AI news ·
State AI Pilots Show Success Through Learning From Failure and Data Improvements
AI pilots in state governments often face setbacks, but learning from failure and strong, well-labeled data drives progress. States like Pennsylvania show success through training and adaptation.
GovTech Biz AI Report From NASCIO Conference Highlights Benefits of Failure and Strong Data
Artificial intelligence (AI) pilots in state governments don’t always succeed on the first try. Yet, a recent report discussed at the National Association of State Chief Information Officers (NASCIO) Midyear Conference in Philadelphia shows that failure can lead to valuable lessons and progress. The report emphasizes the importance of structured, well-labeled data and organizational readiness in driving successful AI initiatives.
According to the AI x Talent report from Govern For America, a nonpartisan nonprofit focused on public service, many states have seen promising results from AI pilot projects. For example, the use of tools like ChatGPT has helped officials save time and reduce technical debt in public agencies. However, state AI leaders acknowledge that initial failures are part of the learning and experimentation process, especially when iteratively training AI models.
Key Findings From the Report
- Successful AI projects often rely on well-structured and labeled data systems.
- Learning from failed pilots is critical to refining AI applications.
- More than a dozen states contributed insights, reflecting growing AI interest nationwide.
- At least 32 states had AI task forces, but fewer have permanent AI leadership roles or dedicated centers of excellence.
- Scaling AI, especially large language models, demands significant advances in state capacity.
The report also notes that the recent surge in AI interest has motivated states to invest in foundational data science capabilities, a necessary step before fully deploying AI solutions at scale.
Real-World Example From Pennsylvania
At the NASCIO conference, Pennsylvania’s director of emerging technologies, Harrison MacRae, shared insights from a yearlong pilot involving ChatGPT Enterprise. The study included 175 workers across 14 agencies, with over 85% reporting positive experiences. MacRae pointed out that a key part of success came from learning how to interact effectively with the new tool, highlighting the importance of training and adaptation.
This example underscores how states can gradually build AI competence by embracing experimentation and user training.
Looking Ahead
As states continue to explore AI, the report encourages a balanced approach that expects some setbacks while focusing on improving data infrastructure and organizational support. Developing AI leadership and permanent centers of excellence can accelerate progress.
For government IT professionals and leaders interested in expanding their AI skills and knowledge, exploring targeted courses and training can be a practical next step. Resources like Complete AI Training’s latest AI courses offer focused learning paths on AI tools, prompt engineering, and public sector applications.