Australia’s Social Media Algorithm Reform: How My Feed, My Way Could Change What People See Online

Australia is moving toward a major rethink of how social-media platforms and other digital services should manage the systems that decide what users see. Rather than concentrating only on removing harmful individual posts after they appear, the proposed approach seeks to examine the technology, design choices and commercial incentives that influence the entire online experience. At the centre of the debate is a proposal known as My Feed, My Way, which could give Australians greater control over personalised recommendation systems and allow eligible users to choose a feed based more heavily on the friends, creators and accounts they deliberately follow. The proposal forms part of a wider Digital Duty of Care framework that would place stronger responsibilities on digital companies to identify foreseeable risks, protect younger users and demonstrate that they are taking reasonable steps to create safer online environments.

The significance of the proposal goes well beyond a new button in a social-media settings menu. Recommendation systems have become one of the defining technologies of the modern internet. They influence which videos become popular, which political discussions gain attention, which products people discover, which creators build audiences and which topics dominate a person's daily online experience. A user may believe that opening an app simply reveals what is happening on that platform, but in reality the content displayed can be heavily filtered, ranked and personalised by systems designed to predict what will capture attention. Two people following similar accounts can therefore encounter very different versions of the same service.

Australia's emerging reform agenda asks whether people should have a stronger right to determine how that process works. Instead of accepting personalised recommendations as an unavoidable condition of participating in social media, users could receive a clearer choice. Those who enjoy algorithmic discovery could continue using personalised recommendations. Those who want more control could select a feed that places greater emphasis on accounts they intentionally follow rather than continuously presenting material selected according to behavioural predictions.

This approach reflects a broader change in online-safety policy. For many years, regulation focused heavily on specific categories of illegal or harmful content. Governments and regulators attempted to remove damaging posts, respond to abuse and create reporting mechanisms. Those tools remain important, but modern digital platforms are increasingly defined not only by the content they host but also by the mechanisms determining which content receives attention. A harmful item that exists somewhere on a platform is one problem. A system that repeatedly recommends similar material to vulnerable users is a different and potentially more complex problem.

The proposed Digital Duty of Care is designed to move some responsibility toward prevention. Digital companies could be expected to assess the foreseeable risks created by their products, features and design decisions rather than waiting until users suffer harm and submit complaints. This principle has particular importance for children and teenagers, who may be especially vulnerable to compulsive design, unrealistic body-image material, bullying, sexually explicit content, dangerous challenges and other harmful experiences.

Australia's reforms are therefore part of a much larger question about the future of the internet: should the digital services used by millions of people be designed primarily around maximising attention, or should users have stronger rights to decide how their feeds operate?

What Australia Is Proposing

The Australian Government released an exposure draft of the Online Safety Amendment (Digital Duty of Care) Bill 2026 on 8 September 2026. An exposure draft is not the same as a final law. It allows proposed legislation to be examined, discussed and potentially changed before progressing through the parliamentary process.

One of the most prominent elements associated with the proposal is My Feed, My Way. Under the announced approach, social-media users over the age of 16 could be offered greater control over whether personally recommended material appears in their default feed.

A user could choose to continue receiving personalised recommendations. Such recommendations may be produced using signals including viewing history, searches, likes, shares, accounts followed, comments, watch time and other forms of interaction.

Alternatively, the user could choose a feed focused more directly on friends and creators they have intentionally decided to follow. The concept is not necessarily to eliminate every algorithm operating inside a service. Modern platforms rely on algorithms for countless functions, including spam detection, security, search, moderation and content delivery. The reform is more specifically concerned with whether personalised recommendation engines should automatically dominate the main feed.

New and existing users could be informed that this choice is available. That requirement may prove crucial because a control that exists but is difficult to locate can have little real-world impact. If users never know that an alternative feed exists, formal choice may not translate into meaningful choice.

For the proposal to work effectively, the presentation of the decision could therefore matter almost as much as the legal requirement itself. Users would need to understand what each option means, how personalisation works, whether their preference remains active and how easily it can be changed later.

The government has presented the reform as an attempt to increase individual control. Instead of technology companies alone deciding what type of recommendation system dominates the user experience, individuals could have a more direct role in selecting how content reaches them.

Why Social Media Algorithms Have Become So Influential

Social media has changed dramatically since the early era of online networking. Many platforms initially revolved around a relatively straightforward idea: people created an account, connected with friends or selected accounts to follow, and then viewed updates from those connections.

Today, many major services operate differently. Following an account remains important, but the platform may continuously place recommended material into a person's feed even when that person has never followed the creator responsible for it.

This change has created enormous opportunities for discovery. A previously unknown musician can reach millions of listeners. A small business can suddenly attract customers. A teacher can distribute educational material far beyond a local classroom. A talented creator can build an audience without having an established media company behind them.

Recommendation technology can therefore be extremely useful.

The concern arises from the objectives used to decide what should be recommended. Platforms compete intensely for user attention. More time spent inside an application can create more opportunities to display advertising, gather behavioural signals and promote additional content.

As a result, recommendation engines may be designed to predict what is most likely to keep a user engaged.

Engagement, however, is not always the same as usefulness, accuracy, balance or wellbeing.

Material capable of generating strong emotions may sometimes outperform calmer content. Anger can generate comments. Fear can encourage sharing. Controversy can attract repeated viewing. Unexpected material can interrupt normal scrolling patterns. Highly emotional stories can keep users watching.

