Trying to track the staggered implementation of the UK Online Safety Act 2023 (“OSA”)? You are in safe hands here. See the latest updates and key dates below.

25 July 2025Online age checks must be in force from 25 July 2025

With the deadline for completing the first children’s risk assessments having passed on the 24 July 2025, the duties to protect children from harms online are now in force. From 25 July 2025, tech firms must implement robust age checks to prevent children from accessing online pornography and other harmful content, such as material related to self-harm, suicide, and eating disorders. Major platforms—including Pornhub, Bluesky, Discord, Grindr, Reddit, and X—have committed to deploying age checks across their services. Ofcom will enforce compliance, extending its oversight to all sites that allow user-shared pornographic content. The regulator will also monitor the largest platforms used by children, requiring them to assess and report on their child safety measures, particularly around content feeds and algorithms. Most UK parents support these new protections, though some doubt tech firms will fully comply. Ofcom warns that non-compliance will result in enforcement action.

21 July 2025Ofcom publishes its final transparency statement and transparency reporting guidance

Ofcom published its final transparency statement and guidance on transparency reporting. This guidance clarifies when and how Ofcom will use its transparency powers, outlining the process for transparency reporting under the Online Safety Act. The guidance details the requirements for providers of "categorised services"—such as social media, search, and pornography platforms listed on Ofcom’s public register—to publish transparency reports. These reports must include information specified by Ofcom in transparency notices, and the guidance explains the factors Ofcom will consider in determining what information must be disclosed. It also sets out how Ofcom will engage with stakeholders during the reporting process. Ofcom will also publish its own transparency report, which will analyse and draw conclusions from the reports submitted by service providers.

18 July 2025Ofcom publishes second consultation on online safety fees and penalties

Ofcom launched its second consultation on implementing the online safety fees and penalties regime. It provides draft guidance to help service providers calculate their Qualifying Worldwide Revenue (QWR), which determines whether they are required to pay fees under the Online Safety Act. The Act mandates that Ofcom’s costs for regulating online safety are recovered from certain providers of regulated services, specifically those whose QWR exceeds a threshold set by the Secretary of State. The recently laid regulations define QWR as the total revenue generated worldwide from relevant parts of a provider’s regulated services and require providers to use a ‘just and reasonable approach’ when apportioning revenue between regulated and other activities. The consultation remains open until 10 September 2025.

11 July 2025Ofcom publishes paper exploring how different tools can be used to identify deepfakes

Ofcom published a discussion paper examining tools to identify deepfakes—AI-generated fake videos, images, and audio—which threaten online safety. The paper evaluates four main ‘attribution measures’: watermarking, provenance metadata, AI labels, and context annotations. These tools aim to help users recognise AI-generated content and understand how it was created. The paper highlights that attribution measures can help users think more critically about content but should not place all responsibility on individuals. It stresses the need for clear communication, standardisation, and combining attribution with other interventions, as these measures can be manipulated or removed. The findings are intended to guide industry and policy, rather than set new rules, and will inform Ofcom’s future work under the Online Safety Act.

11 July 2025MPs demand stronger regulation of misinformation and AI content

A report from the Science, Innovation and Technology Committee (SITC) criticised the Online Safety Act (OSA) for failing to address the algorithmic spread of misinformation and the risks posed by generative AI. The committee highlights that social media companies, motivated by profit, have amplified false and harmful content. MPs warn that the OSA does not cover ‘legal but harmful’ content or provide sufficient oversight of recommendation algorithms. They call for a new regulatory regime based on public safety, free expression, platform accountability, user control, and transparency. Key recommendations include requiring platforms to deprioritise fact-checked misinformation, increasing algorithmic transparency, and introducing legislation to regulate generative AI. The report also identifies a gap in regulating the digital advertising market, urging stronger action against the monetisation of harmful content. The committee stresses the need to protect free speech while imposing proportionate restrictions on the spread of misinformation and calls for urgent government action to strengthen online safety.

9 July 2025Enforcing the Online Safety Act: Children’s risk assessments

Ofcom announced that several online services are now required to submit their written records of children’s risk assessments by 7 August 2025 or face potential enforcement action. This requirement forms part of Ofcom’s ongoing enforcement programme under the Online Safety Act, specifically targeting the protection of children online. On 24 April, Ofcom published its children’s safety codes of practice and accompanying guidance. Providers need to complete a suitable and sufficient children’s risk assessment in line with Ofcom’s requirements by 24 July 2025. Providers must not only carry out these risk assessments, but also make and keep a detailed written record, including how the assessment was conducted, its findings, and how it will be kept under review. Ofcom has now issued formal statutory information requests to a number of service providers.

30 June 2025Ofcom launches consultation on additional safety measures

Ofcom opened a new consultation seeking views on a further set of targeted safety proposals under the Online Safety Act. These latest proposals aim to limit the spread of illegal content through improvements to recommendation features and crisis response protocols, expand the use of proactive technologies, such as hash-matching and fraud detection, and enhance protections for children via stronger age assurance and livestream interaction controls. The consultation remains open until 20 October 2025.

Other Topics