Where There Are Rights™

Search Legal Theories

EU digital immunity and its global extensions reveal the fault lines of modern international law. Across Europe, the UK, and beyond, lawmakers have entrenched liability shields that misidentify harm and leave victims without remedies. Instead of addressing the root causes in platform architecture and immunity structures, regulation prunes at the branches through content moderation and compliance tweaks. This paper traces how fault-based liability, a foundation of legal systems worldwide, breaks down in digital governance. It exposes the structural failures that protect platforms at the expense of people and calls for a re-anchoring of law in rights and remedies.

Regulatory Critique: EU Digital Services Act (DSA) and UK Online Safety Act (OSA)

EU Digital Immunity and Beyond: Fault Lines in Modern International Law

By Katherine Starr Expert Witness | Legal Theorist | Creator of Negligent Digital Access™

My career has been spent inside systems that claimed to protect schools, sports institutions, youth programs and I have seen, again and again, how failure hides behind formality. Policies exist, checklists are filled, and safety is assumed. Yet abuse, grooming, and trauma unfold in plain sight. These harms are not unpredictable they follow clear patterns. There are signs that were ignored, and rules and boundaries that were slowly pushed. That is how someone is allowed to manipulate an environment and exploit the very gaps created by inaction or blind trust. The Digital Services Act[i] (DSA) and the Online Safety Act[ii] (OSA) follow this same failed blueprint. These laws create the illusion of action while leaving the root of digital harm untouched: platform design.

We are not facing a moderation problem. We are facing a technology platform design problem. These platforms are not passive vessels that happen to host harm they are architectures that facilitate it. And the laws that now govern them refuse to name that truth.

The DSA and OSA attempt to address digital harm, but do so without applying the most critical standard of institutional accountability: the doctrines of negligent hiring, negligent supervision, and breach of duty. In the U.S., these doctrines are foundational to civil liability in institutional contexts. However, in the UK and EU, there is no identical terminology. Instead, the principles manifest through concepts such as ‘duty of care’ under UK tort law and ‘obligations de sécurité’[iii] under French civil law. In the UK, the common law duty of care established in cases such as Caparo Industries plc v Dickman [1990] 2 AC 605[iv] outlines foreseeability and proximity as core to liability. This foundational duty extends to scenarios involving institutional responsibility for individuals whose conduct may foreseeably cause harm. The principle was further affirmed in Home Office v Dorset Yacht Co Ltd [1970] AC 1004, where the government was held liable for its failure to supervise young offenders who went on to damage private property. This case reinforces that institutional duty arises where there is control, proximity, and a foreseeable risk of harm[v]. In many civil law countries, including France, Germany, and the Netherlands, liability is grounded in the principle of fault (faute, Verschulden) under their respective civil codes[vi]. These frameworks emphasize the foreseeability of harm and the failure to act where one has the capacity to prevent it aligning with the core structure of institutional negligence. While the doctrinal language differs, the conceptual alignment is clear: institutions digital or otherwise bear responsibility when they fail to prevent known or predictable harm within systems they control. They form the bedrock of liability in sectors where proximity to vulnerable individuals demands oversight. But in digital regulation, they are conspicuously absent.

My position papers on Negligent Digital AccessTM[vii] and Negligent DatingTM[viii] laid out precisely how these legal analogies should be applied in the online context. Platforms that allow unverified users access to intimate environments, that ignore prior reports of harm, and that refuse to implement available safety technologies, are not just overlooking risk they are enabling it. The duty to prevent foreseeable harm does not disappear because the harm occurs online.

These regulatory frameworks, in their current form, prioritize content flagging and moderation cycles while leaving the architecture of abuse intact. When regulators fine TikTok for failing to label advertisements[ix] or reprimand platforms for weak child data controls[x], while continuing to ignore the systemic design failures that make those breaches inevitable. These enforcement actions do not address what my work defines as platform-enabled foreseeable harm: the result of architectural choices made to scale vulnerability for profit. That harm begins long before any piece of content is published it begins at the point of access.

Both the DSA and OSA fail to apply the logic that governs physical institutions: if a school hires a teacher without background checks and that teacher harms a student, the school is liable. If a dating app allows a known scammer to return under a new name with no meaningful block or warning, and that user exploits another victim, the platform should bear the same responsibility. This is not theoretical. These are the same doctrines adapted for digital environments that I have applied across youth-serving institutions for over a decade.