If an algorithm receives signals suggesting that a particular type of content captures attention, it may recommend more of it. The user may then respond to that material, producing additional signals. Over time, a feedback loop can develop.

This process helps explain the widespread concern about online "rabbit holes." A person who begins by watching ordinary content about a subject may gradually receive increasingly intense versions of that material because each interaction becomes another signal indicating interest.

The underlying technology does not necessarily intend to create harm. It may simply be pursuing the objective it was designed to optimise. Yet if the objective gives too much importance to attention while giving too little importance to safety, diversity or wellbeing, harmful patterns can emerge.

That is one reason policymakers are increasingly focusing on platform design rather than examining only individual posts.

What My Feed, My Way Could Mean in Practice

Imagine that a user joins a social-media service. Under a stronger user-choice model, the platform could explain that there are different ways to experience the feed.

One option could be personalised discovery. The platform analyses behaviour and recommends content that it predicts the user will find interesting.

Another option could focus primarily on chosen connections. Posts from friends, subscribed creators and followed accounts would become more central to the default experience.

The user would then make the choice rather than having one model silently imposed.

For existing users, the service could similarly provide a clear opportunity to review the setting.

This might appear to be a small interface change, but default settings have substantial power. Many people never explore complicated configuration menus. If a platform activates personalisation automatically and places the alternative several menus deep, most users may simply remain with the default.

A meaningful reform would therefore need to consider whether the choice is understandable, prominent and neutral.

The language used to describe the options would also matter. A platform could theoretically describe personalised recommendations as an exciting, modern experience while portraying the alternative as limited or outdated. Even if both choices technically exist, such wording could influence user decisions.

Similarly, repeated prompts encouraging people to reactivate personalisation could weaken the value of an opt-out.

Long-term persistence is another important issue. If a user chooses a follower-focused feed, that preference should ideally remain effective after app updates, device changes or redesigns unless the user consciously changes it.

The quality of the alternative feed will matter as well. If companies intentionally make the non-personalised experience less useful, users may feel pressured to return to recommendation systems.

Implementation will therefore determine whether My Feed, My Way becomes a meaningful consumer right or simply another setting that few people use.

Personalisation Would Still Be Available

An important point is that the proposal does not require personalised recommendation systems to disappear entirely.

Many people genuinely value recommendation technology. It helps them find music, recipes, travel ideas, educational content, sports highlights, technology tutorials and creators they would otherwise never encounter.

For these users, personalisation can make an overwhelming amount of information easier to navigate.

A complete prohibition could therefore remove significant benefits.

The proposed approach instead focuses on whether personalisation should be voluntary.

This distinction allows two different types of users to coexist. Someone who enjoys discovering unexpected material could continue using a recommended feed. Someone who feels overwhelmed by endless recommendations could choose a different experience.

The policy debate may eventually focus on whether the appropriate model is opt-out or opt-in.

Under an opt-out system, personalised recommendations could operate unless the user disables them.

Under an opt-in system, personalisation would begin only after the user actively chooses it.

The difference has major practical consequences because defaults strongly influence behaviour.

Australia's parliamentary debate could therefore involve not only whether choice should exist but also how that choice is presented and when consent should be required.

The Digital Duty of Care Goes Much Further

My Feed, My Way has attracted attention because it can be explained easily, but it represents only one part of the broader Digital Duty of Care concept.

The proposed framework is based on the idea that digital companies should take reasonable steps to identify and reduce foreseeable harm.

That would shift the regulatory emphasis from reaction toward prevention.

A company introducing a new engagement feature might need to consider whether the feature creates predictable safety risks.

A gaming platform developing a new communication function may need to think about how younger users could be affected.

An artificial-intelligence service may need to evaluate how vulnerable users interact with its system.

A social-media platform changing its recommendation engine may need to consider whether the new ranking system increases exposure to dangerous or psychologically harmful material.

This type of approach treats online safety as part of product design rather than something handled only by a moderation team after launch.

The principle resembles safety expectations in other industries. Companies producing physical products are generally expected to consider foreseeable risks. A manufacturer cannot reasonably argue that safety is exclusively the customer's responsibility when a dangerous design feature could have been identified beforehand.

Digital services are obviously different from physical products, but policymakers increasingly argue that similar preventive thinking should apply.

The challenge is determining what "reasonable steps" should mean for rapidly changing technology.

A global social-media platform serving millions of Australians has different resources and risks from a small online service. Regulation may therefore need proportionality so that obligations reflect the scale and nature of the service.

Overly vague rules could create uncertainty.

Overly detailed rules could quickly become outdated as technology changes.

The success of the Digital Duty of Care could therefore depend on whether it establishes durable safety principles while allowing enough flexibility for innovation.

Why Children and Teenagers Are at the Centre of the Reform

Protection of younger users is one of the strongest themes in Australia's proposed digital policy.

Children and teenagers increasingly experience a significant part of life online. They communicate with friends, watch entertainment, play games, learn, create videos, follow celebrities and interact with AI systems.

The online environment can provide enormous benefits, but younger users may also face risks that adults are better equipped to recognise.

A teenager encountering highly edited images may interpret unrealistic appearance standards as normal.

A child exposed repeatedly to dangerous challenge videos may underestimate the risks involved.

A young user interacting with harmful material about eating disorders may receive additional recommendations because the system interprets attention as interest.

Someone experiencing bullying may repeatedly encounter related content at a moment of emotional vulnerability.

