The Online Safety Act and UK Journalism: What Reporters Need to Know

What the Online Safety Act 2023 Does

The OSA creates a new regulatory framework for online services, imposing duties on platforms to protect users — particularly children — from illegal content, content that is harmful to children, and (for the largest platforms) content that is harmful to adults. Ofcom is the designated regulator, with powers to issue codes of practice, conduct investigations, and impose fines of up to £18 million or 10% of global turnover (whichever is higher).

The Act categorises online services into three tiers: Category 1 (the largest platforms with the most significant reach), Category 2A (search services), and Category 2B (smaller services meeting certain thresholds). Each tier carries different duties, with Category 1 services facing the most extensive obligations including duties around “legal but harmful” content for adults.

The Journalistic Content Exemption

After sustained lobbying from press freedom organisations, the Act includes specific protections for journalistic content. Under Section 16, Category 1 services must have systems and processes designed to ensure that “journalistic content” of democratic importance is not removed or restricted without adequate consideration of its significance.

The Act defines “journalistic content” broadly — it includes content generated for the purposes of journalism by any person, not just professional journalists employed by recognised news publishers. This is significant because it extends protection to freelancers, bloggers, and citizen journalists who produce content with a journalistic purpose.

Recognised News Publisher Status

The Act creates a distinct category of “recognised news publisher” that benefits from enhanced protections. To qualify, a publisher must meet specific criteria including:

• Having a standards code or being subject to an independent regulatory regime (such as IPSO, IMPRESS, or the BBC Editorial Guidelines)
• Having a registered office or principal place of business in the UK
• Having a named editor who has editorial control over content
• Having policies and procedures for handling complaints

Content from recognised news publishers is exempt from many of the Act's content moderation requirements. Platforms cannot remove or restrict access to content from recognised news publishers unless they have reasonable grounds to believe the content constitutes a relevant offence. This creates a two-tier system that has drawn criticism for potentially disadvantaging independent journalists and smaller outlets.

Ofcom's Enforcement Powers and Journalism

Ofcom's enforcement powers under the OSA are extensive and include the ability to issue information notices, conduct audits, publish codes of practice, and impose substantial financial penalties. For journalists, the key concern is how Ofcom will balance its duty to protect users from harmful content against the right to freedom of expression under Article 10 of the European Convention on Human Rights.

Ofcom has published codes of practice that platforms must follow (or demonstrate equivalent measures). These codes include specific references to the protection of journalistic content, and Ofcom has committed to considering press freedom implications in all its enforcement decisions. However, critics argue that the sheer volume of content moderation required may lead platforms to over-remove content — including legitimate journalism — to avoid regulatory risk. For background on Ofcom's broader role, see our guide to the Ofcom Broadcast Code.

Implications for Online News Publishers

Online news publishers face several practical implications under the OSA:

• Comment sections: If your website allows user-generated comments, it may fall within the scope of the Act as a “user-to-user service.” You would then be subject to duties around illegal content and potentially harmful content, depending on your size and reach.
• Distribution via social media: When you publish stories on social media platforms, those platforms must apply the journalistic content protections before removing or restricting your content.
• Audience engagement tools: Forums, live blogs, and interactive features may trigger OSA obligations if they allow user-generated content.
• Archive content: The Act applies to existing content as well as new publications. Review legacy content for compliance with the new framework.

Social Media Reporting Under the OSA

Journalists who report on social media content face particular challenges under the OSA. The Act's provisions on illegal content include specific offences such as the sharing of intimate images without consent, cyberflashing, and the encouragement of self-harm. When reporting on these issues, journalists must balance the public interest in exposure against the risk of amplifying harmful content.

The Act also creates new communications offences — including a “false communications” offence and a “threatening communications” offence — that replace and expand upon the old Section 127 of the Communications Act 2003. Journalists should be aware that these offences are broadly drafted and could potentially catch certain forms of undercover or investigative reporting if not handled carefully.

