Meta, the company that runs Facebook and Instagram, has announced plans to repurpose most of the personal data that they ever collected about you, to train their “artificial intelligence (AI) technologies” — without, of course, asking your permission to do so.

Our challenge to Meta’s plans

Their move lands as gently as a bull in a china shop: AI is a promising but highly experimental technology which has already proven its ability to fail big and ruin peoples’ lives. The unsafe deployment of AI and other algorithmic tools has already resulted in people being wrongly accused for sexual harrassment, wrongly convicted for being black, wrongly investigated for fraud, discriminated against in their A-level exam results for being poor, or simply run over by a “self-driving” car.

Despite the obvious and well-documented risks, Meta doesn’t seem too bothered about turning all of us into involuntary (and unpaid) test subjects for their experiments — after all, what could possibly go wrong when an unproven and highly defective technology meets a ‘move fast and break things’ culture?

Fortunately, both European and UK data protection law protects us from illegal, unfair and harmful uses of our personal data. This is why Open Rights Group has joined None of Your Business (NOYB) and raised a formal regulatory complaint in the UK against Meta. NOYB’s intervention has already forced Meta to halt their plans in the EU and the UK, but their commitment is not legally binding and nothing, as it stands, would prevent them from resuming their plans at any point in time. Thus, we are asking the Information Commissioner’s Office (ICO) to urgently adopt an enforcement order to prevent Meta from trumping our data protection rights, and using our data to train their artificial intelligence without our consent.

WHAT ARE META’S PLANS?

With the latest update of their privacy policy, Meta has stated their intention to feed all the data they hold (about you) on their Facebook and Instagram platforms for the purposes of training “AI technology”. While Meta also stated that they would only use “public posts”, their privacy policy contradicts this by including any personal data collected through these platforms, with the sole exclusion of private chats.

There is more to that. Meta doesn’t tell us what kind of product or AI they will be developing, but they claim to have a “legitimate interest” for this purpose, which overrides your rights and legitimate expectations. In practical terms, this means that they are denying your right to consent or refuse to the use of your data for training their “AI technology”: instead, UK users would be asked to undergo a complicated 11-step procedure to object to Meta’s plan. Meta also does not tell you what this “AI technology” will look like or will be used for, but it requires you to explain how this new technology will impact you and how this overrides their “legitimate interest”, if you wish to object.

Finally, if you do not to object, or fail to persuade Meta to honour your objection, you will never have an opportunity to reverse this choice: as Meta states, you will Lose both your right to object to the use of any of your data which has already been fed to their system, nor you will be allowed to ask your data to be erased.

WHAT’S THE PROBLEM?

Meta’s plans are an obvious abuse of your data and a violation of European data protection standards. In short,

• Under data protection law, Meta must state what they will use your data for. By choosing not to disclose their end goals, Meta would be operating completely out of the law from the outset.

• Meta also claims that they have a legitimate interest that overrides your rights, and thus they do not need to ask for your permission. However, they do not tell what this “legitimate interest” is, nor they try to justify at any time how this mysterious “legitimate interest” overrides your rights.

• Meta has the obligation to facilitate the exercise of your data protection rights, but they force you through a lengthy 11-step procedure to exercise your right to object to personal data processing. For reference, the same result could be achieved by providing an “opt-out” link, as most marketing emails do, but Meta gratuitously chose not to do it this way.

• Finally, data protection law gives you several rights concerning your data, including that of accessing your data, having your data rectified if found to be incorrect, and have your data deleted if you wish so. Meta, however, explicitly states that you will not have the right to delete this data once they have been fed to their “AI technology”. Meta also does not explain how you would be able to exercise your right to access or rectification in relation to this new “AI technology”.

CALL THE COPS

Following Meta’s announcement, 11 regulatory complaints were immediately filed in the European Union (EU). As a result, Meta had to suspend their plans in Europe and in the UK. Interestingly, Meta also mentions that they have been in touch with the Information Commissioner’s Office (ICO), which would have formulated some “recommendations” concerning Meta’s plans.

This leads as to why we are filing a formal regulatory complaint before the ICO. On the one hand, Meta’s suspension is, as it stands, a decision which can be reversed at any moment. The ICO has, however, the power to issue a formal order to Meta to stop processing personal data for this purpose, which would be legally binding and not dependent on Meta’s mood of the day. We’re also asking the ICO to adopt this order under an urgency procedure: as you would lose your right to objection or erasure, this violation of your rights would be irreversible.

On the other hand, there is little detail about what corrective measures or “recommendations” were communicated by the ICO to Meta. As we’ve detailed above, the extent and gravity of Meta’s violation would require radical changes to their plans, including the requirement to ask for free, informed and affirmative consent before feeding your data to the facebook’s AI. By formalising our complaint, we hope to prevent the ICO—and hold them to the account if they don’t—from giving the green light to Meta’s plans unless they implement the radical changes they need in order to comply with the law.

THE UK AT A CROSSROAD

Is a purely commercial interest enough to ignore regulatory standards and exclude us exclude us from important conversations? Does innovation empower us, or trumps our agency, rights and legitimate expectations?

Meta’s plans hijack the meaning of words as “innovation” and “growth” and provides the worst, most dystopic answer to important questions that we face, now that Artificial Intelligence starts being widely adopted in many aspects of our lives.

The UK is at a crossroad: Will AI be used to our benefit or against us? In tomorrow’s King’s Speech, we’re expecting that the new Labour government will announce plans to bring in legislation that will regulate how AI is used in the UK. We need to ensure that AI is used in ways that enhance our lives rather than corporate profits, and that we have regulators that will take decisive action to protect us.

Digital Privacy

Hands Off Our Data

Find Out More