Due Diligence 11 November 2026 · Gumshoe

Mobius v Inoteq: The Court Case That Made Supplier Bank Verification Your Job

A hacker sits in a director's sent folder. They've read the rhythm of the business — the cadence of invoices, the tone of approvals, the exact letterhead. On 28 April 2022, they used the Mobius Group

A hacker sits in a director's sent folder. They've read the rhythm of the business — the cadence of invoices, the tone of approvals, the exact letterhead. On 28 April 2022, they used the Mobius Group director's real email to ask Inoteq to update bank details. A fake letter on Mobius letterhead followed. The money moved. The real Mobius never saw it.

The hacker didn't need to spoof the address — they owned it. Real credentials. Real history. Real letterhead. Inoteq's accounts payable team did what looked reasonable: they phoned. The line crackled. They asked for email confirmation. The hacker, still inside the director's inbox, replied "confirmed." That was the entire verification chain. One call on a bad connection, then a reply from the compromised source. Judge Gary Massey ruled it prudent but inadequate. Inoteq wore the $190,000 loss plus interest because they were better placed than Mobius to guard against the fraud — the payer's duty, not the victim's luck.

The National Anti-Scam Centre's Targeting Scams report puts the national bill for payment redirection scams at A$166.8 million in 2025. [2024] WADC 114 is the precedent that makes 'we called them' a failed defence. The mechanism is repeatable: compromised inbox, authentic letterhead, a single verification channel that loops back to the attacker. The control that fails is the one that trusts the channel it's trying to verify.

The Anatomy of the Breach: From Compromised Inbox to Second Payment

A hacker logged into the genuine email account of a Mobius Group director and sent Inoteq a request to update Mobius's bank details. A follow-up email from the same address attached a fake letter on Mobius letterhead announcing the new bank details. Nothing in the message triggered a spam filter because it came from the real sender's mailbox.

Inoteq's accounts payable clerk phoned Mobius to verify. The connection was poor; the person on the other end sounded like the director. Inoteq then asked for written confirmation by replying to the same email thread. The hacker, still inside the compromised inbox, replied "confirmed." That single loop — call on a bad line, then email confirmation from the very account that initiated the fraud — was the entire verification chain.

The fraudulent request arrived on a Thursday. Inoteq then paid about $190,000 owed to Mobius into the new account. The fraud only surfaced when Mobius chased its unpaid invoices. The money moved through a major bank's fast-clearing system and was dispersed before Mobius discovered the diversion. No second invoice, no second approver, no callback to a number on file from onboarding. The clerk treated the compromised email thread as a closed loop.

What makes this breach instructive is not the sophistication of the attack — it was a credential compromise and a forged PDF — but the ordinariness of the failure. The verification steps Inoteq took would satisfy most internal audit checklists: a phone call was made, written confirmation was requested, the request came from a known sender. Each step was documented. Each step was also routed through channels the attacker already controlled. The court found that Inoteq's process, while executed in good faith, did not meet the standard of reasonable care because it never broke the chain of compromise.

Why 'Prudent' Isn't Enough: The Legal Weight of Inadequate Verification

Judge Gary Massey's ruling in Mobius Group Pty Ltd v Inoteq Pty Ltd [2024] WADC 114 draws a line that every AP team needs to internalise: a verification step that stays inside the attacker's control is not verification at all. The court accepted that Inoteq's clerk made a phone call to Mobius — a step that looked prudent on paper. The line was poor. The clerk then asked for email confirmation. The hacker, still sitting in the director's inbox, supplied it. That sequence — one call, then written confirmation from the compromised account — constituted the entire defence. It failed.

The $190,000 loss wasn't written off as an unavoidable fraud. It was sheeted home to Inoteq because the payer, not the compromised supplier, held the procedural leverage to break the chain. The court found the organisation controlling the outbound payment bore the responsibility — a formulation that shifts the burden squarely onto the payer. Good faith is not a defence when the verification protocol never left the fraudster's reach.

The ruling makes the standard concrete: verification must occur on a channel the attacker cannot reach. A phone call to a number on file — not a number in the suspicious email — followed by a callback to a known contact at the supplier, documented with date, time, and who authorised the change. Anything less is not a control; it is a ritual. The $190,000 judgment plus interest is the price of learning that lesson in court rather than at onboarding.

Inoteq's clerk acted in good faith. The fraudster exploited the gap between "I called" and "I verified independently." That gap is where every payment redirection scam lives. The court has now priced it.

The Liability Shift: Why the Payer, Not the Victim, Carried the Loss

Judge Massey's finding that Inoteq was "better placed than Mobius to guard against the loss" reframes payment redirection from a supplier risk to an accounts payable governance failure. The court treated the outbound payment function as a control point with a professional duty of care — not a clerical pass-through. Inoteq controlled the payment instruction, the verification protocol, and the release of funds. Mobius, by contrast, had no visibility into Inoteq's internal process once its director's email was compromised.

This aligns with the risk allocation logic in a typical regulatory case where a licensee was found to have breached its obligations by failing to manage cybersecurity risk across systems it controlled. For finance leaders the takeaway is practical: a verification process that can be completed from inside a compromised inbox is unlikely to look reasonable after the fact. The resulting judgment and costs illustrate the market price of that gap.

