It requires actually knowing who your suppliers are, including beyond the first tier, assessing each for modern slavery risk against real indicators, taking proportionate action where risk is found, and keeping the assessment current as suppliers and conditions change. The old approach could rest on generalisations. The new one rests on evidence.
Disclosure Reporting vs Risk-Based Due Diligence
| Metric | Annual Statement | Continuous Due Diligence | Improvement |
|---|---|---|---|
| Supplier visibility | First tier, described broadly | Mapped across tiers | Deeper |
| Risk assessment | General narrative | Screened against real indicators | Evidenced |
| Cadence | Once a year | Ongoing monitoring | Continuous |
| Response to risk | Optional, described | Proportionate action expected | Active |
| Evidence for a regulator | The statement itself | A documented audit trail | Provable |
This is where the manual approach that produced an annual statement quietly fails. A large business might work with hundreds or thousands of suppliers, each with its own suppliers beyond them. Assessing that network by hand, in spreadsheets and email threads, was never really feasible, which is part of why the disclosure model tolerated generalisation. Once due diligence becomes mandatory and penalties attach to failure, the gap between a general narrative and a genuine, evidenced process becomes a legal exposure rather than a presentational one. The reasonable steps defence signalled in the reform rewards businesses that can show a real process and holds to account those that cannot, and a real process across a large supplier base is a data problem before it is anything else.
Where AI Earns Its Place
Artificial intelligence does not discharge a modern slavery obligation on its own, and it does not replace the human judgement the law expects a business to exercise about its own risk. What it does well is carry the volume and the continuity that make genuine due diligence feasible across a large supply chain, so that the accountable people spend their time on decisions and remediation rather than on data collection. There are four places it earns its keep.
The first is supplier mapping and tiering. Before a business can assess risk, it has to know who its suppliers are, and for many organisations that information is scattered across procurement systems, accounts payable, contracts and inboxes. AI can help consolidate and structure that information, extract supplier relationships from documents, and build the tiered map of the supply chain that due diligence depends on. This is the same discipline of pulling structured meaning out of messy commercial documents that we cover in contract review with AI to extract key terms and risks.
The second is risk screening. Not every supplier carries the same modern slavery risk. Risk concentrates by geography, by sector, by the nature of the work and by the length and opacity of the chain. AI can screen a supplier base against these indicators and against public risk information, surfacing the suppliers that warrant closer attention so that limited human effort is directed where it matters rather than spread thin across everyone equally.
A Due Diligence Workflow
The third is continuous monitoring. A once-a-year assessment cannot satisfy an obligation that is meant to be current. Supplier circumstances change, new suppliers are added, and public risk signals shift. Automated monitoring can keep the risk picture up to date and flag material changes for human review, turning due diligence from an annual event into a living process. This is the same shift from periodic reporting to continuous, evidenced compliance that we examine in automated compliance reporting for Australian businesses.
The fourth is evidence and statement drafting. Every due diligence step carries an implicit need to prove it happened. Automated logging of what was assessed, what was found and what action followed produces the audit trail that a reasonable steps defence relies on, and it also feeds the annual statement, which becomes an accurate summary of a real process rather than a document written from generalities.
Where the Effort Goes, Manual vs Automated
What Reasonable Steps Looks Like in Practice
The reasonable steps defence signalled in the reform is worth dwelling on, because it shapes what a business should actually build. A defence framed around reasonable steps does not demand perfection, and it does not demand that a business eliminate all modern slavery risk from a supply chain it does not fully control. What it demands is a genuine, proportionate, documented process, one a business can put in front of a regulator to show it took the risk seriously and acted on it. That is a meaningfully different bar from certainty, and it is one a business can realistically meet if it builds the right foundations.
Consider a typical 200-person importer sourcing goods from several countries. It cannot audit every factory in every tier of its chain, and the law is not asking it to. What a reasonable steps position looks like for that business is a supplier map that reaches beyond the first tier where it matters, a risk assessment that concentrates scrutiny on the higher-risk sourcing, evidence that it acted where risk was found, and a monitoring process that keeps the picture current. None of those elements requires a large compliance department. Each of them requires structured data and a repeatable workflow, which is exactly what automation provides. The failure the reform targets is not imperfect knowledge of a complex chain. It is the absence of any real process at all, the annual statement written from generalities with nothing genuine behind it.
This is why the businesses that struggle most under a due diligence regime are not the ones with the most complex supply chains. They are the ones with the least structured data about the supply chains they have. A business that already knows its suppliers, that already captures its procurement in systems rather than inboxes, and that already treats its disclosures as evidence, converts to due diligence at modest cost. A business that has treated modern slavery as an annual writing task, with no underlying data discipline, faces a much steeper climb, because it has to build the visibility from scratch at the same time as the rules are firming up around it.
The Trajectory, and What It Rewards
It helps to see this reform as one point on a longer line. The Modern Slavery Act introduced disclosure. The statutory review and the Anti-Slavery Commissioner pushed toward action. The July 2026 announcement signalled penalties and a failure-to-prevent offence for large organisations. Consultation on the detail is the current stage. The specific penalty levels and the exact scope of any due diligence duty are still being worked through, so no business should treat any particular figure or start date as settled. What is settled is the direction, and it favours businesses that build a genuine capability early rather than waiting for the final drafting.
The Direction of Travel
The businesses best placed for this are the ones that already treat their non-financial disclosures as a data discipline rather than an annual write-up. Modern slavery due diligence sits alongside a wider tightening of supply-chain and sustainability obligations, including the phased climate-related financial disclosures under the new standard, which we cover in AASB S2 climate disclosure and AI for emissions. The common thread is that regulators increasingly expect evidence, not narrative, and the businesses that have built the data plumbing to produce that evidence, as products like Carbonly do for supply-chain sustainability data (detailed in our Carbonly case study), find each new obligation far cheaper to meet than those starting from spreadsheets each time.
What to Do Now
The reform is not fully drafted, but the sensible response does not depend on the final detail, because the foundational work is the same under any version of the rules. The first task is visibility: know who your suppliers are, across tiers, in structured form. A business that cannot answer that question cannot conduct due diligence under any regime, and building that map pays off regardless of exactly how the law lands. The second task is risk focus: rank the supplier base so that attention goes to genuine risk rather than being diluted across everyone. The third is evidence: log what you assess and what you do, so that a reasonable steps position is demonstrable rather than asserted.
For an importer, manufacturer or distributor, the strategic read is that modern slavery is joining the set of obligations that can no longer be met with an annual document and good intentions. It is becoming a continuous, evidenced, data-driven discipline, and that is both a burden and, handled well, an efficiency. The same supplier map and monitoring that satisfies a modern slavery duty also strengthens procurement, resilience and quality across the business, benefits that connect directly to the operational gains we discuss in AI use cases for Australian manufacturing and order automation for wholesale distributors. The businesses that come through this reform well will be the ones that stopped treating modern slavery as a writing task and started treating it as what it has become: an operational capability, built on data, that has to work continuously and hold up under scrutiny.
Related Reading
- AASB S2 climate disclosure and AI for emissions
- Contract review with AI to extract key terms and risks
- Automated compliance reporting for Australian businesses
- AI use cases for Australian manufacturing
- Order automation for wholesale distributors
- Carbonly: supply-chain sustainability data case study
This article is general information, not legal or compliance advice. Modern slavery reforms are still under consultation, and your obligations depend on your revenue, your supply chain and the final form of the law. Confirm your position against Australian Government and Anti-Slavery Commissioner guidance and qualified advice.