Book

As generative AI systems absorb personal data into billions of parameters, and as blockchain ledgers lock transaction histories into place forever, regulators and technologists alike are asking the same uncomfortable question: once data is in, can it ever really come out? The European Data Protection Board weighed in with fresh guidance at the end of 2025. Commentators have declared erasure "impossible" in the age of large language models. And yet, this book argues, the panic is premature — the problem was never unsolvable. It was simply never designed for. Blockchain promises trust without intermediaries. Data protection law promises dignity through control. This book confronts what happens when those two promises collide: the fault line between blockchain's architectural commitment to permanence and Article 17 of the GDPR — the individual's right to erasure. It is a conflict as old as Bitcoin itself, yet one that has never felt more urgent than now, as blockchain-anchored data increasingly feeds the very AI training pipelines at the center of today's privacy reckoning. Moving fluidly across cyber law, information systems design, and corporate governance theory, the authors dismantle the false binary that has trapped so much of the public conversation: the assumption that immutable technology and enforceable privacy rights are fundamentally incompatible. Instead, they show that immutability exists on a spectrum, that erasure is a qualified right rather than an absolute one, and that the real question facing every enterprise isn't "how do we delete the undeletable" — it's "how do we build systems where deletion was never a crisis to begin with." At the heart of the book is a five-layer compliance-sensitive architecture — governance and controllership, data categorisation, cryptographic design, consent management, and audit infrastructure — built as a practical playbook for executives and system architects navigating this exact moment. And its sharpest twist is this: the very immutability that makes deletion hard is what makes blockchain-based AI systems auditable, contestable, and trustworthy.