These risks become more complicated when platforms use continuous recommendation systems.

A child does not necessarily need to search actively for harmful material. The service may deliver it.

This difference is central to the argument for system-level regulation.

Traditional online-safety approaches often assume that the user searches for content or encounters it randomly. Recommendation systems introduce another possibility: the platform itself selects what appears next.

If the technology continually amplifies problematic material, removing individual posts may not be enough.

The broader Digital Duty of Care proposal is intended to apply additional expectations relating to younger users across digital environments including apps, online games and artificial-intelligence services.

This reflects the reality that children's online lives no longer occur within a single category of platform.

Harmful Content and Design Risks

The proposed framework identifies several areas of concern for people under 18.

These include material that encourages or accelerates eating disorders, pornography, hostile attitudes toward women or gender equality, glorification of crime, dangerous stunts, abuse, bullying and other material capable of causing serious psychological distress.

The proposal also looks beyond content to design.

This distinction is important.

An individual video may not necessarily be the only source of risk. The way a service repeatedly presents material can also shape behaviour.

Autoplay can remove stopping points.

Infinite scrolling can make sessions continue indefinitely.

Notifications can repeatedly pull users back into applications.

Visible engagement counts can create pressure to compare popularity.

Disappearing content can encourage frequent checking because users fear missing something.

Personalised recommendations can continuously supply material matched to previous behaviour.

Each feature may appear relatively harmless in isolation. Combined, they can create a highly compelling environment.

For younger users whose habits and self-image are still developing, these design characteristics may deserve additional scrutiny.

Relationship With Australia’s Under-16 Social Media Rules

The algorithm reform should be understood alongside Australia's existing rules relating to social-media access by people under 16.

The minimum-age framework and My Feed, My Way address different stages of the online experience.

The under-16 system focuses primarily on whether younger Australians can maintain accounts on designated age-restricted social-media platforms.

My Feed, My Way focuses on how eligible users experience feeds once they are able to use the services.

The wider Digital Duty of Care then creates another layer by considering how digital services should manage foreseeable risks, including risks affecting people under 18.

Together, these policies represent a more comprehensive approach.

One part addresses access.

Another addresses recommendation systems.

Another addresses product design and safety obligations.

This layered model suggests that Australian online-safety regulation is moving away from the idea that a single age limit or content-removal mechanism can solve the entire problem.

Digital safety involves multiple questions: who can use a service, what the service recommends, how it encourages engagement, what risks it creates and how companies respond when harm becomes foreseeable.

Why a Follower-Focused Feed Could Appeal to Users

Many people originally joined social-media platforms because they wanted updates from people and organisations they intentionally selected.

Over time, feeds on some services have become increasingly dominated by recommendations from unfamiliar accounts.

A user may follow family members, local organisations, sports teams, community groups, businesses and favourite creators yet still find much of the feed filled with unrelated recommended material.

A stronger follower-focused option could change that relationship.

Users may gain greater predictability.

Someone following local news providers could see more of those updates instead of unrelated viral videos.

A person interested in photography could concentrate on photographers they specifically selected.

A small-business customer could more reliably see updates from shops and services they intentionally followed.

The feed could become more directly connected to conscious user choices.

This does not necessarily mean chronological order. A service could still use software to sort content from followed accounts. But personally recommended material from unfamiliar accounts would play a smaller role.

For users who feel that modern platforms are increasingly deciding what they should care about, this could restore a sense of control.

Could Algorithm Choice Reduce Doomscrolling?

"Doomscrolling" describes the tendency to continue consuming negative, worrying or emotionally intense information for longer than intended.

Recommendation algorithms are not the sole cause of this behaviour, but they can contribute.

A personalised platform always has another item available.

After one video finishes, another begins.

After one article, another appears.

After one controversial discussion, the platform may recommend several related posts.

There is rarely a natural stopping point.

A feed based more heavily on intentionally followed accounts may behave differently. Once the newest material from those accounts has been viewed, there may simply be less content waiting.

That natural limit could help some users stop.

However, switching recommendation systems off would not automatically eliminate excessive use.

Infinite scrolling could remain.

Autoplay could remain.

Notifications could remain.

Messaging could remain.

Engagement counters could remain.

Users could still search actively for endless content.

The potential benefit is therefore better understood as one component of a broader effort to reduce compulsive design rather than a complete solution to problematic social-media use.

Privacy Questions Could Become More Important

Personalised recommendations depend heavily on data.

A platform may analyse what users watch, how long they watch it, what they search for, which posts they save, which accounts they follow, what they ignore and what they click.

These signals help create a behavioural profile.

A user choosing a less-personalised feed may naturally wonder whether the platform will stop collecting those signals.

The answer is not necessarily.

Recommendation preferences and privacy practices are different areas.

A service could respect a user's decision not to receive personalised feed recommendations while still collecting certain information for security, advertising, analytics or other purposes.

Clear communication will therefore be essential.

Users need to know whether choosing a follower-focused feed changes only what appears on the screen or also changes how behavioural information is processed.

They may also want to know whether advertisements remain personalised.

Search suggestions could involve separate recommendation systems.

Suggested accounts could continue to be personalised.

Notifications might still be influenced by behavioural predictions.

The phrase "turn off the algorithm" can therefore create unrealistic expectations.

Almost every modern digital platform relies on algorithms.

The meaningful question is which form of personalisation the user can control.

