After extensive research, TNO reports that government agencies are increasingly using artificial intelligence (AI). An exploratory survey found that the number of AI applications found within government is more than 1.5 times higher compared to 2021 (165 cases). Municipalities in particular are large users: out of a total of 266 AI applications found, 105 (39%) were used here. In addition, there is also increased transparency regarding the use of AI, making it easier to see exactly what technologies the government is using and for what purposes.
Why this is important:
AI has the potential to improve society, and it also benefits governments.
Types of AI Applications and Government Agencies
The majority of applications (105) were found in municipalities, and many applications were developed in partnerships (46), followed by institutions such as the Rijkswaterstaat, the DUP and the RVO (23).
For the first time, there are more implementations (102) than experimental applications (95), indicating that AI applications are becoming more mature.
Knowledge processing, archiving, and anonymization
AI is most often used for knowledge processing, archiving and anonymization (60), followed by inspection and enforcement (54). The increase in anonymization software can be explained by the influence of the Open Government Act (Woo).
Several examples of successful applications in the public domain are given: anonymization tools to protect sensitive data, algorithms to detect ship movements in ports using camera images, algorithms for fisheries and road inspections, and virtual policy assistants to answer parliamentary questions quickly and appropriately.
The most used technology is image recognition (80), followed by speech and text recognition (70). Interestingly, only eight applications use generative AI and LLMs (large-scale language models). This is because generative AI can be considered an online service and is primarily used by individuals within an organization.
Apple postpones launch of AI products in Europe, but here's what's coming next
Europe will likely have to wait even longer for Apple's “smart Siri” and other new AI products.
transparency
By 2024, there will be more transparency regarding the application of AI in public services. For example, more case studies will tell us which AI techniques were applied. This year, there were only 38 cases (14%) where it was unclear what type of AI techniques were applied, compared to 72 cases (44%) in 2021.
