Meta logo illustration (Chee, 2024)

What happens when the only way to avoid surveillance on social media is to pay for it? When it comes to Meta’s “pay or consent” model, it is easy to think of this as a straightforward choice, but it is also a question of privacy, rights and power. If we have to pay to say no to personalised ads, then privacy is no longer a right. It starts to look like a premium feature.

A simple choice or a deeply unfair one?

Meta’s pay or consent model looks straightforward at first. You can pay for an ad-free version of Facebook or Instagram, or keep using the service for free and agree to personalised advertising in exchange for the use of your personal data.

The difficulty is that regulators have already warned that personal data is not a commodity. If the right to data protection is a paid service, privacy is not being used to provide minimum protections anymore (EDPB, 2024). This is precisely the case in Meta’s model. It does not provide an alternative payment service. It is the default model, and a subscription option to opt out.

So, this is more than one subscription option. It is a question of whether privacy in the time of platform services is still a right or if it is being turned into a commodity.

Why does this matter more than one subscription option?

It is not only cost or user experience. It is, in fact, concerning how the digital platforms control engagement in society. Social media platforms “are the spaces that connect and communicate, but also the spaces that govern communication” (Suzor 2019, p. 12). This is because Meta is not just a company competing in a marketplace. It is managing access to a digital environment that is essential to many people. Meta has infrastructural power because users are interacting with a common service that is also a common platform for communication and social participation. It’s business decisions, then, that also become rules about how users can connect and stay connected.

In this context, the pay or consent model is not a business model. It is a glaring instance of platform governance by a private company that sets the terms of engagement for users of a digital platform.

The biggest problem is the word “choice”

Meta’s best defence is the freedom of the term ‘choice’. It can be claimed that users are being given choices and that, therefore, consent is meaningful.

But the press and regulators have called that line of argument into question. The report on the European case notes that the binary choice requires users to consent to the combination of their data, and does not offer a less personal but equivalent alternative version of Meta’s social networks (Chee, 2024). It is not a technical quibble. It is fundamental to whether there is really a right to refuse.

The European Commission’s preliminary opinion also reinforces this argument. This binary choice compels consent and does not provide a less personalised but otherwise equivalent service (European Commission, 2024). As these responses demonstrate, the issue is not whether users are confronted by two buttons. The issue is that the terms of the choices are already biased.

For many users, Facebook and Instagram are not digital conveniences. They are integral to family life, academic communities, business promotion and cultural exchange. Opting out may be legally optional, but it can be socially expensive, hence the misleading use of the word “choice”.

Privacy should not be a commodity.

This is where a digital rights perspective becomes essential. The best way to respond to Meta’s model is not to say you don’t like the price, but that the model shifts the understanding of privacy.

Human rights-based policies are increasingly being seen as an alternative to more market-oriented systems rather than just as a binary choice between rights and regulation (Karppinen, 2017, p. 99). This is a useful way to understand Meta’s model because it reveals what is at stake. Rather than privacy limiting the business model, privacy is part of the business model and subject to negotiation within the market.

This is even more important in terms of how platforms create norms. Chen and Cheung (2018, p. 288) demonstrate how the ubiquity of WeChat in everyday communication allows it to manage users’ privacy and create a new privacy culture. Meta’s model works similarly. It does not just control data collection. It also helps to normalise the concept that surveillance is the default state of being, while privacy becomes optional, conditional or for sale.

So a consent screen might appear neutral, but it is also encouraging users to accept a new privacy paradigm where we have access to the platform first and the right to privacy second.

It is also a class issue, not just a privacy issue

The pay or consent approach is usually discussed as if it is a universal choice. But this is not the case.

Marwick and Boyd (2018, p. 1158) put this point well when they note that “privacy is particularly challenging for those who are marginalised in other domains”. This is important in the current context because while a subscription fee might be affordable for some users, it is not the same for students, low-income users, migrants, part-time workers, and so on.

What appears to be a neutral design decision then creates a two-class system on the internet. Some users can choose to be less surveilled, but others are stuck in the full behavioural advertising system because they cannot pay to get out of it. This is why the model proposed by Meta is not a privacy issue. It is also a class issue because it is structurally built into the digital economy.

