Automated Payroll When No Internet or Access to Your Cloud-Hosted Payroll System: What Next 

The preservation of critical business systems needs a reliable backup if, for any reason, including a cyber attack by an agent or the need to secure data, Loss of internet for any reason, Cloud environment unavailability, or any other interruption, it is no more impactful than on payday.

Payroll not processing impacts every employee in profound ways. Funds to meet financial commitments, or to be available for daily food shopping, cannot be unavailable.

“It’s (issues that shut down internet access) is the start of a tsunami because more and more vulnerabilities will be discovered by AI,” says Yigal Unna, the former head of Israel’s national cyber defences.

Inzenius, the innovator behind its Patented Payroll platform, leads the way in payroll and its associated disaster plan for any internet-connected outages. Inzenius’s plan includes the ability to process its payroll and create the ABA file for the banks.  

That’s Inzenius  Australian Compliant Award Interpreter

www.inzenius.com

Recent no-internet cases include: 

CrowdStrike,  PHP security breach to the Northern Territory Government internet shutdown, reminds us of a potential critical vulnerability that is high-risk, payroll-related, with a high impact on the people we employ and could cost a lot of time and expense to repatriate the payroll 

https://www.reuters.com/world/asia-pacific/eu-agency-warns-websites-may-need-be-taken-offline-php-prepares-patches-2026-10-21/

https://www.cyber.gov.au/alert/critical-vulnerabilities-in-php 

https://www.abc.net.au/news/2024-07-20/what-happened-crowdstrike-global-outage-explainer/104122582

https://www.abc.net.au/news/2026-10-08/nt-government-departments-hit-by-network-outage-triple-zero/107245122

AFR Article

Start of a tsunami’: Global web shutdowns signal dystopian AI future

https://www.afr.com/technology/start-of-a-tsunami-global-web-shutdowns-signal-dystopian-ai-future-20261022-p617qz

 Paul Smith AFR Technology editor, wrote

Artificial intelligence may be pitched as the good guy in Thursday’s stories about how it discovered cybersecurity flaws in most of the world’s websites, but the events are an ominous watershed moment that shows how the game has changed forever.

The existential warnings about rising power of the bots are starting to eventuate before our eyes.

Home Affairs Minister Tony Burke and the Australian Signals Directorate were quick to tell organisations to patch the bugs found by AI. Alex Ellinghausen

The unearthing of flaws in the PHP programming language that runs about 70 per cent of the world’s websites, which had laid undiscovered by human eyes for years, left government agencies and businesses scrambling to apply patches.

An advanced OpenAI model had been used to find the gap in cyber defences, setting off a frantic race across time zones. European websites were the first alerted via emergency bulletins and sites began disappearing to take pre-emptive action. Warnings here followed quickly, with the Australian Signals Directorate and Home Affairs Minister Tony Burke alerting organisations to take immediate action, before hackers could take advantage.

“It’s a start of a tsunami because more and more vulnerabilities will be discovered by AI,” says Yigal Unna, the former head of Israel’s national cyber defences, who is visiting Australia for a series of conferences.

“This is great when it is in the hands of the right people … but even as someone who appreciates the tremendous advantages to mankind of technology, I see we need to recalculate our approach towards AI as we are almost on the verge of losing control.”

The PHP weaknesses may be patched by most organisations, but not everyone exposed will be quick enough to know about it, or to know what to do about it. And next time it could easily be the bad guys that use increasingly capable open weight models, or find a way to use frontier models for nefarious purposes.

“Critical bugs have always surfaced, but AI finds more of them, faster, and turns them into attacks in days. Defenders are losing time,” says Alastair MacGibbon, a former national cyber adviser to then-prime minister Malcolm Turnbull.

“You can’t patch what you don’t know you have. Australia runs on thousands of obscure pieces of code most people have never heard of. Every organisation needs a software and firmware bill of materials, vendors need to provide one, and government now needs to map the shared dependencies across sectors that no single company can see.”

Ageing technology

In reality many critical services run on ageing technology, which has been viewed as too risky or costly to replace, and MacGibbon warns that things are coming to a head. It will mean organisations taking themselves offline before hackers can use AI to do it, to figure out how to make themselves AI-safe.

“We’ll have to accept planned outages to maintain systems because scheduled downtime beats catastrophe,” MacGibbon adds.

The threat posed by increasingly capable AI models became a global diplomatic issue earlier this year when Anthropic disclosed its unreleased Mythos model was capable of finding hidden flaws in websites and other tech systems. It formed a group of US companies called Project Glasswing, which looked to plug flaws in major systems before it could be released to the public.

The Trump administration imposed an export ban in June, which temporarily forbade any non-US citizens from accessing the system. This outraged cyber professionals in Australia, who reasoned that Australia was being kept in the dark by a major ally, and highlighted the lack of sovereign capability of middle powers as their economies increasingly rely on US- or Chinese-owned AI models.

Senior executives from major AI labs such as Anthropic spent most of September publicly debating the chance of AI eventually wiping out humanity, with the doomsday scenarios often starting with AI agents committing cyberattacks to knock out critical infrastructure.

Former director-general of the Australian Signals Directorate and special adviser to professional services firm Island Bell, Rachel Noble, says Thursday’s PHP problems have been responsibly handled by all involved, but this is a portent of many such future threats.

“Unfortunately, we should expect to see more events like this, and not all will be benign. AI can assist in finding previously undiscovered vulnerabilities but equally, in the hands of bad actors, it can be weaponised,” she warns.

Having spent years on the global cyber frontline in Israel, Unna admits he is worried that AI is ushering in an unprecedented era of human history. He is fearful about a total absence of global regulations in AI, and the fact that – even in Project Glasswing – society was forced to trust individuals making billions of dollars from the technology to act for the greater good.

“Yes, they immediately closed it to the public once they realised what a monster they created, but then they petted it for a bit and trained it to be a good monster and were left with nobody accountable if something bad happens,” Unna says.

“We are definitely in trouble, but I’m also hopeful because we still have the upper hand and are still smarter than our machines and our technology. Maybe I have seen too many Hollywood movies where Schwarzenegger helps humans in the end, but I’m still optimistic that mankind’s overall wisdom and wish to live peacefully will contain this monster as well.”

Rogue OpenAI agent breach shows Australia exposed on cyber defence

Paul Smith edits the technology coverage and has been a leading writer on the sector for 20 years. He covers big tech, artificial intelligence, business use of tech, the fast-growing Australian tech industry and start-ups, and national innovation policy from our Sydney newsroom. Send tips to @sayssmithy.99 on encrypted messaging platform Signal. Connect with Paul on Twitter. Email Paul at psmith@afr.com

Frequently Asked Questions

How can you process payroll when the internet is down? 

In simple terms, Inzenius facilitates the use of the backup connected to an in-pay office system duplication system

What data will be available to the system? 

Up to the latest full backup before the system was not available from 

How can the bank distribute the funds?

This will be up to the bank to have its own process, but as a baseline principle, they only need to take the ABA file and run the net pay to each employee’s bank account to enable their scheduled payments and all funds to be accessed 

What happens after the internet connection to the live payroll is restored?

The backup from the processed payroll is activated on the cloud-hosted environment.

Can the entire system and data be easily relocated from one cloud-hosted environment to another or brought back in-house? 

Yes