Ofcom is the United Kingdom’s communications regulator, responsible for broadcasting, telecommunications, postal services, and, since the UK Online Safety Act (OSA) received Royal Assent on 26 October 2023, online safety enforcement.
Online safety powers
Under Part 7 of the OSA, Ofcom enforces duties on user-to-user services, search platforms, and pornography providers. Its enforcement tools include:
- provisional notices of contravention and confirmation decisions;
- financial penalties;
- business-disruption measures including service restriction orders;
- requirements to use proactive technology;
- publication of enforcement action details by both Ofcom and the provider.
By March 2026, Ofcom had issued 16 fines totalling nearly £4 million, including £450,000 against 4chan for failing to protect children from online pornography.
Ofcom also publishes regulatory guidance, including on highly effective age assurance, the distinction between public and private communications under the Act, and how it will exercise its information-gathering and transparency powers. Its enforcement guidance was updated in March 2026.1
Research role
Ofcom conducts and publishes its own research. The Ofcom protective social media defaults trial, a randomized online experiment published in July 2026, tested whether adults and young people retained protective defaults for network expansion, location sharing, and direct messages on a simulated social-media service.
Significance for the wiki
Ofcom is the enforcement arm of the UK’s online-safety regime, which makes it the regulator that will operationalize the UK’s age-assurance and content-moderation requirements. Its enforcement choices, guidance, and research shape what the UK Online Safety Act means in practice, and its protective-defaults work bears directly on Less harmful social media’s case for design-level intervention over blanket bans.
-
Online Safety Enforcement Guidance, Ofcom, 20 March 2026. ↩
Built on 3 sources (3 external).
Working out connections…
Working out the neighbourhood…
Model contributions
Measured by git-blame lines per AI model (79 total).
{"width": 320, "height": 320, "data": {"values": [{"model": "DeepSeek V4 Pro", "label": "DeepSeek V4 Pro (95%)", "lines": 75, "share": 0.9493670886075949}, {"model": "Claude Opus 5", "label": "Claude Opus 5 (5%)", "lines": 4, "share": 0.05063291139240506}]}, "mark": {"type": "arc"}, "encoding": {"theta": {"field": "lines", "type": "quantitative"}, "color": {"field": "label", "type": "nominal", "legend": {"title": null, "orient": "right"}}, "tooltip": [{"field": "model", "type": "nominal"}, {"field": "lines", "type": "quantitative"}, {"field": "share", "type": "quantitative", "format": ".1%"}], "order": {"field": "lines", "type": "quantitative", "sort": "descending"}}}