Meta is not selling ads. It is selling access to our behaviour

To see why this is an issue, it is helpful to think beyond the interface and consider the bigger picture of the advertising system.

Meta’s system is more than just a transaction between one person and one piece of advertising. According to the ACCC (2021), ad tech allows a very large number of publishers to sell digital ad inventory to a very large number of advertisers. It is part of a broader system of commerce that delivers value to the digital advertising market through data, targeting and matching. This is datafication, in which social practices are constantly turning into behavioural data that can be tracked, measured and sold.

Behavioural advertising is not just a matter of convenience. The EDPS (2018, p.3) advances this point by noting that a major business model on the internet now involves tracking people online, collecting information about their personality, health, relationships and opinions and beliefs. This means that the ulterior motive is to monetise digital advertising. It relies on extensive tracking and inference, which means that the product is not just about attention. It is the ability to know, predict and sell people’s behaviour.

In this sense, Meta is not selling ads. It is selling access to the data that makes people knowable, sortable and targetable.

Here is where we see platform governance

The pay or consent model also renders platform power more visible because it renders the relationship between commerce and governance more visible.

Meta is not operating as a platform. As Flew (2018, p.28) puts it, large platform companies are increasingly monitoring, regulating and removing content, as well as blocking and banning users editorially. Meta, thus, is already engaging in governance activities that determine visibility, access and norms, and while appearing as a private business offering choices to consumers.

This is the reason why we should think of the pay or consent option as a governance strategy, not a subscription option. It sets the conditions for users to continue to be visible and connect within a private digital infrastructure. Once that becomes visible, the key question changes. It becomes a matter of whether that is the default option. The question is whether large platform services should be permitted to make surveillance the default option for engagement.

Why regulators are pushing back

The most substantive regulatory challenge to Meta’s approach concerns whether or not it offers a real choice. The ICO (2025, p.28) makes clear that there needs to be equivalence between the service available under the consent option and that available under the pay option if users are to make a freely chosen decision about consent to personalised advertising. This is a vital criterion because, without a genuinely equivalent alternative, users are not being offered an equivalent choice. They’re being forced to choose the option that preserves the platform’s surveillance-for-profit business model.

So, it is not just the cost of the subscription that is problematic. The fundamental issue is that the user is being put in the position of having to take responsibility for their privacy. A rights-based approach would start with the assumption that nobody should have to buy into effective protection from profiling.

This is more than about privacy. It’s also about automation and platform power.

While this case is most obviously about privacy and digital rights, it also has implications for automation, datafication and algorithms.

The meta model cannot be dismissed as just one product design decision. Nissenbaum (2015, p.832) suggests that privacy threats come from sociotechnical systems, which are technologies embedded in environments influenced by social, economic and political dynamics. These are designed for particular purposes. Similarly, Meta is part of a larger system in which data extraction, algorithmic classification, and commercial optimisation are all embedded in the design of the platform. This is its second problem of algorithmic governance, since the infrastructures that enable ad targeting are also used to rank, show visibility, exposure, and prominence on the site.

Infrastructures that facilitate targeted advertising, therefore, can also affect what is viewed, what is amplified, how individuals are categorised, and how harms are exaggerated using the platform. Consumer privacy is not only about this sense. It demonstrates the support of platform power by data collection more broadly.

An improved internet cannot be a luxury item.

When behavioural advertising is made optional, and you must pay, you have already lost the game. It has institutionalised surveillance and shifted the burden to the consumer.

Another reason is evidenced by Australian research on the topic of digital rights. Goggin et al. (2017) point out that around 78% of users are interested in knowing how social media companies use their personal data. This brings out the expectation to be more open, having more control among the people. Individuals are concerned with information. They already want more than the platform accountability.

The underlying principle of rights is even more obvious in this statement of the UN General Assembly. Human dignity and personal autonomy require the right to privacy, which upholds freedom where each person has the right to develop and to express their own personality (OHCHR, 2025). Here, the market status of individual users in the commercial platforms should not be a determinant of privacy. Online activity should at least be a requirement.