Proposed Reform Areas and Their Potential Value

What the Reform Could Mean for Social Media Companies

For major social-media companies, these changes could affect both technology and business strategy.

Recommendation engines are deeply connected to the economics of digital platforms.

If personalised systems keep users engaged for longer periods, companies can show more advertisements and collect more behavioural information.

If significant numbers of Australians choose less-personalised feeds, usage patterns could change.

Average session lengths may decrease for some users.

Content discovery could become less dependent on automated recommendations.

Platforms might need to improve the experience of follower-based feeds.

They might develop new voluntary discovery tools that users access only when they want recommendations.

Companies could also need new systems for storing user choices and demonstrating compliance.

Engineering teams may need to ensure that a selected preference remains effective across different devices and versions of an application.

Compliance teams may need evidence showing that users receive proper notices.

Design teams may need to ensure that one choice is not presented unfairly.

Risk teams may need to conduct stronger safety assessments before new features are launched.

These changes could move online safety deeper into the product-development process.

What Smaller Platforms Could Face

Large global companies have substantial technical, legal and compliance resources.

Smaller services may not.

That creates an important regulatory challenge.

If obligations become excessively complicated, compliance costs could unintentionally favour the largest corporations.

A small Australian digital business may struggle to employ specialised lawyers, safety engineers and policy teams.

Effective regulation therefore needs clarity and proportionality.

The objective should be to improve safety without making ordinary innovation impossible.

Risk-based obligations could help.

A service with limited functionality and a small user base may present different risks from a massive platform using sophisticated behavioural profiling.

Rules that recognise those differences could maintain strong safety expectations while avoiding unnecessary burdens.

How Creators Could Be Affected

Recommendation algorithms have transformed the creator economy.

Previously, creators often needed established audiences before achieving large-scale distribution.

Modern recommendation systems can place content from an unknown account in front of thousands or millions of people almost immediately.

This has allowed musicians, educators, comedians, artists, small businesses and independent commentators to grow rapidly.

If a substantial portion of users chooses follower-focused feeds, creators may receive fewer automatic impressions from people who have never encountered them before.

That could reduce certain forms of viral discovery.

At the same time, it might increase the value of genuine follower relationships.

Creators could place greater emphasis on encouraging viewers to follow intentionally.

Audience loyalty could become more important than temporary viral reach.

Creators may also diversify distribution across newsletters, websites, direct subscriptions and other channels rather than relying exclusively on recommendation engines.

The reform could therefore alter how online audiences are built.

What Australian Businesses Could Experience

Businesses increasingly depend on social-media discovery.

A restaurant may reach nearby customers through recommended videos.

A retailer can introduce products to users who have never followed the business.

A tourism operator can attract travellers through viral content.

A local professional can build awareness using educational posts.

If algorithmic discovery becomes less dominant for some users, businesses may need to strengthen direct customer relationships.

Encouraging users to follow an account could become more valuable.

Search visibility may become more important.

Email lists, direct communication and repeat visits could gain significance.

At the same time, users opting out of constant recommendations may still want voluntary discovery tools.

Platforms may therefore create new systems in which users intentionally enter a discovery area rather than having recommendations continuously inserted into the main feed.

That model could preserve opportunities for businesses while giving consumers more control.

News Publishers and Distribution Power

The proposal also raises important questions for publishers.

News organisations have spent years adapting to platform algorithms.

A ranking change introduced by a technology company can dramatically increase or decrease traffic to news sites.

Publishers may invest heavily in a platform only to discover that the algorithm changes and their reach falls.

A stronger follower-based model could partially change this relationship.

If users deliberately follow news organisations and the platform gives greater priority to those choices, publishers may gain a more predictable connection with their audience.

However, the outcome would depend entirely on implementation.

A follower-based feed could still use ranking systems that determine which publisher posts appear first.

The platform would still control many aspects of distribution.

The reform would therefore not automatically remove platform power, but it could strengthen the importance of deliberate subscriptions and follows.

Free Expression and Regulatory Power

Online-safety legislation regularly creates difficult questions about freedom of expression.

Protecting children from serious harm is a widely recognised public objective.

At the same time, democratic societies depend on open political debate, journalism, artistic expression and access to diverse opinions.

Digital regulation therefore needs carefully defined boundaries.

Rules that are too weak can fail to protect vulnerable people.

Rules that are too broad can create uncertainty around lawful speech.

The Coalition has expressed concerns about aspects of the proposed Digital Duty of Care framework, including regulatory scope, ministerial powers, journalism and the definition of harmful material.

Those concerns form part of the parliamentary debate.

My Feed, My Way itself can be distinguished from censorship because choosing not to receive personalised recommendations does not necessarily prevent another adult from publishing or accessing lawful content.

The wider Digital Duty of Care, however, involves additional responsibilities and powers, so questions about scope, safeguards and oversight remain important.

Clear definitions will matter.

Transparent decision-making will matter.

Appeal mechanisms will matter.

Independent regulatory oversight will matter.

Parliamentary scrutiny will matter.

Good online-safety regulation needs to protect people without creating unnecessarily vague restrictions on legitimate expression.

Why Some Want Stronger Opt-In Rules

Critics of an opt-out approach argue that simply providing a switch may not change much.

Most people use default settings.

If personalised recommendations remain active unless users deliberately disable them, the majority may never make a different choice.

Supporters of stronger controls therefore argue that platforms should obtain affirmative consent before activating behavioural recommendation systems.

