NYTimes: Governments can add AI to services without replacing staff

In short: A new NYTimes opinion piece argues that government can use AI in small, practical ways that improve existing services, instead of trying to replace them.

What's going on

The article’s main point is that AI does not have to remake how people deal with government. Instead of building entirely new systems, agencies can add AI to the tools people already use, like familiar websites, phone menus, and internal staff workflows.

A National Academy of Public Administration roundtable has described this as an incremental approach. For example, an AI chatbot can answer common questions and help route people to the right service, including after business hours. Think of it like adding a helpful front desk assistant to a busy office, not replacing the staff who handle complicated cases.

The World Bank has also outlined common public sector uses for AI, such as helping deliver services, spotting fraud, and supporting policy analysis. At the same time, it notes that benefits are not guaranteed and that real-world rollouts can bring new problems.

What to watch

A “not disruptive” approach still has risks. Public agencies need to check whether the data used by AI is accurate, how personal information is protected, and whether decisions can be explained in plain language. They also need clear ways for people to challenge mistakes, especially when AI influences a decision that affects benefits, taxes, housing, or legal outcomes.

Source: NYTimes