It is no longer sufficient for laws to govern online platforms through the lens of post-hoc moderation. They must evolve to adopt access-point liability a recognition that harm begins not at the moment of abuse, but at the moment of access. That is where negligence lives. And unless these regulatory frameworks explicitly recognize and address design as conduct not just neutral architecture then foreseeable harm will remain legally invisible and systematically unaddressed.

The failure of the DSA and OSA is not in their ambition but in their framework. They regulate behavior rather than structure, shield design from scrutiny, and defer to procedural audits instead of imposing standards of care. My legal theories grounded in both institutional precedent and digital application offer a roadmap toward meaningful accountability. These laws should not merely define digital safety. They should enforce it structurally, preventively, and with full recognition of the duty to protect that every access-granting entity holds, whether physical or digital.

Until they do, they will continue to punish only the visible actor and never the platform that built the stage.

What the Law Protects and What It Fails to See

The structural flaw in both the DSA and OSA is not just what they ignore but what they actively protect. These laws enshrine limited liability models that shield platforms from responsibility for the very conditions that enable harm. In the EU, Article 6 of the DSA[xi] provides immunity to platforms for third-party content so long as they do not play an active editorial role and act quickly upon notification of illegal content. In the UK, the OSA claims to impose a duty of care, but this duty is narrowly framed and does not create a private right of action. Platforms are not exposed to civil litigation by those harmed. Instead, enforcement falls to Ofcom[xii], which focuses on compliance with procedural obligations, not accountability for harm.

These frameworks rely on a flawed assumption: that harm is a matter of content, not conduct of user behavior, not platform structure. As long as a platform has a system in place for content flagging and takedown, it is considered compliant, regardless of whether its architecture consistently places users in harm’s way. This regulatory approach turns procedural minimalism into a defense. The architecture of exposure fast-trust features, anonymous access, disappearing messages, frictionless contact with vulnerable individuals is treated as legally neutral.

What I argue in my position papers is fundamentally different. I am not seeking to regulate speech. I am not asking for platforms to be held accountable for what users say. I am asking that they be held accountable for what they build for how their systems are designed to ignore known risks, override protective boundaries, and create pathways that are predictably exploited. My work reframes harm not as an accident of content, but as a foreseeable outcome of access.

Negligent hiring, negligent supervision, and breach of duty are not theories about what was said they are theories about what was enabled. These doctrines do not criminalize speech; they define liability when an institution fails to act on what it knew or should have known. In a digital context, this means liability arises not from a single post or message, but from the creation of an ecosystem in which those messages are likely to lead to abuse, grooming, coercion, or manipulation.

The current legal shields in both the EU and the UK do not reach this level of scrutiny. They insulate platforms as long as they react quickly and do not directly produce the harm. But the failure to interrogate the system that allowed the harm in the first place is a dereliction of legislative responsibility. It is not that these laws don’t go far enough it is that they fundamentally misunderstand where accountability lives. The fines issued whether for ad transparency failures or weak data protections are attempts to look decisive without being transformative. These financial penalties are symbolic gestures. Regulators levy them not because they understand how to fix the problem, but because they don’t. When a platform is fined, the assumption is that money will motivate change. But change cannot come when the root cause is never named. The flaw is not insufficient labeling or delayed takedown. The flaw is in the system’s construction in how it allows unverified access, fast-trust environments, and manipulation at scale. That is where harm begins, and until regulators are willing to confront that, they will continue treating the wrong wound with the wrong remedy. Fining a platform for procedural violations while leaving its harm-producing architecture untouched is like fining a predator in the hope that they won’t do it again. It mistakes consequence for correction. It implies accountability without demanding structural reform. This is not deterrence. It is theater.

The Digital Services Act (EU)

The DSA presents itself as a milestone in online accountability. But it is a milestone built in the wrong direction. Risk assessments, algorithmic transparency, and voluntary compliance mechanisms are not safeguards they are distraction strategies that feign oversight without demanding structural change. Grooming, disinformation, and algorithmic exploitation are acknowledged but nowhere in the public record has any regulatory body in the EU or UK demonstrated a meaningful or measurable reduction in these harms through enforcement. There is no documented case where a platform’s architecture was restructured to prevent grooming. No regulator has compelled the removal of frictionless onboarding, anonymous access, or fast-trust intimacy mechanisms that enable these harms. Fines and investigations have been levied for procedural violations, but they do not equate to prevention. What has been prosecuted is not grooming by design but rather the failure to disclose ads or delay in takedown. The actual environment in which grooming thrives remains untouched. but the DSA remains silent on the very systems that make these harms inevitable: anonymous user access, rapid trust acceleration, and features designed to eliminate traceability.

