Employers could face penalties for including certain non-compete clauses in employment arrangements under proposed Federal Government reforms, even where they never attempt to enforce the restraint, with corporate penalties of more than $1 million.
If passed, the proposed changes would prohibit non-compete clauses for employees earning at or below the high-income threshold, currently $190,100, as well as restraints preventing former employees from soliciting former colleagues. The definition of a non-compete could also extend to some arrangements requiring employees to repay bonuses, forfeit benefits or face other financial consequences if they join a competitor.
The reforms would also change the way restraints can be drafted for higher-income employees. Cascading clauses containing alternative restraint periods or geographic areas would be prohibited, meaning a drafting approach commonly used to give courts alternative options could instead make the entire restraint unenforceable.
Restraints contained in contracts entered into before the legislation commences, which is expected to occur in 2027, would not be affected.
Holding Redlich Partner Charles Power can discuss:
· how non-compete, co-worker non-solicitation and bonus arrangements could be affected by the proposed reforms
· what the proposed ban on cascading restraints could mean for the way post-employment restraints are drafted
· what employers should consider when changing existing employment arrangements once the new regime, if passed, commences.
New automated decision-making transparency requirements coming into effect on 10 December
Employers using AI and automated tools to screen job candidates, assess employees or support other workforce decisions will need to consider new privacy transparency requirements commencing on 10 December.
The new Australian Privacy Principle requirements apply where personal information and computer programs are used to make or inform decisions that significantly affect an individual’s rights or interests. Importantly for employers, the rules are not limited to decisions made entirely by AI. The OAIC guidance indicates a computer program may be captured where it is a key factor in human decision-making.
This could bring commonly used workplace technologies into focus, including automated recruitment screening and candidate ranking tools and AI-assisted employee assessments. Organisations may also remain responsible where these technologies are provided by third-party vendors.
Holding Redlich General Counsel Lyn Nicholson can discuss:
· how the new requirements could apply in a workplace context, including where automated recommendations materially influence decisions affecting employees or contractors
· the key challenges for employers, including how to identify where automated decision-making already occurs across the business
· practical steps organisations should take before 10 December to ensure compliance.
End-of-year functions: What employers need to consider before the celebrations begin
As workplaces prepare for end-of-year functions, employers should remember that celebrations outside the office can still carry workplace responsibilities and risks.
An employer-organised or authorised function can constitute “work”, meaning inappropriate conduct at an end-of-year event can have consequences for both employees and employers. Risks can range from bullying and sexual harassment complaints to workers’ compensation claims, with alcohol potentially increasing the likelihood of inappropriate behaviour.
Simply having workplace policies in place may not be enough. Employers should actively communicate expected standards of behaviour, ensure employees understand relevant policies and complaints processes, monitor conduct and alcohol consumption at the event, and consider how employees will travel to and from the venue.
Holding Redlich Partner Michael Selinger can discuss:
· when an end-of-year function may be considered an extension of the workplace and what that means for employers
· the workplace safety and conduct risks that can arise at functions organised or hosted by the company
· what employers and HR managers should do before an event to help manage those risks.
Fair Work Commission introduces AI disclosure requirements
The risks of relying on AI in Fair Work Commission claims are in the spotlight after a former ALDI employee was ordered to pay $1,230 towards the supermarket’s legal costs for continuing an unfair dismissal claim despite being warned twice that it was unlikely to succeed.
The Commission found the applicant had been notified of his dismissal three days before completing the six-month minimum employment period required for protection from unfair dismissal.
While the Commission made clear that using AI to assist with drafting is not inherently unreasonable, parties remain responsible for assessing the accuracy, relevance and legal basis of material they submit.
From 20 October, parties lodging documents with the Commission will be required to disclose whether they used generative AI and confirm that they have reviewed the accuracy and relevance of their material. The new guidance follows a 70 per cent increase in the Commission’s overall workload over the past three years, which has been largely attributed to the proliferation and use of AI in unfair dismissal and general protections applications.
Holding Redlich Special Counsel Natasha Jones and Michael Hope can discuss:
· what the new AI disclosure and verification requirements mean for parties appearing before the Commission
· how employers can identify and respond to claims containing irrelevant or inaccurate AI-generated material
· what happens if parties to the dispute fail to comply with the new guidance.
Safe Work Australia to focus on AI, climate change and workplace harassment as WHS laws face next phase of reform
Artificial intelligence, climate change and harmful workplace behaviours will become an increasing focus of workplace safety regulation following the release of Safe Work Australia’s Best Practice Review of the model Work Health and Safety laws final report.
The Review made 21 recommendations to strengthen and maintain harmonisation of the model WHS framework, some of which could result in immediate changes including the development of industry codes of practice, greater involvement by unions to initiate prosecutions and restrictions on insurance or indemnity cover for legal costs incurred in defending a WHS prosecution.
While the recommendations will not necessarily become law in every state and territory, they provide employers with an indication of where workplace safety regulation could be heading.
Holding Redlich Partner Michael Selinger can discuss:
· what the Review signals about the future direction of WHS regulation, particularly around AI, climate change and workplace harassment
· potential changes that could arise from the Review and what they could mean for employers
· what employers should be considering now in light of the Review, before any of the recommendations are implemented.
Digital duty of care to extend online safety obligations beyond social media
Australian businesses operating websites, apps and AI-enabled tools would face new online safety obligations under proposed reforms carrying penalties of up to $109.2 million.
The proposed digital duty of care reforms would shift the focus from responding to harmful content after it occurs to identifying foreseeable online safety risks and putting appropriate safeguards in place. Businesses with customer forums, messaging or chat functions, user-generated content, AI-powered tools, personalised recommendations and other interactive features could be affected.
The changes could also bring greater scrutiny to how digital services are designed, including algorithms, AI tools and recommendation systems that may contribute to foreseeable harm. Businesses could face new risk assessment, governance and record-keeping requirements, including assessments before significant changes are made to a service.
Holding Redlich Senior Associate Louise Almeida can discuss:
· what the shift from responding to online harm to identifying foreseeable risks could mean in practice
· how websites, apps, algorithms, AI tools and personalised recommendations could come within the proposed regime
· what businesses with customer forums, messaging functions and other interactive features should consider
· how businesses can incorporate online safety into existing privacy, cyber security and governance processes.
Unexplained absence does not automatically amount to abandonment of employment
A recent Fair Work Commission ruling has confirmed that an employee who stops attending work and does not respond to attempts to contact them has not necessarily abandoned their employment.
In Mr Md Abdullah Haque v dnata Airport Services Pty Ltd [2026] FWC 1698, the Fair Work Commission found that dnata had dismissed Mr Haque after he stopped attending work and did not respond to repeated attempts to contact him.
Although Commissioner Crawford applied existing Full Bench authority, he expressed doubts about whether the Fair Work Act intends to exclude abandonment of employment from the definition of dismissal.
Holding Redlich Partner Stephen Trew can discuss:
· the enquiries employers should make before concluding that an employee has abandoned their employment
· how the wording of correspondence can determine whether employment is characterised as abandoned or terminated by the employer
· what the Commissioner’s doubts about the existing approach could mean for future abandonment of employment cases
· how employers can respond when an employee stops attending work and cannot be contacted
· the consequences of incorrectly treating an unexplained absence as abandonment of employment.