Under this approach, the user would first choose personalisation.

This difference between opt-out and opt-in regulation may appear technical, but it could dramatically affect adoption.

Suppose 100 users create accounts.

Under an opt-out system, all 100 might begin with personalisation and only a minority may change the setting.

Under an opt-in model, personalisation may begin only for the users who actively select it.

The resulting distribution could be very different.

The final legislative debate may therefore focus heavily on the architecture of consent.

Algorithms Are Not Inherently Harmful

The debate should not create the impression that algorithms themselves are dangerous.

Algorithms are simply methods for processing information and making decisions.

They are used everywhere online.

Search engines use ranking algorithms.

Email providers use algorithms to detect spam.

Banks use algorithms to identify suspicious transactions.

Streaming services recommend entertainment.

Navigation systems calculate routes.

Online stores suggest products.

Educational platforms recommend lessons.

Algorithms can improve safety as well.

A platform can use automated systems to detect abusive behaviour, identify dangerous material or reduce exposure to spam and scams.

The policy problem is therefore not the existence of algorithms.

The important questions are what objectives they optimise, what information they use, what safeguards apply and how much control users possess.

An algorithm designed exclusively to maximise viewing time could produce different outcomes from one designed to balance engagement, diversity, safety and user preferences.

Good regulation should recognise this distinction.

Recommendation Systems and Artificial Intelligence

The importance of algorithm regulation may grow as artificial intelligence becomes more sophisticated.

Future digital services may not simply recommend videos or posts.

AI assistants could select news, suggest products, plan entertainment, summarise political events and decide which information sources users encounter.

Personalisation could become far deeper than current social-media feeds.

A system may learn a person's routines, interests, shopping habits, career objectives and communication preferences.

That could create extraordinarily useful services.

It could also create new forms of influence.

If a digital assistant continually determines what information deserves attention, the question of user control becomes even more important.

Principles established today around consent, transparency and recommendation choice could therefore shape future AI regulation.

Australia's debate may be viewed not only as social-media regulation but also as an early discussion about how people should control automated information systems.

What the Reform Could Mean for Parents

Parents currently carry a significant share of responsibility for protecting children online.

They may need to understand parental controls, privacy settings, screen-time tools, messaging functions, gaming platforms, AI chatbots and social-media features.

The pace of technological change makes this difficult.

A platform can redesign its interface within weeks.

A new service can become popular almost overnight.

A feature that did not exist last year may suddenly dominate teenagers' online lives.

The Digital Duty of Care attempts to place more responsibility on the companies designing these environments.

That does not eliminate the role of parents.

Families still need conversations about online behaviour.

Children benefit from understanding that a recommended video is not necessarily important, accurate or healthy simply because a platform displayed it.

Young people should understand that recommendation systems are designed according to specific goals.

Media literacy therefore needs to include algorithm literacy.

A teenager who understands why certain content keeps appearing may be better equipped to question it.

Teaching Algorithm Literacy

Algorithm literacy could become an increasingly important educational skill.

Students are already taught to consider whether information comes from reliable sources.

The next step is understanding why particular information appeared in the first place.

A recommendation is not neutral.

It may reflect previous behaviour.

It may reflect commercial priorities.

It may reflect popularity.

It may be generated because similar users engaged with the material.

It may be designed to maximise retention.

Understanding these possibilities helps people become more deliberate digital users.

A person can ask:

Why am I seeing this?

Did I choose to follow this account?

Is this content being recommended because it is trustworthy or because it attracts attention?

Am I seeing a representative range of views?

Has my previous viewing behaviour created a narrow feed?

These questions are valuable regardless of the final form of Australia's legislation.

Could the Reform Improve Mental Wellbeing?

It would be unrealistic to claim that changing recommendation settings will solve complex mental-health problems.

Mental wellbeing is influenced by many factors.

However, giving users greater control over repeated exposure may help some people create healthier digital environments.

Someone who repeatedly encounters stressful political arguments could reduce recommendations.

A user recovering from unhealthy eating behaviours may prefer fewer unexpected recommendations related to dieting or body image.

A person feeling overwhelmed by constant viral material could choose a quieter feed.

Parents may prefer greater predictability in the online environments used by older teenagers.

The benefit lies in control.

Users would gain another tool for shaping the environment around them.

Whether this produces measurable improvements would require careful research after implementation.

Enforcement and the Role of eSafety

Rules have limited value if companies can ignore them.

The proposed framework therefore places importance on regulatory enforcement through the eSafety Commissioner.

The government has indicated that serious non-compliance could attract penalties reaching A$109.2 million.

Large penalties can be particularly important when dealing with multinational technology companies.

A small fine may simply become another business expense.

A substantial penalty creates a stronger financial reason to take compliance seriously.

However, enforcement involves more than penalties.

Regulators need technical expertise.

They need access to relevant information.

They may need to understand complex recommendation systems.

They need ways to evaluate whether an apparent user choice actually works in practice.

A platform could state that users have control while designing the interface in a way that strongly discourages the alternative.

Effective oversight may therefore involve examining real user experiences rather than relying entirely on written company policies.

Manipulated Explicit Images and Emerging Harms

The proposed reform package also addresses emerging forms of online abuse associated with manipulated explicit imagery.

Artificial intelligence has made it easier to generate realistic fake images.

This technology can be misused to create sexually explicit material depicting people who never consented to such images.

The harm can be severe.