The European Commission’s own guidance underlines the need for systemic risk mitigation related to mental health and the protection of minors (Regulation (EU) 2022/2065, Articles 34–35)[xiii], yet stops short of naming design features as central to those risks. Disappearing messages, no-verification onboarding, and frictionless cross-platform connectivity are left untouched despite overwhelming data, including academic studies from the Oxford Internet Institute and EU Kids Online, confirming their role in exploitative outcomes[xiv].

By emphasizing protection of minors, the DSA ignores the continuum of digital vulnerability that spans all ages. Adults particularly those isolated, grieving, or navigating emotional loss are consistently targeted and manipulated through the same platform mechanics. These harms are not less severe because the victim is older; they are equally destructive, equally preventable, and equally born from structural negligence. To rely on platforms to assess and regulate their own risk is to repeat the institutional failings of the past the same deference that allowed athletic departments, schools, and religious institutions to conceal misconduct for years under the guise of self-governance.

The DSA does not protect. It performs. It replicates the same cultural denial I have spent my professional life confronting where systemic failure is disguised as responsible oversight.

III. The Online Safety Act (UK)

The UK’s OSA reinforces urgency in language but erodes it in practice. It centers its authority around child protection while ignoring that the same platforms facilitating abuse of minors are also enabling coercion, manipulation, and exploitation of adults. Like the DSA, it refuses to regulate design. It focuses on post-harm response, on content flagging and moderation timelines, but never compels platforms to dismantle the features that cause the harm in the first place.

In a 2023 Ofcom briefing, nearly one in three online harm reports involved adults over the age of 30, with romance scams and coercive manipulation via dating apps as major drivers[xv]. Yet the OSA does not require platforms to implement preventative controls for these scenarios. When a woman in her 50s is groomed and defrauded through a dating app when her vulnerability is exploited through mechanisms built to accelerate emotional intimacy and obscure identity that outcome is not incidental. It is a foreseeable, preventable result of deliberate product design. Yet the OSA offers no language, no framework, no consequences for platforms that engineer the very conditions in which harm becomes routine. The failure is not in how fast a message is removed. The failure is in how easily it was allowed to happen.

The OSA repeats the same error we’ve seen in institutional response: it narrows the definition of risk, limits accountability to edge cases, and ignores the conditions that create the harm. That is not safeguarding. That is systemic evasion.

Harm is Not Age-Based. Negligence is Not Contextual.

We must abandon the fiction that safety is age-bound. Harm is not defined by how old you are. It is defined by the system that exposes you, isolates you, and fails to intervene when coercion unfolds in real time. The digital environment that grooms a fourteen-year-old through vanishing messages and suggestive language is the same environment that grooms a forty-eight-year-old through emotional isolation and relational trust. The failure is architectural, not incidental.

When we blame the actor, we arrest the user. When we blame the structure, we regulate the system. The first provides punishment. The second offers prevention. Platforms are not neutral. They are engineered spaces with behavioral outcomes built in. And when those outcomes repeatedly include grooming, coercion, financial abuse, and emotional collapse, then those platforms are not just tools of harm they are manufacturers of it.

Law That Knows Better Must Do Better

The DSA and OSA codify a performance of concern without enforcing a model of accountability. They define threats without addressing the designs that house them. What we need are laws that do not simply describe harm but disrupt the environments that produce it. We need structural reckoning not self-regulated transparency. We need a legislative framework that treats platform design with the same legal scrutiny we apply to institutional negligence in the physical world.

I have testified, written, and advocated across multiple sectors of harm and the pattern is always the same. The failure was not that people did not know. The failure was that people knew and chose not to act. These laws now carry that burden. The evidence is plain. The design is the problem. And the law must catch up before harm continues not as anomaly, but as consequence.

Citations

[i] European Commission. “The Digital Services Act.” European Commission, 2022.
https://commission.europa.eu/strategy-and-policy/priorities-2019-2024/europe-fit-digital-age/digital-services-act_en

[ii] UK Government. “Online Safety Act: Explainer.” GOV.UK, 2023.
https://www.gov.uk/government/publications/online-safety-act-explainer/online-safety-act-explainer

[iii] French Civil Code. Article 1240 (formerly Article 1382).
Imposes liability for harm caused by one’s fault — interpreted to include obligation de sécurité, especially in institutional or contractual settings.

