GenAI in Government: Navigating Risks and Building Trust in Citizen Services
Mokshita P.
Artificial Intelligence

GenAI in Government: Navigating Risks and Building Trust in Citizen Services

Gartner's survey indicates that less than 25 percent of government organisations will have citizen-facing generative AI services by 2027, citing community trust concerns and fear of failure as adoption barriers.

In a report by Gartner, it has been predicted that less than 25 percent of government organisations will incorporate GenAI-enabled citizen-facing services by the year 2027. According to the annual global survey conducted by Gartner, which involved over 2,400 CIOs and technology executives, 25 percent of governments have either deployed or have plans to deploy GenAI in the next 12 months. An additional 25 percent are planning to deploy within the next 24 months.

The survey highlights that, despite the increasing interest in GenAI over the past 15 months, concerns regarding public trust and the fear of failure are major factors slowing down adoption for external use with citizens. Dean Lacheca, VP Analyst at Gartner, noted, “While governments have been benefiting from the use of more mature AI technologies for years, risk and uncertainty are slowing GenAI’s adoption at scale, especially the lack of traditional controls to mitigate drift and hallucinations.”

Governments have initially focused on establishing governance frameworks to support experimentation and limited adoption. However, the lack of empathy in service delivery and failure to meet community expectations are cited as potential obstacles to public acceptance of GenAI in citizen-facing services.

To overcome these challenges, Gartner recommends that governments continue to actively deploy GenAI solutions that enhance internal aspects of citizen services. The report advises aligning the adoption of GenAI with the government's risk appetite to prevent early missteps that could undermine community acceptance.

Governments are encouraged to prioritise back-office opportunities, addressing use cases that predominantly impact internal resources and avoiding perceived risks associated with citizen-facing services. Building knowledge and skills associated with GenAI and establishing transparent AI governance and assurance frameworks are also essential steps for accelerating adoption.

Lacheca stated the need for empathy-focused practices in human-centred design when implementing citizen or workforce-facing AI solutions. This ensures that GenAI solutions align with community expectations and adhere to guidelines for responsible use, addressing concerns such as inaccurate results, data privacy, and secure conversations.

In conclusion, Gartner's recommendations aim for a gradual and well-managed adoption of GenAI by government organisations, taking into account both internal and external considerations to build trust and ensure successful integration into citizen services.