Victims may experience humiliation, harassment, reputational damage and psychological distress.

Younger people may be particularly vulnerable.

The proposal to strengthen eSafety's ability to respond to services associated with such material demonstrates how rapidly online-safety regulation must evolve.

Traditional regulatory categories may not adequately address technology that changes within months.

Future frameworks therefore need enough flexibility to respond to new forms of abuse without requiring completely new legislation every time a technology develops.

International Significance

Australia's digital reforms could attract attention beyond the country.

Governments around the world face similar questions.

How much control should platforms have over recommendation systems?

Should users have a legal right to choose non-personalised feeds?

How should children be protected from addictive design?

How should artificial-intelligence services manage foreseeable harm?

How much transparency should large technology companies provide?

Australia has already taken a prominent position in debates about youth access to social media.

Algorithm choice could become another area in which international policymakers watch the Australian experience closely.

If My Feed, My Way proves practical and popular, other jurisdictions may consider similar requirements.

Technology companies could also decide that maintaining consistent global controls is easier than building a completely different version of an application for every country.

In that scenario, an Australian regulation could indirectly influence users elsewhere.

The opposite is also possible.

If the system proves confusing, difficult to enforce or easy to circumvent, critics may argue that recommendation regulation is ineffective.

Implementation will therefore be crucial.

Why Technical Details Will Decide Whether It Works

Legislation can establish broad rights, but technical details determine how those rights feel in everyday life.

Consider a user choosing a non-personalised feed.

Will posts be displayed chronologically?

Will advertisements still be personalised?

Will suggested accounts appear?

Will search results use behavioural information?

Will notification recommendations remain personalised?

Can the platform periodically reset the preference?

Can it repeatedly ask the user to switch back?

What happens when the user installs the application on a new phone?

Does the preference apply across devices?

Will every section of the platform respect the choice?

These questions could determine whether people view the system as meaningful or cosmetic.

Clear regulatory guidance will therefore be important.

The Importance of Interface Design

A technically available choice can still be manipulative.

Digital companies have enormous expertise in interface design.

Button size, wording, placement and colour can influence behaviour.

A service might place a large button saying "Keep My Experience Personalised" beside a small text link saying "Use limited feed."

Both choices technically exist, but they are not presented equally.

Regulation may therefore need to consider so-called dark patterns: interface techniques designed to push users toward outcomes that benefit the company.

Meaningful consent should be informed and voluntary.

Users should not need technical expertise to understand the options.

Could Social Media Become Less Addictive?

The broader Digital Duty of Care raises the possibility of platforms reconsidering features designed primarily to maximise engagement.

This does not mean social-media companies would stop competing for user attention.

But stronger safety expectations could encourage designers to examine whether specific features create foreseeable negative behavioural effects.

Potential changes might include clearer stopping points, better controls over autoplay, more transparent recommendation settings, improved notification management and stronger protections for minors.

Any such changes would depend on the final legal framework and regulatory implementation.

The important shift is conceptual.

Instead of assuming that every engagement feature is acceptable until someone proves harm after deployment, companies could face greater responsibility to evaluate risks before and during use.

Economic Questions

Regulation often creates both costs and benefits.

Companies may face expenses associated with engineering, safety assessments, compliance systems and reporting.

Some advertising models could be affected if fewer users choose deep personalisation.

Creators may experience changes in discovery.

Businesses may need new marketing strategies.

At the same time, stronger trust could benefit digital services.

Users who feel they have genuine control may be more comfortable remaining on platforms.

Parents may feel more confident allowing older teenagers to use certain services.

Companies investing in safer design could strengthen their reputation.

Innovation could also shift toward technology that provides useful personalisation without relying on aggressive engagement techniques.

The economic outcome will therefore depend on how platforms respond.

Why the Reform Is About Power as Much as Safety

At its deepest level, the algorithm debate is about power.

Who decides what appears on the screen?

For years, the answer has increasingly been the platform.

Users make choices, but recommendation systems interpret those choices and then determine what comes next.

A person may follow fifty accounts, yet the service still decides which posts receive attention.

The platform controls ranking.

It controls recommendation criteria.

It controls interface design.

It controls experimentation.

It can change the system without asking users.

My Feed, My Way challenges part of that arrangement.

It suggests that users should have a stronger right to determine the type of system governing their attention.

This is significant because attention has become an economic resource.

Platforms compete for minutes and hours.

Advertisers compete for visibility.

Creators compete for reach.

Political campaigns compete for engagement.

News organisations compete for clicks.

The algorithm sits between all of them.

Giving users more control therefore has implications far beyond convenience.

What Happens Next

The proposal remains part of an evolving legislative process.

The exposure draft was released on 8 September 2026, with consultation scheduled to close at noon on 22 September 2026.

Feedback may influence the final form of the legislation.

Political parties may propose amendments.

Industry groups may raise technical concerns.

Child-safety advocates may seek stronger protections.

Civil-liberties organisations may focus on free-expression safeguards.

Technology companies may argue for clearer definitions or longer implementation periods.

Researchers may recommend evidence-based standards.

The final legislation could therefore differ from the exposure draft.

Australians following the issue should distinguish between what has been proposed and what ultimately becomes law.

After legislation passes, regulatory implementation can create another stage of change.

Detailed guidance may determine how companies demonstrate compliance and how the user-choice system operates.

What Australians Should Watch

Several issues could become particularly important during the next stages of the debate.