[iv] Caparo Industries plc v Dickman [1990] 2 AC 605 (HL).
Established the three-part test for duty of care in UK tort law: foreseeability, proximity, and whether it is fair, just, and reasonable to impose liability.

[v] Home Office v Dorset Yacht Co Ltd [1970] AC 1004 (HL).
Held the UK government liable for failing to supervise young offenders who caused foreseeable harm, reinforcing institutional duty of care where control and proximity exist.

[vi] Donoghue v Stevenson [1932] AC 562 (HL).
Established the foundational “neighbor principle” in UK tort law that one must take reasonable care to avoid acts or omissions likely to injure foreseeable others.

[vii] Katherine Starr, Negligent Digital Access: When Platform Design Enables Harm, KStarr Enterprises, LLC (2025), https://www.katherinestarr.com/negligent-digital-access/.

[viii] Starr, K. (2025). Negligent Dating exposed: How a billion-dollar industry fails to protect. KStarr Enterprises, LLC. https://www.katherinestarr.com/negligent-dating/

[ix] European Commission. “EU takes action against TikTok over online content rules.” Financial Times, May 15, 2025.
https://www.ft.com/content/6feebcf5-a1cb-4507-a691-211173dc84d6

[x] Information Commissioner’s Office (ICO). “Twitter reprimanded for failing to protect children’s data.” ICO.org.uk, January 2023.
https://ico.org.uk/about-the-ico/media-centre/news-and-blogs/2023/01/twitter-reprimanded-for-failing-to-protect-children-s-data

[xi] Regulation (EU) 2022/2065 of the European Parliament and of the Council of 19 October 2022 on a Single Market for Digital Services (Digital Services Act), Article 6.

https://eur-lex.europa.eu/eli/reg/2022/2065/oj

[xii] Ofcom. “How we will regulate the Online Safety Act.” Ofcom.org.uk, 2023.
https://www.ofcom.org.uk/online-safety/illegal-and-harmful-content/how-we-will-regulate

[xiii] Regulation (EU) 2022/2065 of the European Parliament and of the Council of 19 October 2022 on a Single Market for Digital Services (Digital Services Act), Articles 34–35.
https://eur-lex.europa.eu/eli/reg/2022/2065/oj

[xiv] Livingstone, S., Stoilova, M., & Nandagiri, R. “Children’s Data and Privacy Online: Growing Up in a Digital Age.” London School of Economics and Political Science / EU Kids Online, 2019.
https://www.lse.ac.uk/media-and-communications/assets/documents/research/projects/childrens-privacy-online/Childrens-Data-and-Privacy-Online-Report-2019.pdf

Oxford Internet Institute. “Youth and Digital Technologies: A Review of the Evidence.” University of Oxford, 2022.
https://www.oii.ox.ac.uk/publications/youth-digital-technologies-review.pdf

[xv] Ofcom. Online Experiences Tracker – Wave 7, 2023 (Set 2). Ofcom.org.uk, 2023.
https://www.ofcom.org.uk/research-and-data/media-literacy-research/online-experiences-tracker

How to Cite This Paper:

APA (7th Edition)

Starr, K. (2025). Systemic blindness: How EU and UK law misidentify harm and protect the platforms that enable it. KStarr Enterprises, LLC. https://www.katherinestarr.com/systemic-blindness/

MLA (9th Edition)

Starr, Katherine. Systemic Blindness: How EU and UK Law Misidentify Harm and Protect the Platforms That Enable It. KStarr Enterprises, LLC, 2025. www.katherinestarr.com/systemic-blindness/.

Bluebook (Legal)

Katherine Starr, Systemic Blindness: How EU and UK Law Misidentify Harm and Protect the Platforms That Enable It, KStarr Enterprises, LLC (2025), https://www.katherinestarr.com/systemic-blindness/.

 

About the Author

Katherine Starr is a Legal Theorist and Expert Witness specializing in institutional negligence, platform accountability, and digital harm architecture. She is the originator of Negligent Digital Access™, Negligent Digital Architecture™, Negligent Dating™, and the Digital Maritime Doctrine™ — a series of original legal frameworks designed to expose systemic design failures across digital platforms. Her work draws on direct case experience, policy critique, and lived expertise in institutional misconduct.

Katherine Starr

Articles and Insights on Digital Harm

Contact me

Let's Start Talking

Feel free to reach me via email or schedule an appointment through my customer portal.

I’m available to discuss a variety of topics, including athlete life coaching, book series adaptations, motivational speaking or sexual abuse expertise.