The practical implication for CFOs and procurement leads is that "we called the number in the email" is no longer a defensible control — it is evidence of a control gap. The court's reasoning places the onus on the paying entity to design verification that survives the compromise of the supplier's communications. That means independent, out-of-band confirmation: a known contact at a pre-registered number, a portal-based approval workflow, or a physical sign-off by an authorised officer. Anything less leaves the payer carrying the loss, regardless of good faith.

The question a court asks is what a reasonable payer would have done, and after this decision a single unconfirmed call is a hard position to defend. A single phone call to a number supplied in the change request, followed by email confirmation to the compromised account, fails the test. The $190,000 judgment shows how quickly an ordinary bank-change request becomes a liability. The only control that holds is one that does not trust the channel the request arrived on.

Building the Defensible Wall: Beyond the Single Phone Call

The court's reasoning draws a bright line: verification must occur on a channel the fraudster cannot reach. In Mobius v Inoteq, the payer rang the supplier but the line was poor, so it asked for confirmation by email — and the hacker answered from the same compromised inbox that originated the request. That is not a second check; it is the same check repeated. A defensible control requires an independent channel: a pre-registered phone number held in your master vendor file, a portal-based approval workflow with multi-factor authentication, or a physical sign-off by an authorised officer who has never seen the change request email.

The "second check" must be documented, time-stamped, and performed by someone other than the processor who received the initial request. If the AP clerk who actioned the change also made the confirmation call, the segregation of duties collapses. The standard the court has set is not "did we try?" but "did we verify through a channel the attacker did not control?" That means the contact point used for confirmation must exist before the change request arrives — not be supplied by it.

That pre-existence requirement changes how onboarding works. Every new vendor record must capture a verified, out-of-band contact — landline, mobile, or portal credential — at the point of creation, not when a change request lands. If the only number on file is the one printed on the first invoice, the control is already compromised. The master vendor file becomes the single source of truth for verification channels, and any amendment to that file triggers the same multi-layered check as a bank detail change.

Weak Control vs Defensible Control for Bank Detail Changes
Control StepWeak Control (Failed in Mobius)Defensible Control (Court-Resilient)
Initial request receivedEmail from known domain with attached letterhead PDFPortal submission with MFA or signed physical form
First verificationPhone call to number on the request letterCall to pre-registered landline in ERP master record
Second verificationEmail confirmation to same compromised inboxPortal acknowledgment or second authorised signatory wet-ink
DocumentationFile note of call and email trailSigned verification checklist with timestamps and sign-off IDs
Audit survivabilityCollapses under single-point compromiseStands — each check independent of the request vector

Documentation is the evidence trail that survives cross-examination. A defensible process records: the pre-registered contact used, the identity of the person who performed the check, the date and time, and the exact confirmation received. Screenshots of portal approvals, signed physical forms, or file notes of verbal confirmations against known numbers all meet this standard. An email thread that begins with the fraudulent request does not. The cost of these controls is measured in minutes per change; the cost of their absence, as Mobius v Inoteq demonstrates, is the full value of the misdirected payment plus interest and legal fees.

Uncommon Insights

The Mobius v Inoteq breach wasn't a technology failure — it was a trust architecture failure. Inoteq's AP team executed exactly what their procedures demanded: they called, they confirmed, they documented. The flaw was procedural blindness to the fact that every verification channel they used — phone, then email — originated from or could be intercepted by the same compromised source. Trust, once established digitally, became a single point of failure.

1independent channel minimum
0digital-only paths accepted
$190kcost of assuming 'prudent' was enough

Digital workflows embed trust into efficiency: an email from a known domain, a phone number on a letterhead, a portal login. Fraudsters don't break these channels; they inhabit them. The court's ruling makes clear that "prudent" reliance on any channel that shares provenance with the change request is legally insufficient. The only defensible control is a verification path that exists entirely outside the digital thread — a pre-registered landline, a physical wet-ink sign-off, a portal credential issued at onboarding. Trust must be anchored before the transaction, not verified within it.

Mobius v Inoteq The Court

Most AP teams still treat a returned phone call as the gold standard. It isn't. A call to a number on a fraudulent letterhead, followed by an email to the compromised inbox, is a single channel wearing two masks. The court drew the line at independence: the verification must originate from a source the fraudster never touched — a landline registered at onboarding, a director's personal mobile stored in the ERP master file, a wet-ink authority signed before the first invoice.

Key Takeaways

  • STOP — Treat every bank detail change request as compromised until verified through a channel the requester did not provide.
  • CHECK — Call the supplier on a pre-registered landline from your ERP master record, not a number on the invoice, letter, or email.
  • CONFIRM — Require a second, independent authorisation: portal acknowledgment with MFA, a wet-ink sign-off from a second authorised officer, or a callback to a known contact at a different number.
  • DOCUMENT — Record every verification step with timestamps, sign-off IDs, and the specific channel used; a file note of a phone call and email trail will not survive scrutiny.
  • ENFORCE — Build the workflow into your AP system so the payment run cannot proceed without the completed checklist attached.
VERIFY NOW

Run a free supplier check in seconds

Search by business name, ABN, or ACN. Instant PASS/WARN/FAIL against 20+ Australian government registers.

Start verifying →
VERIFY A SUPPLIER
Run a free check in seconds

Search by business name, ABN, or ACN. Get a real-time PASS/WARN/FAIL report across 8 verification checks.

Start verifying →

Contains data sourced from the Australian Business Register and ASIC, © Commonwealth of Australia, licensed under CC BY 3.0 AU.