Ofcom has set out its approach to enabling safe and secure AI adoption across the sectors it regulates and within its own work. The regulator’s report explains how it is applied broadly in a technology-neutral way where outcomes-based regulatory principles to AI are highlighted, and takes a case study approach to demonstrate its role in supporting the sectors it oversees to safely innovate with AI.

“Our approach to AI aligns with our broader mission of making communications work for everyone,” the British communications regulator said yesterday in a statement.

The industries Ofcom regulates have technology and innovation at their heart. As technologies evolve, new opportunities emerge that have the potential to drive better outcomes for consumers and businesses.

The regulator said it was taking proactive steps to help realise these benefits by enabling safe and secure AI adoption across the communications sector.

The Ofcom publication includes eight case studies of its upcoming work and work from the last 12 months detailing how it has supported sectors and helped drive growth.

These steps include the following:

• Building a pilot data lake to make spectrum licensing and online safety data more accessible and usable helping create the right foundation for AI adoption in our sectors.
• Engaging with innovators to gain a clearer view of where regulatory uncertainty might hinder AI adoption, so we can respond accordingly.
• Helping organisations safeguard against deepfakes by setting out technology-led mitigation techniques.
• Examining trust in AI chatbots to understand how people perceive and engage with AI, and where harms may develop as a result.
• Assessing the impact of AI on telecoms customer experience to support safer AI adoption through assessing whether current consumer protections are sufficient.
• Exploring AI deployment within the broadcasting sector to better understand the opportunities and risks of AI in content creation.
• Understanding AI use in the cybersecurity of telecommunications networks and information systems and whether current regulatory requirements may present unintended barriers to adoption.
• Exploring how AI is being applied, or could be applied, by telecoms providers to support network management and optimisation, to help inform our policy development on network resilience for telecoms providers.

“While both industry and consumers benefit from AI deployment, the risks created or exacerbated by AI primarily flow to the consumer.  At Ofcom, we’re proactively monitoring and evaluating emerging harms. Where intervention is required, we’re taking swift and decisive action, “ the regulator stressed.

As an example, it cited the coordination with AISI and the National Cyber Security Centre (NCSC) to update stakeholders on frontier AI’s cybersecurity implications after Anthropic’s preview of Claude Mythos caused widespread concern, and launched a formal investigation into X’s Grok chatbot.

It said it was working closely with the UK Government to prepare for increased regulatory responsibilities relevant to AI, including any introduced by both new legislation and changes to existing legislation.

What does this approach to AI mean in practice? Understanding the impact of new technologies such as AI is a crucial part of helping ensure communications works for everyone.  “We are investing heavily in building our understanding of the day-to-day impact of AI upon citizens and consumers, including the impact of agentic AI as we prepare for our sectors to use this technology more widely,” Ofcom said.

How is Ofcom using AI? Alongside supporting businesses and consumers, it has been investing in internal AI capabilities and piloting responsible AI adoption across many elements of its work.

It said it was developing its own AI tools and algorithms to help streamline the consultation process and to support work in tracking developments in technical standards fora.