Australia’s algorithm off‑switch: what it changes for users and platforms
According to BBC News, the Australian government will introduce a law that forces social‑media companies to give every user a clear option to turn off recommendation algorithms, showing only posts from accounts they follow.
The move follows years of complaints that algorithm‑driven feeds push harmful content – from dangerous diet challenges to extremist rhetoric – into users’ screens, often despite safety settings.
The proposed algorithm switch
The draft Digital Duty of Care bill requires platforms to provide an "algorithm switch" that, when activated, disables the personalised recommendation engine. Users would then see a chronological or "following‑only" feed. The law also adds a duty of care, meaning services must take reasonable steps to keep their environments safe for children and for adults against "seriously harmful" material. Breaches could attract fines up to A$100 million.
How recommendation feeds work today
Most platforms use machine‑learning models that analyse a user’s past likes, comments, watch time, and network connections. The model predicts which posts are most likely to keep the user scrolling, then orders the feed accordingly. This process happens in seconds on the server and is invisible to the user.
The benefit is that users are exposed to a steady stream of content that feels relevant, even if they do not follow the original creators. The downside is that the system optimises for engagement, not wellbeing. Content that provokes strong emotions – outrage, fear, desire – tends to generate more clicks, so the algorithm amplifies it.
What changes – the practical trade‑offs
| Feature | Algorithmic (current) | Following‑only (proposed) |
|---|---|---|
| Content relevance | Tailored by past behaviour; high engagement | Limited to accounts you follow; may feel less “fresh” |
| Exposure to new creators | Frequent; platforms push new videos, posts | Rare; only new posts from existing follows appear |
| Mental‑health impact | Studies link to anxiety, addiction‑like brain responses | Potentially lower exposure to triggering material |
| Platform revenue | Higher ad impressions and longer session times | Shorter sessions, fewer ad slots |
| User control | Settings exist but are hidden, often reset on app reopen | Explicit on/off switch; must stay active to work |
The biggest trade‑off is between engagement‑driven revenue and user safety. Turning the algorithm off may reduce the time users spend on the app, cutting ad revenue. Platforms argue that the switch could break features that rely on algorithmic ranking, such as “Explore” pages or recommended stories, and could even limit direct‑message functionality, as seen on Instagram where the feed toggle also disables certain UI elements.
Who wins, who loses
Users: People vulnerable to harmful content – for example, those with eating disorders or mental‑health conditions – gain a tool to limit exposure. However, casual users may miss out on discovering new creators or trending topics they would otherwise enjoy.
Platforms: Large firms (Meta, TikTok, X, Snap) face additional development work to meet the requirement and risk reduced engagement metrics. Smaller platforms might find the rule a level‑playing field if they already rely on simpler feeds.
Advertisers: Fewer impressions could lower the value of ad inventory, prompting a shift toward more targeted, higher‑price placements.
Regulators: The law gives a concrete lever to enforce online safety, but enforcement will be challenging. Past attempts, such as the EU’s 2023 algorithm controls, showed low uptake of optional switches.
The hidden catch and what to watch next
The legislation promises that the switch must be "obvious" and that turning it off should not degrade core functionality. In practice, platform developers may implement the feature in a way that nudges users to keep it on – for example, hiding the toggle deep in settings or resetting it after a period of inactivity. Early tests in the Netherlands, where a court forced Meta to keep the choice, revealed that users often ignore the option because the default algorithmic feed is more convenient.
Another subtle issue is the definition of "seriously harmful" content. The bill distinguishes between material that is harmful to children (broad) and to adults (narrow). Determining what falls into each bucket will likely involve lengthy legal battles, potentially delaying enforcement.
Watch the following developments:
- Implementation guidelines – the Australian Communications and Media Authority will release technical standards. The clarity of those guidelines will determine whether the switch truly works or becomes a token compliance step.
- Industry response – if major platforms lobby for amendments or propose alternative safeguards, the final law could look different from the draft.
- User uptake data – early adoption rates will reveal whether people actively use the off‑switch or let the default remain.
- Cross‑border pressure – success in Australia could spur similar moves in Canada, the UK, and other jurisdictions, creating a de‑facto global standard.
Practical steps for everyday users
- Check your app settings now and note where the feed control is located. Even if the switch isn’t mandatory yet, many platforms already offer a "following only" mode.
- If you experience distress from certain content, use platform‑provided tools to hide or report specific posts, and add relevant keywords to block lists.
- Consider a short‑term break from scrolling when you feel a binge‑watch habit forming. A timer or a phone‑level screen‑time limit can act as a manual off‑switch.
- For those managing mental‑health conditions, discuss social‑media habits with a professional and explore safe‑use plans.
By keeping an eye on how the law shapes platform behaviour, users can decide whether to rely on the new switch or continue using the existing algorithmic experience.