User-Generated Content and Platform Liability

The OSA fundamentally changes the liability landscape for platforms hosting user-generated content. Under the old framework (primarily the E-Commerce Directive as retained in UK law), platforms benefited from broad hosting immunity provided they acted expeditiously to remove illegal content upon notification. The OSA imposes proactive duties — platforms must now take steps to prevent users from encountering illegal content and must have systems in place to identify and remove it.

For journalists, this creates both opportunities and risks. On one hand, platforms may be more responsive to takedown requests for content that harasses or threatens journalists. On the other, the pressure on platforms to remove content proactively may lead to the over-removal of legitimate journalistic material, particularly investigative content that includes references to criminal conduct, extremism, or other sensitive topics.

Ofcom's Regulatory Timeline and What's Live Now

The Online Safety Act received Royal Assent in October 2023, but its provisions are being brought into force in stages. Understanding what is actually live — as opposed to what has been legislated but not yet commenced — is essential for accurate reporting:

• Illegal content duties (commenced 2024): Platforms must now take steps to prevent users from encountering illegal content, including child sexual abuse material (CSAM), terrorism content, fraud, and the new communications offences introduced by the Act itself. These duties apply to all in-scope services regardless of category.
• Children's safety codes (commenced 2025): Ofcom published its Children's Safety Code in 2024, and Category 1 and 2 services had a compliance deadline in 2025. These codes include age verification requirements and restrictions on recommending harmful content to children.
• Transparency reports: Category 1 services must now publish annual transparency reports disclosing their content moderation decisions, including the number of journalistic content takedowns and the outcomes of appeals.
• Adult content duties (phased): The most controversial element — duties on Category 1 services around “legal but harmful” content for adults — is still being phased in. This area has significant implications for investigative reporting that covers difficult but lawful topics.

Legal but Harmful Content: What It Means for Journalism

The concept of “legal but harmful” content was among the most debated aspects of the OSA's passage through Parliament. Critics, including many press freedom organisations, warned that requiring platforms to moderate content that is legal could have a chilling effect on journalism covering difficult but newsworthy topics.

Under the Act as passed, Category 1 services must:

• Have regard to certain categories of “priority content that is harmful to adults” (such as content promoting eating disorders or encouraging self-harm) in their systems and processes
• Give users tools to control their own exposure to such content
• Be transparent about their policies for moderating such content

For journalists, the concern is that platforms responding to these duties may over-remove content that discusses harmful topics in a journalistic context — for example, investigative pieces on anorexia, suicide, or extremism. The journalistic content exemption in Section 16 should provide protection, but its practical application by content moderation teams (often automated systems with limited editorial sophistication) is uncertain. Media organisations should monitor their content removals carefully and maintain records to challenge unjustified takedowns.

Practical Steps for UK Journalists

1. Understand your publisher's status: Check whether your employer or client qualifies as a “recognised news publisher” under the Act. If so, your content benefits from enhanced protections on platforms.
2. Document content removals: If a platform removes your content, screenshot the notification, note the reason given, and use the expedited appeals process. Keep records for potential legal challenges.
3. Review your own platforms: If you run a personal blog or news website with comment sections or forums, assess whether it falls within scope of the OSA and take steps to comply.
4. Stay updated on Ofcom codes: Ofcom is publishing its codes of practice in phases. Monitor these for changes that may affect how platforms treat journalistic content.
5. Engage with press freedom organisations: Bodies such as the NUJ, the Society of Editors, and the News Media Association are actively monitoring the OSA's impact on journalism. Their guidance will evolve as the Act is implemented.

The OSA and Protecting Journalists From Online Abuse

One underreported aspect of the Online Safety Act for journalists is how it can be used to seek protection from online abuse. UK journalists — particularly women, journalists from minority backgrounds, and those covering politically sensitive beats — face disproportionate levels of online harassment and threats. The OSA creates new tools that may assist.

