Regulatory Watch – September 2026

EU proposes a Kids Act

The European Commission has announced plans for a new Act designed to enhance the protection of children online. The Kids Act will bar under-13s from accessing social media while 13-15 year-olds will only have access through a mini account managed by a parent or guardian. Services used by minors will have to offer stronger protection, including prohibiting unsolicited contact from strangers, limiting addictive features such as infinite scroll and offering easy ways to block or mute other users. Platforms will also have to check users’ ages while protecting their privacy. The Commission states that the Act will ‘shift the burden from regulators to platforms’, with companies required to demonstrate that their services are safe. The text will now be considered by the European Parliament and the Council. (Click here to read more).

13 EU countries call for regulatory ‘deep clean’

Austria is leading a new alliance on regulatory simplification, arguing for a review of the impact of new rules on businesses and a ‘deep cleaning’ of existing rules, according to Euronews. Thirteen member states – Austria, Czechia, Denmark, Germany, Estonia, Hungary, Italy, Latvia, Lithuania, Poland, Portugal, Slovenia and Slovakia – signed a non-paper (an informal diplomatic document) saying that ‘EU legislation cannot be an end in itself. Therefore, regulatory restraint to avoid over-regulation and the introduction of new administrative burdens is a condition for sound EU law-making’. Euronews reports that a new ‘Alliance for the Reduction of the Bureaucratic Burden’ will be announced shortly. (Euronews)

Meta launches new legal challenge against Ofcom

UK media regulator Ofcom faces a fresh legal challenge from Meta over the categorisation of the company’s services. Both WhatsApp and Instagram are classified as ‘Category 1’, which means they face additional obligations under the Online Safety Act, including greater transparency, more user controls and additional reporting of illegal content. Meta launched a legal challenge in May over the methodology for the calculation of charges which, the company says, should not be based on global revenue. TikTok and X are also taking legal action against the regulator over the provision of data required to monitor their services. Ofcom Director for Online Safety, Oliver Griffiths, said in evidence to a parliamentary committee that in a litigious environment, ‘we are absolutely looking to drive change’ but that ‘if we take unnecessary risks we are going to get absolutely clobbered by deep-pocketed individuals and firms’. (The Guardian)

AI rivals join proposal for ‘AI slowdown’

Sam Altman and Elon Musk have rallied behind a call from Anthropic’s Dario Amodei for a slowdown in the development of AI, saying that the technology is advancing ‘faster than our ability to keep it safe’. The development comes after a week in which Anthropic researcher Jacob Coxon resigned with the claim that ‘the people building AI earnestly believe that it could kill us all by the end of the decade’. Amodei responded in an essay that ‘a co-ordinated pacing strategy’ would allow for the testing of frontier systems to catch up, proposing that independent researchers should be given have ongoing access to every frontier AI company. He also called for government support in allowing AI developers in democratic countries to co-ordinate safety standards and limits, a step which might otherwise fall foul of antitrust laws. China has rejected the idea of international AI governance, with a spokesman criticising ‘threat narratives’ and warning against ‘engaging in confrontation and malicious competition’. (Financial Times / BBC)

Google warns of ‘reduced quality’ as it makes changes to comply with the DMA

Google has rolled out changes to its online search engine results designed to comply with the EU’s Digital Markets Act (DMA). The EU regards the changes as necessary to ‘level the playing field’ for companies appearing in Google searches. Google was fined 460 million euros in July for favouring its own services in shopping, hotels, transport and sports results. The new results page will feature one specialised search engine followed by two others but with fewer details. Google claim the changes will ‘degrade the user experience for Europeans’ and remove helpful features while ‘boosting online intermediaries at the expense of local businesses’. The changes will not affect users outside the EU. (Reuters)

Meta agrees to $18 billion settlement for online harms to children

Meta has reached a settlement with US states and territories to resolve claims that Facebook and Instagram harmed children. The money will be paid to 48 states plus the District of Columbia and three US territories. Meta has denied wrongdoing and says the payment will be distributed in annual instalments over a 10 year period. Meta will also implement a series of changes including daily time limits and night-time blocks. While the settlement covers almost all of the US, New Mexico has separately imposed a fine on Meta of nearly $1 billion for being a ‘public nuisance’. (Reuters)

UK government rejects idea for an AI ‘kill switch’

A proposal by a cross-party grouping of UK MPs to introduce a law allowing the government to ‘switch off’ models in UK data centres has been rejected. Fears have risen after incidents in which AI models have exhibited ‘rogue behaviour’. But researchers note that switching off an AI model will provide little protection without international agreement and the government says that the UK ‘cannot simply turn AI off’. (BBC)