Bluesky’s Next Challenge is How User Data is Used

Since Bluesky began to take off last November with a surge of new users after the US election, and as an attractive place for people looking for alternatives to the toxicity of X, the social network has been growing at an impressive pace.

It surpassed 33 million users in mid-March – a remarkable achievement for a decentralised social network still in its early stages that, less than six months ago, had just about reached 15 million users since starting up in early 2023.

As Bluesky expands, so does increasing discussion around key issues of interest for users and others, such as content moderation and user safety,.

The newest debate is around a draft proposal from Bluesky on giving its users more control over how their public data is used for AI training, web archiving, and other external applications.

Such a proposal should stimulate timely and necessary conversation. The role of user data in AI training has become a contentious issue across social platforms, with ambiguity often being the default stance of many companies. The lack of clear guidelines and transparent policies has led to waves of public concern, legal scrutiny, and pushback from users who feel they have little say in how their content is repurposed.

Against this backdrop, Bluesky’s approach – asking users directly for their input on a proposed framework – is a good move that will encourage transparency and engagement.

Bluesky’s Proposal: "User Intents for Data Reuse"

Bluesky’s draft proposal introduces a mechanism through which users can express their preferences regarding how external parties use their publicly available data. This could include AI training, web archiving, and other forms of data reuse.

The proposal suggests that users will be able to specify these preferences in their account settings, with their choices stored in a machine-readable format in their Bluesky repository. It functions similarly to the robots.txt file that websites use to guide search engine crawlers – a widely recognised but unenforceable standard that operates on good faith.

The idea is to create a clear, public signal that platforms and developers can follow when deciding how to handle user-generated content.

Why This Matters

Social networks have long profited from user-generated content, with little oversight or explicit consent regarding its use beyond the original platform. The most egregious example of use without consent is the Cambridge Analytica scandal in 2018 where personal data belonging to millions of Facebook users and their friends was collected for political advertising.

While some companies have introduced opt-out mechanisms or offered limited transparency, the overall landscape remains murky. Many users remain unaware that their content is being leveraged in ways they never anticipated.

Bluesky’s proposal represents a firm step in the right direction, not because it legally binds third parties to comply, but because it establishes an ethical precedent. It acknowledges the need for clearer communication and gives users an accessible way to signal their preferences.

The Challenge of Compliance

Of course, there are limits to how enforceable such a system can be. While Bluesky can introduce this intent-based approach, external entities – particularly AI developers and web scrapers – are under no legal obligation to honour it. Much like robots.txt, compliance is voluntary, and there will inevitably be actors who disregard user preferences.

This raises a fundamental question: should social networks be taking a stronger stance on data protection, perhaps through stricter technical barriers or legal advocacy? Or is transparency and user choice enough? These are conversations worth having as we grapple with the ethics of AI training and data use.

Open Dialogue is Better Than No Dialogue

Bluesky’s decision to solicit user feedback on this proposal is a positive move, although some initial reactions on the social network itself were far from supportive, with some claiming that Bluesky was reversing its earlier commitment not to permit use of user data for AI training purposes.

Still, instead of quietly implementing changes or allowing the status quo to persist, Bluesky is inviting discussion on what an ethical approach to data reuse should look like. This is in stark contrast to some platforms, which have faced backlash for implementing AI-related data policies without user consultation.

As social networks continue to grow, the issue of user data and its use in AI training will not go away. Bluesky’s initiative is a welcome attempt to introduce some clarity into an otherwise ambiguous landscape. It may not be a perfect solution, but it is a step towards a more transparent, ethical approach to data use.

If you're concerned about how your content is repurposed beyond the platforms your audience uses, now is the time to pay attention. The conversation is happening, and user voices will help shape the future of data ethics in social media.

How do you see this development? Share your thoughts in the comments below.