Research by the Reuters Institute has documented the extent of online abuse targeting journalists in the UK, finding that political journalists, sports reporters, and foreign correspondents are among the most targeted groups. The abuse ranges from pile-on harassment (coordinated mass reporting of accounts, sudden floods of critical or abusive replies) to individual direct threats. The OSA's new offences — particularly the threatening communications offence — provide a clearer legal basis for police action than the previous legislative framework, though enforcement remains inconsistent and varies significantly between police forces.

Documenting abuse systematically is the foundation of any effective response. Screenshot and archive every piece of abusive content immediately — many abusers delete posts quickly after sending them. Tools like the Internet Archive's Wayback Machine and the browser extension Save Page Now allow you to create permanent records of online content. When making a complaint to a platform or to the police, comprehensive, dated documentation significantly improves the chances of action being taken.

• New criminal offences: The Act creates a specific offence of sending threatening communications (Section 181) that could be used to prosecute those who make credible threats against journalists online. This offence is broader than the previous Section 127 Communications Act 2003 offence and carries a maximum sentence of five years.
• False communications offence: Coordinated disinformation campaigns targeting journalists (fabricated quotes, falsified screenshots) may constitute the false communications offence under Section 179 where there is an intent to cause harm. This is not straightforward to prosecute, but the existence of the offence may deter some bad actors.
• Platform duty to remove threatening content: Category 1 platforms now have an explicit duty to remove threatening communications. Journalists who report threatening content to platforms should expect more timely removal than under the previous voluntary framework.
• Ofcom complaints mechanism: Journalists who believe a platform has failed to apply its OSA obligations can complain to Ofcom. Ofcom cannot act on individual complaints but uses them to inform its systemic enforcement approach. Documenting a pattern of platform failures across multiple journalists on the same platform can be an effective way to trigger Ofcom scrutiny.
• NUJ support: The NUJ has specialist support for members facing online abuse, including legal advice and a dedicated online abuse reporting tool. Its Be Safe Online resources provide practical guidance on documenting and reporting abuse effectively.

Practical Checklist

Use this checklist to audit your OSA exposure as a journalist or publisher:

• Confirm whether your outlet qualifies as a “recognised news publisher” under the OSA criteria
• Check whether your website's comment section or forum triggers OSA user-to-user obligations
• Set up a system to document and appeal any platform content removals within the expedited window
• Review legacy archive content for potential compliance issues under the new framework
• Monitor Ofcom codes of practice as they are published and note any journalism-specific provisions
• Take legal advice before any undercover reporting that involves online deception
• Ensure your site's data transfer arrangements with US platforms are covered by valid transfer mechanisms
• Subscribe to Ofcom's Online Safety newsletter to receive notification of new codes, consultations, and enforcement decisions as they are published
• Join the News Media Association or Society of Editors industry briefings on OSA developments — both publish member-only summaries of regulatory changes as they occur
• Document any instance where your editorial content appears to have been suppressed or reduced in reach by a platform algorithm — this data is valuable for advocacy and potentially for an Ofcom complaint

Common Mistakes

• Assuming the journalistic exemption is a blanket protection: It requires platforms to consider journalistic purpose before removal — it does not prevent removal, and unrecognised publishers have weaker protections.
• Not appealing content removals: Platforms must provide expedited appeals for journalistic content. Many journalists accept removal without using this mechanism, missing a chance to restore content and build a record.
• Ignoring comment section obligations: Running an open comments section without an OSA risk assessment can expose a publisher to compliance failures, even at relatively small scale.
• Conflating the OSA with GDPR: These are separate regimes with different scopes. OSA governs online service duties; GDPR governs personal data. Both can apply simultaneously.
• Assuming the Act only covers big platforms: Smaller services meeting certain user-number thresholds also carry obligations. Check the Category 2B thresholds if you run a community journalism site.
• Confusing Recognised News Publisher status with complete OSA immunity: RNP status provides enhanced protections and procedural rights, but it does not exempt publisher content from all platform moderation — platforms retain the right to remove content that violates their terms of service, provided the decision is properly documented and appealable.
• Ignoring the interaction between UK OSA and EU DSA: UK publishers distributing content to EU audiences may need to consider both regimes simultaneously. The DSA's transparency report requirements, algorithmic accountability rules, and advertising restrictions may apply to platforms on which you distribute content, even if the OSA is your primary domestic regulatory framework.

