UK public sector must leverage AI for digital ID – Tony Blair Institute

The integration of artificial intelligence (AI) into digital identity systems holds immense potential for the public sector, according to a new report by the Tony Blair Institute (TBI).

The report maintains that AI can help transform public-service delivery, enhance efficiency, improve cybersecurity and promote inclusivity.

Citing the success of digital identity implementations in Singapore, Estonia and Finland, it said: “Governments must leverage AI’s capabilities and implement robust security measures and regulatory frameworks to ensure that digital-ID systems are secure, trustworthy and beneficial for all citizens.

“By addressing the challenges of deepfakes, remote transaction manipulation and data protection head on, nations can harness digital-ID innovation and AI capabilities responsibly and unlock extensive benefits: personalised education, improved health care, seamless access to services and data-driven policymaking.”

TBI analysis shows that the UK could implement a digital-ID system within three years and generate cumulative net savings of almost £4 billion during this Parliament and nearly £10 billion during the next term.

How AI can support digital ID outcomes

TBI said AI can support and augment digital ID outcomes in several key areas:

Enhanced user experience, support and accessibility: AI chatbots and virtual assistants can provide real-time support for citizens using digital ID systems. For example, the mAigov tool, an AI-driven chatbot launched by Greece’s Ministry of Digital Governance, is embedded within the country’s digital-government portal (gov.gr). It enables citizens to access 1,832 digital public services and receive personalised responses to their questions.

Personalised and targeted services: By analysing data linked to digital ID, AI can enable governments to deliver more targeted social services, healthcare interventions and educational resources. It said Singapore is working to integrate AI for disease diagnosis, patient monitoring and predictive healthcare analytics. In the education sector, AI-powered digital tutors such as OpenAI’s GPT-4o analyse students’ progress and needs, providing personalised and adaptive materials, explanations and support.

Data-informed decision-making: AI, when trained on trusted data, can analyse new information and identify patterns, enabling smarter decision-making. In healthcare, AI can flag scans requiring specialist attention, reducing unnecessary diagnostic tests. It can also enhance pharmaceutical trials, though this requires robust security measures for accessing patients’ data and managing consent. It said the NHS App in the UK exemplifies how access to health data and appointment booking can be securely managed, with over 30 million sign-ups. Looking ahead, integrating health and medication information into digital wallets with selective disclosure features could give patients greater control over their records, further revolutionising access to and management of healthcare data.

Automated identity verification and fraud detection: AI algorithms can automate the identity-verification process, detect fraudulent activities through pattern recognition and enhance security. For example, the UK Home Office uses biometric technology to automate applications to the EU Settlement Scheme, allowing secure remote identity verification without any in-person appointments. Governments in particular require the highest levels of security afforded by AI. The European Union Agency for Cybersecurity’s (ENISA) Threat Landscape 2023 report identified public administration as the “most targeted sector”, drawing 19 percent of targeted cyberattacks.

“The stakes are high but so are the potential rewards,” concluded the report. “As stewards of this transformative change, governments must seize the opportunity to leverage AI’s potential while safeguarding against its risks, paving the way for a digitally empowered society that leaves no one behind.”