Why government is key to unlocking AI’s true potential

With artificial intelligence (AI) conquering domains previously thought to be exclusive to the human mind, such as visual arts and language, it may seem like the promise of machines solving our biggest challenges is just around the corner.

Unfortunately, AI alone is not going to improve our lives.

It’s certainly impressive that ChatGPT, the most advanced chatbot yet, can create mystery novels, write code, and pass MBA exams, while Dall-E can draw anything you want. But these are parlor tricks compared to the hard work of tackling our thorny issues like poverty and inequality.

The true potential of AI has remained elusive as those charged with tackling these issues – governments – have long struggled to harness technology.

Governments have the key building blocks to unleash the potential of AI – data – but first governments need to learn how to walk with technology before they run.

I led two state labor departments, one during the COVID-19 pandemic, so I saw firsthand how some of our most critical services, such as public labor and unemployment programs, remain rooted in Great Depression-era practices and driven by yesterday’s technology. But I also saw how technology can and cannot help.

So, why is government so rarely successful in leveraging technology and what can be done?

The government needs to change its mindset. Technology is not a standalone solution. It is deeply integrated with processes. So technological solutions must be accompanied by changes in business operations or they will fail.

Our unemployment insurance systems are a case study. At the outbreak of the pandemic, millions of workers were suddenly without a job and applied for unemployment insurance in record numbers. States did their best to process claims quickly, but long delays were the norm.

Unemployment insurance problems weren’t just due to old technology, so the solution isn’t just new technology. Proof of this is that no state unemployment insurance program performed well during the pandemic – not even those with “modernized” systems that spent millions of dollars.

The unemployment insurance systems I know are powered by refrigerator-sized mainframes from the 1980s that use a programming language that few know. Nevertheless, during the pandemic, my team was the first to accept Pandemic Unemployment Assistance claims and the second fastest to pay them.

We were successful because we focused on applications of technology that would complement the process reforms we were implementing at the same time. The process needs better technology, but the process also needs to change to take advantage of better technology.

Our country has managed unemployment insurance in the same way since it was established in the 1930s: applicants provide information about their employment situation; employees manually review each claim, while also manning the call center and answering questions.

This process is not scalable. To stay afloat during the pandemic, we have enabled every operational solution to free up staff and streamline the process. Our motto was: “Let technology handle simple things, so that people can help people.”

Modern technology is necessary, but not sufficient. The latest technology in the world is only as good as the underlying bureaucracy.

Another barrier to government innovation is a procurement process that aims to easily buy ready-made technical solutions. Unfortunately, there are no ready-made solutions to do the hard work of actually improving service.

This backward process means that the government buys a solution, but does not work on the problem. That is why, after spending millions of taxpayer dollars, technology rarely drives public outcomes.

Successful projects in both the public and private sectors use an iterative, resident-first approach. That means developers work closely with residents and their state partners to understand problems, test solutions, and further refine them.

The good news is that this process is getting easier with modern technology, as it also frees up the data that fuels a virtuous cycle of continuous improvement that transforms service – and this is key to unlocking the true power of AI.

For example, states have decades of quarterly wages. By using machine learning and AI to mine this wealth of data, we can transform workforce training programs, which is what I’ve encouraged states to do as CEO of RIPL, a not-for-profit civic-tech organization that works with governments to make better use of data.

In Maryland and Hawaii, they now have a digital tool that analyzes an unemployed worker’s resume and makes data-based recommendations for training opportunities, job opportunities and new careers that are proven to lead to higher wages. This is made possible by machine learning that works across tens of millions of state wage records to understand what careers similar job seekers have moved into, held on to, and earned more.

Massive amounts of data enable machine learning to do such impressive things. And there is no shortage of data that only governments can access and use to improve lives. All it takes is the hard work it takes to approach technology differently.

Shot Jensen is the former director of Labor in Rhode Island and the current CEO of RIPL, a nonprofit social impact technology organization that works with governments to use data, science and technology to improve policies and lives.