Red Flags

• A platform removing journalistic content without providing any removal notice or appeals route
• Content removals that appear to target politically sensitive journalism rather than genuinely illegal content
• Automated moderation systems flagging investigative stories covering extremism or crime as harmful content
• A platform failing to apply the recognised news publisher protections after being notified of publisher status
• Ofcom enforcement decisions that do not give adequate weight to press freedom considerations
• A pattern of coordinated reports against your content by organised groups attempting to misuse the OSA complaints mechanisms to suppress journalism
• Platform age verification systems that prevent your news content from reaching adult readers without justification under the OSA framework
• An Ofcom code of practice that fails to include a specific safe harbour or carve-out for journalism covering the same topics as regulated categories of harmful content — challenge this in the consultation process and through your professional association

Reporting on the Online Safety Act: Covering Ofcom and Platform Regulation

As well as understanding how the OSA affects your own journalism, there is a distinct reporting opportunity in covering the Act itself as a regulatory and policy story. The implementation of the OSA is one of the most significant ongoing regulatory processes in UK technology policy, and dedicated coverage of Ofcom's decisions, platform responses, and the Act's real-world effects is valuable journalism:

• Ofcom's public register: Ofcom maintains a public register of in-scope services. Monitoring additions to this register as Ofcom designates new services is an ongoing story in its own right.
• Transparency reports: Category 1 platforms must publish annual transparency reports under the OSA. These will contain detailed data on content moderation decisions, appeals, and enforcement actions. The first wave of reports will provide genuine insight into how platforms are (or are not) implementing their OSA duties in practice.
• Enforcement decisions: Ofcom's enforcement decisions against platforms will be public documents. Each decision will generate specific lines for platforms compliance stories and may reveal patterns in which types of content are being removed and which are being preserved.
• Parliamentary scrutiny: The Digital, Culture, Media and Sport (DCMS) Select Committee scrutinises Ofcom's online safety work. Its evidence sessions with Ofcom officials and platform executives, and its published reports, are valuable primary sources for journalists covering the OSA beat.

The OSA in Global Context: Comparing UK and International Approaches

The Online Safety Act places the UK in a small group of jurisdictions that have enacted comprehensive online safety legislation. Understanding how the UK approach compares with international equivalents is useful both for journalists covering the OSA as a regulatory story and for journalists whose work may engage multiple regulatory regimes. The EU's Digital Services Act (DSA), which entered full force in February 2024, is the most significant comparator — it applies to the same large platforms as the OSA but uses different mechanisms and enforcement structures. Where the OSA is enforced by Ofcom, the DSA is enforced by the European Commission (for very large platforms) and by national Digital Services Coordinators.

The US First Amendment creates a fundamentally different regulatory environment: Section 230 of the Communications Decency Act provides platforms with broad immunity from liability for user-generated content, and there is no equivalent of the OSA or DSA. This means that US-headquartered platforms (Google, Meta, TikTok, X/Twitter) are primarily regulated for UK users by UK and EU law, not US law — a significant practical point that is relevant when covering platform compliance decisions that have different impacts in different markets. The regulatory arbitrage between jurisdictions — where platforms implement different content moderation standards in different markets — is a significant ongoing journalism story.

• DSA vs OSA comparison: The DSA focuses more heavily on advertising transparency and algorithmic accountability than the OSA; the OSA focuses more heavily on safety by design and harm-specific content duties. The interaction between the two regimes for publishers serving both UK and EU audiences is a developing area of media law.
• Australia's Online Safety Act: Australia enacted its own Online Safety Act in 2021, predating the UK legislation and providing a body of regulatory experience and precedent that has influenced UK implementation. eSafety Commissioner decisions in Australia are sometimes indicative of directions Ofcom may take on comparable issues.
• Section 230 reform debates: US debates about Section 230 reform are directly relevant to UK journalists because any significant change to platform liability under US law would have global implications for how platforms moderate content. Monitoring Congressional hearings and FTC/DOJ actions against platforms provides advance insight into potential changes.