A more appropriate web would begin with a less data-heavy design, non-profiled options and improved access control and data control by dominant platforms.

Final takeaway

The obvious advantage of the pay or consent model of Meta is that it is too ordinary. It appears to place a simple option at the fingertips of the user. But behind the interface is a far larger struggle concerning the digital rights, governing platforms and the political economy of data.

Privacy is negotiable in the model of Meta. It entraps behavioural advertisement. It is costly to choose otherwise. And it unveils how the corporations utilise the private law to regulate the regulations of interaction in the virtual world. In case privacy is a service that users are obliged to pay for, then it is no longer a right. It is a luxury. Once privacy is a luxury, the internet no longer remains a marketplace. It rather becomes a domain of privacy, where privacy safeguards can be afforded by few.

References

ACCC. (2021). Digital Advertising Services Inquiry: Final report, Australian Competition and Consumer Commission. https://www.accc.gov.au/system/files/Digital%20advertising%20services%20inquiry%20-%20final%20report.pdf

Chee, F. Y. (2024, July 1). Meta’s pay or consent model in crosshairs for breaching EU tech rules. Reuters. https://www.reuters.com/technology/meta-charged-with-failing-comply-with-eu-tech-rules-2024-07-01/

Chen, Z. T., & Cheung, M. (2018). Privacy perception and protection on Chinese social media: a case study of WeChat. Ethics and Information Technology, 20(4), 279–289. https://doi.org/10.1007/s10676-018-9480-6

EDPB. (2024). Opinion 08/2024 on Valid Consent in the Context of Consent or Pay Models Implemented by Large Online Platforms. European Data Protection Board. https://www.edpb.europa.eu/system/files/2024-04/edpb_opinion_202408_consentorpay_en.pdf

EDPS. (2018). Opinion 3/2018 EDPS Opinion on online manipulation and personal data. European Data Protection Supervisor. https://www.edps.europa.eu/sites/default/files/publication/18-03-19_online_manipulation_en.pdf

European Commission. (2024). Commission sends preliminary findings to Meta over its “Pay or Consent” model for breach of the Digital Markets Act. https://dpo-india.com/Resources/Digital_Markets_Act(DMA)_EU/Commission-sends-preliminary-findings-Meta%E2%80%9CPay-Consent%E2%80%9Dmodel-breach-Digital-Markets-Act.pdf

Flew, T. (2018). Platforms On Trial. International Institute of Communications, 46(2), 22–29. https://www.iicom.org/wp-content/uploads/im-july2018-platformsontrial-min.pdf

Goggin, G., Vromen, A., Weatherall, K., Martin, F., Webb, A., Sunman, L., & Bailo, F. (2017). Digital Rights in Australia. Departments of Media and Communications, Government and International Relations, Faculty of Arts and Social Sciences, and the University of Sydney Law School. https://ses.library.usyd.edu.au/bitstream/handle/2123/17587/USYDDigitalRightsAustraliareport.pdf

ICO. (2025). Consent or pay: Summary of call for views. Information Commissioner’s Office. https://ico.org.uk/media2/d5bn5kj1/consent-or-pay-summary-of-call-for-views.pdf

Karppinen, K. (2017). Human rights and the digital. In H. Tumber & S. Waisbord (Eds.), The Routledge Companion to Media and Human. Routledge. https://doi.org/10.4324/9781315619835-9

Marwick, A. E., & Boyd, D. (2018). Privacy at the Margins| Understanding Privacy at the Margins—Introduction. International Journal of Communication, 12, 1157–1165. https://ijoc.org/index.php/ijoc/article/view/7053/2293

Nissenbaum, H. (2015). Respecting Context to Protect Privacy: Why Meaning Matters. Science and Engineering Ethics, 24(3), 831–852. https://doi.org/10.1007/s11948-015-9674-9

OHCHR. (2025). The right to privacy in the digital age. United Nations High Commissioner for Human Rights. https://documents.un.org/doc/undoc/gen/g22/323/93/pdf/g2232393.pdf

Suzor, N. P. (2019). Who Makes the Rules? In Lawless: the secret rules that govern our digital lives (pp. 10–24). Cambridge University Press.