The first is whether personalised recommendations operate under an opt-out or stronger opt-in model.

The second is how clearly platforms must present feed choices.

The third is how the law defines affected digital services.

The fourth is the scope of obligations relating to children and teenagers.

The fifth is how harmful material and foreseeable risks are defined.

The sixth is how regulatory powers are supervised.

The seventh is whether small businesses receive proportionate compliance expectations.

The eighth is how privacy and recommendation controls interact.

The ninth is how the framework addresses emerging AI systems.

The tenth is whether enforcement focuses on genuine user outcomes rather than formal compliance.

These details will determine the practical significance of the reform.

The Larger Future of Digital Choice

For much of the internet's development, users accepted platform design largely as it was offered.

Terms of service could be accepted or rejected.

Recommendation systems could often be influenced through behaviour but not fundamentally controlled.

The next phase of digital regulation may focus more heavily on meaningful choice.

Users may increasingly expect to control personalisation.

They may expect clear explanations of why content is recommended.

They may want the ability to select chronological or follower-focused feeds.

They may expect AI systems to explain how information is selected.

They may seek stronger privacy controls.

They may want platforms to respect attention rather than continuously maximising engagement.

Australia's proposed reforms fit within this broader transition.

The issue is no longer simply whether an individual piece of content should remain online.

The debate increasingly concerns the architecture of digital environments.

Conclusion

Australia's proposed social-media algorithm reforms could mark an important change in the relationship between users and digital platforms. My Feed, My Way places a straightforward principle at the centre of the debate: people should have a meaningful say in whether personalised recommendation systems determine what dominates their default social-media experience.

For Australians over 16, the proposed model could create a clearer choice between continuing with personalised discovery and choosing a feed focused more closely on intentionally followed friends, creators and accounts. The proposal does not seek to eliminate every algorithm. Instead, it questions whether behavioural recommendation systems should automatically control the main feed without meaningful user choice.

The wider Digital Duty of Care goes considerably further. It seeks to make prevention a stronger part of online safety by requiring digital companies to identify and manage foreseeable risks. Children and teenagers are a central focus, with particular attention given to harmful content, psychologically damaging experiences and product features capable of encouraging unhealthy behavioural patterns.

The potential consequences extend across the digital economy. Social-media companies may need to redesign products and strengthen compliance systems. Creators could place greater value on direct followers. Businesses may need stronger customer relationships. Publishers may reconsider their dependence on recommendation engines. Parents could receive more support from safety-by-design requirements. Regulators would face the challenging task of supervising rapidly evolving technology.

Important questions remain unresolved. Australians will need to watch how the final framework balances online safety with lawful expression, how algorithm choices are presented, whether personalisation becomes opt-in or opt-out, how privacy interacts with recommendation settings and how regulators determine whether companies are genuinely reducing foreseeable harm.

The reform also carries international significance. Recommendation systems have become central to modern communication around the world. Governments everywhere are considering how to protect children, preserve innovation, maintain freedom of expression and prevent large platforms from exercising excessive control over attention.

Australia's approach offers one possible answer: instead of banning useful recommendation technology, give people a stronger right to choose when and how it shapes their experience.

The long-term importance of My Feed, My Way may therefore extend well beyond one social-media setting. It represents a broader challenge to a digital model in which platforms decide what is most likely to keep people watching and users simply accept the result.

The emerging alternative is a model in which people play a more active role in shaping their own information environment.

Whether that model succeeds will depend on legislation, technical design, regulatory enforcement and the willingness of digital companies to treat user wellbeing and autonomy as important measures of product quality.

But the principle behind the proposal is likely to remain influential: technology can help people discover information without requiring them to surrender control over what they see.

Frequently Asked Questions About Australia’s Social Media Algorithm Reforms

What is My Feed, My Way?

My Feed, My Way is a proposed Australian initiative intended to give eligible social-media users greater choice over the way their default feed operates. Users over 16 could choose whether they want personalised recommendations or a feed focused more strongly on friends and creators they intentionally follow.

Is Australia banning social-media algorithms?

No. The proposal does not amount to a general ban on algorithms. Platforms use algorithms for many necessary functions. The reform is focused more specifically on giving users greater control over personalised recommendation systems in the default feed.

What is a recommendation algorithm?

A recommendation algorithm is software that predicts which content a user may want to see. It can analyse signals such as viewing history, watch time, likes, comments, searches, shares and accounts followed.

Why are recommendation systems controversial?

Recommendation technology can be useful for discovering relevant content, but systems designed primarily to maximise engagement may repeatedly amplify sensational, emotionally intense or harmful material.

When was the Digital Duty of Care exposure draft released?

The exposure draft discussed in the proposal was released on 8 September 2026.

Is the Digital Duty of Care already law?

The September 2026 framework discussed here is an exposure draft rather than a final law. It remains subject to consultation and parliamentary consideration.

When is consultation scheduled to close?

The consultation period described in the proposal is scheduled to close at noon on 22 September 2026.

Who would be able to choose a different feed?

Under the announced model, Australians over the age of 16 would be offered greater control over personalised recommendations in their default social-media feed.

What happens if someone chooses a follower-focused feed?

The person could receive a feed focused more closely on friends and creators they deliberately follow instead of having personally recommended material dominate the default experience.

Would personalised recommendations still exist?

Yes. Users who prefer personalised discovery could continue choosing that experience.

Would turning recommendations off remove every algorithm?