Freedom of Expression Safeguards in the Online Safety Act

Critics of the Online Safety Act from press freedom and civil liberties perspectives have focused on the risks it poses to freedom of expression online. The Act requires platforms to take a broad range of actions that could, if implemented carelessly, result in the removal of legitimate journalistic content. Understanding the specific freedom of expression safeguards built into the Act — and how they can be invoked — is important for journalists whose work may be affected by platform enforcement decisions.

The most significant freedom of expression safeguard is the duty on Category 1 platforms to protect “content of democratic importance”, journalistic content, and content that is simply legal. Platforms must have “particular regard” to freedom of expression when implementing safety measures, and their systems must not be designed in ways that disproportionately restrict legal content. The Joint Committee on Human Rights scrutinised the OSA bill closely for ECHR Article 10 (freedom of expression) compatibility and secured several amendments. However, rights organisations including Index on Censorship and the Open Rights Group have argued that the final Act does not provide adequate protection for edge-case content that is legal but unwelcome to platforms.

• The “legal but harmful” clause removal: The removal of mandatory moderation of legal-but-harmful content for adults was a significant victory for press freedom advocates, as it prevents the OSA from being used to pressure platforms to censor legal commentary and criticism. However, voluntary moderation of such content by platforms remains lawful, and commercial incentives to restrict controversial but legal content remain strong.
• User appeals: The OSA requires Category 1 platforms to provide an accessible, transparent, and effective appeals process for content removal decisions. Journalists whose content is removed should use this process systematically — documenting the removal decision, the grounds given, and the outcome of any appeal. This documentation is valuable both for individual redress and for monitoring platform behaviour across the journalism sector.
• Ofcom's complaints system: Users and publishers who believe a platform is failing its OSA duties can make a complaint to Ofcom. Ofcom has investigatory powers that it can deploy in response to complaints, including requiring platforms to produce information and imposing improvement notices. This is a new accountability mechanism for platform conduct that journalists should be aware of and willing to use when appropriate.

How the OSA Applies to News Publishers: Recognised News Publishers

The Online Safety Act creates a specific category — “Recognised News Publisher” (RNP) — for journalistic organisations that meet certain criteria. Content published by RNPs is treated differently from user-generated content: it is exempt from certain content moderation duties that apply to ordinary user content, and platforms that suppress or remove content from RNPs are subject to additional obligations to notify the publisher and provide an explanation. This is an important protection for journalism, but the criteria for RNP status are specific and not all journalistic outlets will qualify.

To qualify as a Recognised News Publisher under the OSA, an organisation must: operate as a body corporate; have editorial standards that reflect established practice in the industry; be subject to a relevant regulator (IPSO or IMPRESS) or a comparable standards code; be UK-based; and publish news-related material as a significant part of its activities. This means that individual freelance journalists, personal websites, and many hyperlocal operations that are not corporate entities or not regulated by IPSO/IMPRESS may not qualify for RNP status — and therefore do not receive the enhanced protections. For organisations that are eligible, applying for and maintaining RNP status is a practical priority.

• IPSO membership for RNP status: IPSO membership is the simplest route to RNP qualification for most publishers. IPSO regulates publications of all sizes, including small regional and hyperlocal titles. Annual fees are tiered by circulation and revenue, and are typically very low for small publishers.
• IMPRESS as an alternative: IMPRESS is the only press regulator in the UK recognised under the Royal Charter on press self-regulation, which makes it the preferred regulator under the terms of the Leveson settlement. IMPRESS membership also qualifies a publisher for RNP status under the OSA.
• Practical implications of RNP status: If a major platform removes or suppresses content from your news publication, as an RNP you are entitled to a specific explanation and a right of appeal. This is a meaningful additional protection, but you need to be aware of it and be prepared to invoke it if your content is affected.
• Monitoring platform enforcement: As the OSA's systems take effect, tracking which types of journalistic content are being removed or suppressed by platforms — and comparing treatment of RNP content against non-RNP content — will be an important monitoring function for press freedom organisations and media reporters alike.

