EU and UK: Technology Law Shorts

Commercial technology on the horizon: what do I need to know?

EU: EC launches digital connectivity package and consultation

On 21 February, the European Commission (EC) launched a package to improve Europe’s digital infrastructure. The package aims to boost Europe’s digital strength and ensure reliable connectivity for the future. It includes a:

1. White Paper with twelve scenarios for meeting Europe’s digital targets for 2030 and encouraging innovation and investment to achieve a Digital Single Market. It suggests, among other things, creating a Connected Collaborative Computing Network to develop integrated connectivity; and
2. Recommendation setting out national and EU-level actions to improve submarine cable security and resilience via better governance and funding. It suggests that Member States, among other things, create a Cable Security Toolbox and simplify permit processes.

Impact: Share your views about the White Paper’s twelve scenarios by responding to the EC’s consultation. It is open to Member States, civil society, industry, and academics until 30 June 2024. The outcomes will shape future policy actions, including laying the groundwork for a Digital Networks Act.

EU: New AI Office

On 24 January, the European Commission (EC) announced its decision to establish the European Artificial Intelligence Office (AI Office). The decision entered into force on 21 February.

The AI Office will:

• play a key role in implementing the AI Act by supporting Member States in making sure that they consistently apply the AI Act by setting up EU-level advisory bodies;
• enforce rules for general-purpose AI models;
• use its powers granted under the AI Act to request information and measures from providers;
• monitor compliance, investigate potential rule violations and apply sanctions;
• support the development and use of trustworthy AI and protect against AI risks;
• work with expert groups and AI developers to combine knowledge and create codes of practice setting out rules for AI implementation; and
• oversee the AI Pact, which allows businesses to collaborate with the EC and other stakeholders.

While the AI Office will be part of the EC, it will operate independently with a separate budget. It will not affect the powers of national and EU bodies in AI supervision.

Impact: External experts and stakeholders will have the chance to support the work of the AI Office through a call for expression of interest. Sign up to receive updates from the AI Office.

EU: AI Act update

On 13 February, EU Parliament committees adopted the final text of the EU’s AI Act (Act). The Act aims to protect fundamental rights, democracy and the environment from high-risk AI while promoting innovation in Europe. Obligations vary depending on AI’s potential risks and level of impact.

Some key points of the agreement are:

• Further detail added to the list of banned AI practices, such as using real time biometric identification in public places and emotion recognition in a work or education setting;
• Certain AI systems classed as high-risk will be subject to strict rules (like having transparent, detailed information and having in place technical documentation and robust record-keeping processes), while others classed as low risk will be subject to lighter rules; and
• General-purpose AI (GPAI) systems providers will be subject to additional rules like making sure that their outputs are marked as AI-generated, carrying out risk assessments related to possible systemic risks and maintaining robust cybersecurity.

Impact: The Act will affect many groups involved in the AI process within and outside the EU. The European Parliament will now vote on the text (scheduled for mid-April). If it is approved, it will return to the Council of the EU for final adoption. Once adopted, a two-year implementation period will apply, but some provisions will apply before then (bans on certain AI systems will apply after 6 months and GPAI rules will apply after 12 months).

UK: Online Safety Act update

On 10 January, the Online Safety Act 2023 (Commencement No. 2) Regulations 2023 brought into force the majority of the Online Safety Act 2023 (OSA). The OSA requires online platforms to limit harmful content. Non-compliant providers will answer to Ofcom, now also the UK’s online safety regulator.

On 31 January, Part 10 of the OSA came into force, creating new offences, including sending false and threatening communications; sharing (or threatening to share) intimate photos / films; and encouraging or assisting serious self-harm.

Refer to Ofcom’s summary of the new rules issued on 27 February.

Impact: Full implementation of the regulations is not expected until 2026. Additional laws, statutory guidance and codes of practice are needed before most of the OSA’s obligations become operational. Based on Ofcom’s roadmap, Ofcom will next issue a call for evidence related to its approach on transparency, user empowerment and other categorised services duties (Phase 3) in early 2024, followed by a consultation on draft transparency guidance in mid-2024.

UK: Updated policy notes on the Automated Vehicles Bill

The Automated Vehicles Bill (Bill) aims to introduce safety laws for self-driving vehicles. On 16 February, the government updated its policy notes setting out how it plans to exercise its power in the Bill and potential policy approaches.

One of the key policies set out in the notes relates to setting safety expectations for self-driving vehicles, guided by a set of National Safety Principles (referred to as a ‘Statement of Safety Principles’ (SSP) in the Bill). Additional rules may be set through other regulations. The SSP informs safety assessments within the self-driving regulatory framework. The goal is to make sure that self-driving vehicles reach a safety level equal to or surpassing that of a cautious and capable human driver. This minimum road safety standard guides the SSP.

The Secretary of State for Transport will publish the SSP as statutory guidance and consult with relevant interest groups and the public during its preparation.

Impact: On 5 March, the Bill had its second reading in the House of Commons. Expect further developments with more planned drafting and consultations this year – the policy notes are a live document, expected to be updated during the Bill’s parliamentary passage.

Edited by Kirath Bharya (Regulatory Insights).

Further reading:

• Flash Update – Global AI treaty negotiations
• Patenting AI innovations in the UK
• Global AI Regulatory update - February 2024
• Artificial Intelligence: UK and EU regulatory round-up
• Legal Telescope AI special edition : A view from our technology lawyers - 2023