No. Platforms would still use algorithms for functions including security, spam detection, content delivery and moderation. The proposal concerns personally recommended feed content rather than all algorithms.

Would the new feed necessarily be chronological?

Not necessarily. The detailed implementation could determine how posts from followed accounts are ranked and displayed.

Would platforms need to tell users about the choice?

Under the proposed model, new and existing users would be notified that feed choices are available.

Why does the way the choice is presented matter?

Default settings and interface design strongly influence behaviour. A difficult-to-find opt-out may produce much less practical change than a clear and neutral choice presented directly to users.

What is the Digital Duty of Care?

It is a broader proposed framework intended to require digital providers to take reasonable steps to identify and manage foreseeable online risks.

Does the framework apply only to social media?

No. The broader proposal discusses risks across digital services such as apps, online games and artificial-intelligence chatbots.

Why are children and teenagers a major focus?

Younger users may be particularly vulnerable to harmful content, compulsive design, bullying, unrealistic body standards and dangerous recommendations.

What kinds of content could receive stronger attention?

The proposal highlights concerns including eating-disorder material, pornography, hostile attitudes toward women or gender equality, dangerous stunts, crime glorification, abuse and bullying.

What does addictive design mean?

The term can describe features such as endless feeds, autoplay, repeated notifications and other mechanisms that encourage users to remain engaged for long periods.

Is My Feed, My Way the same as the under-16 social-media rules?

No. The under-16 framework concerns account access on designated services, while My Feed, My Way concerns feed choice for eligible users. The wider Digital Duty of Care contains additional protections.

Could the reform reduce doomscrolling?

It could help some people by reducing continuous personalised recommendations, but doomscrolling is influenced by many other features including infinite scrolling, autoplay and notifications.

Will opting out of recommendations stop data collection?

Not automatically. Feed personalisation and broader privacy practices are separate issues. Platforms may continue collecting certain information for other legitimate or commercial purposes.

Could advertising still be personalised?

That will depend on the detailed rules and each platform's practices. Choosing a non-personalised feed should not automatically be assumed to disable all forms of personalised advertising.

How could creators be affected?

Creators who rely heavily on recommendation systems to reach new audiences could experience changes in discovery if many users choose follower-focused feeds. Direct followers may therefore become more valuable.

Could businesses be affected?

Yes. Businesses that depend on algorithmic discovery may need to focus more heavily on direct followers, search visibility, customer relationships and other marketing channels.

Could publishers benefit?

Potentially. If users deliberately follow news organisations and follower-based feeds give those choices greater weight, publishers could gain a more direct relationship with audiences.

What penalties have been proposed?

The government has indicated that serious failure to comply with the proposed Digital Duty of Care could attract penalties reaching A$109.2 million.

Who would enforce the framework?

The independent eSafety Commissioner would have a central compliance and enforcement role under the proposed system.

Why are manipulated explicit images included in the reforms?

AI tools can be misused to create fake sexually explicit images of real people. The proposal seeks stronger regulatory tools for responding to services involved in this type of abuse.

Does the proposal affect freedom of expression?

Supporters argue that stronger online safety and user choice can coexist with lawful expression. Critics have nevertheless raised concerns about the scope of some wider Digital Duty of Care provisions, making definitions and safeguards important parts of the legislative debate.

What is the difference between opt-in and opt-out?

Under an opt-out system, personalisation may operate automatically until the user disables it. Under an opt-in model, personalisation would begin only after the user actively chooses it.

Why does that difference matter?

Most users retain default settings. An opt-in requirement could therefore result in considerably fewer people using personalised recommendations than an opt-out approach.

Are algorithms always harmful?

No. Algorithms provide many valuable services, including search ranking, spam detection, fraud prevention, navigation and useful recommendations. The debate focuses on design objectives, accountability and user control.

Could the reform influence artificial intelligence?

Yes. As AI systems increasingly recommend information, products and services, principles involving consent and user control could become relevant well beyond conventional social media.

What could parents gain from the reform?

Product-level safety requirements could reduce the need for parents to manage every risk manually, although family guidance and digital literacy would remain important.

What is algorithm literacy?

Algorithm literacy means understanding that the content displayed by a platform may have been selected using automated systems and behavioural information rather than appearing naturally or neutrally.

Could Australia’s reforms influence other countries?

Potentially. Governments worldwide are considering similar problems involving recommendation systems, youth safety and digital-platform accountability. Australia's experience could provide lessons for future regulation elsewhere.

What should Australians watch next?

Important developments include the outcome of consultation, the final legislation introduced to Parliament, possible amendments, rules concerning opt-in or opt-out personalisation, the definition of affected services and detailed eSafety enforcement guidance.

What is the central idea behind My Feed, My Way?

The central idea is that the user should have a meaningful role in deciding how personalised recommendations shape the default digital experience instead of leaving that decision entirely to the platform.

Why could this reform be important in the long term?

Recommendation systems influence an enormous share of modern information consumption. Establishing stronger principles of consent, safety and user control could influence not only today's social-media platforms but also future AI-powered digital services.

What is the biggest unanswered question?

The biggest question is whether the final framework will create genuine, practical user control or simply provide another option buried inside platform settings. The details of interface design, consent, enforcement and regulatory oversight will determine the difference.

Quick Enquiry

*** For Buying and selling quires you can contact us to this Mail Id: gkarthikeyan2k@yahoo.com ********* Contact Us