Misinformation, Disinformation, and the OSA: What Journalists Need to Know

The Online Safety Act does not contain a general prohibition on online misinformation or disinformation — a deliberate decision by Parliament that reflected concerns about restricting legitimate expression. The government had initially proposed a “legal but harmful” clause for adults (Category 2A) that would have required platforms to moderate harmful-but-legal content including health misinformation, but this was removed during the Bill's passage after lobbying from free speech advocates and publishers concerned about scope creep.

What the OSA does require is that platforms provide users with tools to filter or reduce exposure to content they do not want to see, including content that they personally regard as misinformation. This “user empowerment” duty shifts some responsibility from platforms to users, a design philosophy that has been widely criticised by public health researchers and digital rights organisations alike. For journalists covering the OSA as a policy story, the exclusion of misinformation from mandatory content moderation obligations — and the political controversy around that exclusion — is a significant and continuing story angle.

• The Future Internet Act: The OSA is likely to be revised or supplemented as AI-generated content and synthetic media become more prevalent. Track proposals from Ofcom, DSIT (Department for Science, Innovation and Technology), and parliamentary committees for planned amendments.
• AI-generated content labelling: Separate to the OSA, the government has been consulting on mandatory labelling requirements for AI-generated content. Journalists covering elections and political campaigns should monitor this area closely, as synthetic media that mimics real politicians is a growing editorial challenge.
• Counter-disinformation unit: DSIT runs a Counter Disinformation Unit that monitors online narratives and shares intelligence with platforms. It does not remove content directly. FOI requests to the CDU can reveal which narratives are being monitored and what interventions (if any) have been taken.

Child Safety Provisions: A Journalist's Reporting Guide

The child safety provisions in the Online Safety Act are among the most expansive in the legislation and are generating significant ongoing controversy and reporting opportunities. Category 1 platforms must not only address clearly harmful content affecting children (child sexual abuse material, content promoting suicide or eating disorders to minors) but also take steps to prevent children accessing age-restricted content. The mechanisms for enforcing this — age verification — are themselves highly contested:

Age verification for adult content sites is one of the most significant and contested implementation challenges. Ofcom's guidance on what constitutes robust age verification has been closely watched by civil liberties organisations concerned about data privacy. Journalists covering this area should familiarise themselves with the work of the Open Rights Group, the Age Verification Providers Association, and Ofcom's published guidance on “highly effective age assurance”. The tension between child safety (requiring effective age checks) and adult privacy (requiring minimal data collection) is a continuing policy story.

• Children's User Advocacy: Platforms must appoint a Children's Advocate (or equivalent) under Category 1 duties. Monitoring who is appointed and how effective they are — by speaking to child safety charities — is one ongoing story angle.
• The NSPCC and Internet Watch Foundation: Both organisations are active commentators on the OSA child safety provisions, with strong research bases. They are valuable sources for child safety dimensions of OSA reporting.
• Research exemptions for academic journalists: The OSA includes provisions to ensure that journalism and academic research accessing otherwise restricted content for legitimate public interest purposes is protected. Understanding the scope and limits of this exemption is important if you investigate platforms or extremist content.

Primary Sources

• Online Safety Act 2023 — Full text on legislation.gov.uk
• Ofcom Online Safety guidance — Codes of practice and regulatory updates
• NUJ: Online Safety Act and Journalism — Union guidance for members
• News Media Association: Online Safety — Publisher-focused briefings
• DCMS Select Committee — Parliamentary scrutiny of Ofcom and online safety
• Social Media Guide for UK Journalists — Best practices for reporting on and via social platforms
• IPSO Editors' Code 2025 — Understanding the regulatory framework that determines recognised publisher status