Work with us
AI regulation in Australia / Digital Duty of Care (draft legislation)

Digital Duty of Care (draft legislation)

Draft legislation putting a legal duty of care on digital services, including AI chatbots, to protect Australian users and especially under-18s, with penalties up to $109.2 million. Out for consultation; bill promised for 2026.

Status
Proposed · Australia (Commonwealth) · eSafety Commissioner
Applies to
Digital service providers: social media platforms, apps, online games and AI chatbots used by Australians, with specific duties toward under-18s
Primary source
Official document →
Last reviewed
28 September 2026

What it is

Draft legislation, released 9 September 2026 for targeted consultation, that would put a legal duty on digital service providers to keep Australians safe online, enforced by eSafety with penalties up to $109.2 million. It names AI chatbots alongside apps and games. It is not law yet; the government says it will introduce the bill this year.

What it requires

Social media platforms must offer every user a choice between an algorithmic feed and a plain feed of accounts they follow (“My Feed, My Way”). Digital services including online games, apps and AI chatbots must protect under-18s from design features with harmful behavioural effects, such as addictive mechanics, and from listed categories of harmful content. Platforms must document the measures they take against identified risks of harm to Australian users and show those measures keep working. eSafety gains removal-notice powers over nudify apps and faster cyber-abuse schemes.

Does this reach your business?

Only if you provide a digital service that Australians, and in particular under-18s, use directly: a consumer app, an online game, or a public-facing AI chatbot. A business that uses AI internally, or a chatbot that serves adult customers on a business-to-business site, is not the target as drafted. The scope will settle in the bill; if minors can reach your chatbot, check the bill when it is introduced.

What we recommend

If you are in scope, our advice is to start the record now: what your service does, who can reach it, what could harm a young user, and what you already do about it. That is the documentation the duty will ask for, and it is the same record Certrak keeps for any AI system. If you are not in scope, note the date and move on.

Questions people ask

No. Draft legislation was released for targeted consultation on 9 September 2026. The government says it will introduce the bill to Parliament this year.

Yes, as drafted. Online games, apps and AI chatbots that under-18s use would have to protect them from harmful design features and listed categories of harmful content, and document the measures they take.

No. The draft duty falls on providers of digital services to the public. Internal AI use, and a chatbot serving adult customers of a business, are not the target as drafted.

Up to $109.2 million for failing to meet the duty, enforced by the eSafety Commissioner, according to the government's announcement of the draft.

Updates on this