The Australian Financial Complaints Authority (AFCA) opened consultation on 31 August 2026 on its proposed Scam Rules. These are the new rules for handling scam complaints under the Scams Prevention Framework (SPF), and consultation closes on 28 September 2026. It would be easy to treat this as a matter for complaints teams. It isn't. It is really a question about how well your mule account controls will hold up when someone outside your organisation reviews them.

The timeline, in plain terms

AFCA became the authorised external dispute resolution scheme for SPF scam complaints from 1 July 2026. Organisations in the regulated sectors of banking, telecommunications and digital platforms had to be AFCA members from 1 September 2026. From 31 March 2027, consumers and small businesses can bring scam complaints to AFCA under the SPF.

The proposed Scam Rules sit in their own section of the AFCA rulebook, separate from the rules for ordinary financial services complaints. They cover how SPF complaints will be handled, which complaints are excluded, monetary limits, and a full set of definitions. Any final changes need approval from the AFCA board and from ASIC, which will consult the other SPF regulators. AFCA says it will develop more detailed operational guidelines in early 2027.

The shift that already happened

The more important change for banks came earlier. Since 12 March 2026, under AFCA's updated rules, AFCA can look at the conduct of a receiving bank (the bank the scam money was paid into) even where the complainant isn't that bank's customer. It can consider complaints about accounts opened by scammers without permission, and it can join several banks to one complaint and split responsibility between them.

AFCA has said this does not create new legal duties to non-customers. But AFCA decides what is fair in all the circumstances, taking into account good industry practice as well as the law. In practice, when a victim's money passed through one of your accounts, the question will increasingly be whether you could reasonably have stopped it.

Where mule controls get tested

A mule account is one used to receive and pass on criminal proceeds, sometimes by a willing participant and sometimes by someone who has been deceived. Under this model, the controls that matter most are:

Onboarding. Could you show what checks were done when the account was opened, and why they were enough for that customer?

Early detection. Does your monitoring flag the classic mule pattern of money arriving from unrelated people and leaving again quickly? How long does it take for an alert to be reviewed?

Response. When another bank tells you an inbound payment was a scam, can you act out of hours, freeze what is left and reply with a clear record of what you did?

Evidence. Six months later, could someone who wasn't involved rebuild the sequence of events from your records alone?

What to do before March 2027

If you are in scope, read the consultation paper even if you don't make a submission. The definitions and exclusions will shape which complaints reach AFCA. Then run a handful of recent scam cases where your bank was on the receiving end, and test them against the questions above. Fixing gaps now costs far less than finding them in an AFCA determination.

The SPF is changing the question from "whose customer was this?" to "who could have prevented this?". Receiving banks need to be ready to answer the second question.