How to speed up your work in government (using AI)

AI's true power lies in its capacity to streamline tasks, boost productivity and elevate the efficiency of daily operations. When we discuss AI with public servants, we often hear: "How can AI help me accomplish mundane tasks faster, save time and improve productivity?"

The appeal of AI is its ability to manage routine tasks efficiently and offer insights that would otherwise demand significant time and effort. Whether you're new to AI or have experience, learning how to leverage these tools effectively can greatly enhance your work.

Earlier this month, Apolitical hosted an online event focused on using AI to accelerate work in the public sector. Chantal Smith and Duncan Brown shared practical tips, innovative ideas and personal experiences on integrating AI into your workflow. We turned what they said into a 1-page summary you can use to guide you through the early stages of experimenting with AI. They also provided guidance on addressing common challenges like data privacy concerns.

Continue reading for actionable advice on harnessing AI to improve your productivity and work efficiency. And, if you prefer to learn through a short and snappy video, check out the highlight reel from the event at the end of the article.

Key considerations before you start using AI

Before diving into the world of AI, it’s crucial to understand some foundational principles:

Continuous improvement: AI is not static. Models are constantly updated and refined, requiring ongoing learning and adaptation.
Critical use: AI’s utility is not universal; it excels in some areas and falters in others. Understanding these boundaries — and recognising that they are shifting — will help you use AI more effectively.
Compliance and ethics: Always align your AI use with your department’s guidelines and broader governmental policies. This includes being vigilant about data privacy, avoiding the input of sensitive information into AI systems and recognising the potential for bias in AI outputs.

Practical applications of AI in government work

To get the most out of AI, it’s essential to match the tool to the task. Here are some practical ways you can start integrating AI into your daily work:

1. Enhancing productivity

AI can act as a “second pair of hands” to speed up processes that are often tedious and time-consuming:

Meeting management: You can use AI to record and transcribe meetings and automatically identify action items. Just make sure everyone on the call is happy with you doing so!
Document translation: In multilingual environments, DeepL offers superior translation services, ensuring accuracy and speed without compromising on quality.
Design and presentation: Tools like Figma, enhanced with AI features, allow for the rapid creation of presentations, adjusting the tone and style of text in real-time to fit your needs.

2. Strategic planning and research

AI can also serve as a “second brain,” helping you think more creatively and strategically:

Scenario planning: When considering future initiatives, you can use AI to model different scenarios. For example, you might ask an AI model to predict the adoption of electric vehicles by 2030, giving you several strategic options based on varying levels of infrastructure development.
Research assistance: AI tools like Elicit or Perplexity can rapidly synthesise vast amounts of information, offering summaries and insights that can guide your decision-making. This is particularly useful in policy research, where time is often limited.

3. Data analysis and visualisation

For tasks involving data, AI can provide quick and accurate analysis:

Excel and Google Sheets: While AI features are increasingly being built into tools like Excel, you can also leverage AI like ChatGPT to guide you through complex spreadsheet functions, offering step-by-step instructions.
Data summarization: If you work with large datasets, AI can help distill key insights, allowing you to focus on the most relevant information without getting bogged down in details. Just be careful — very careful — about what data you’re inputting.

Being cautious with AI outputs

Despite its benefits, AI is not infallible. Here are some important considerations to keep in mind:

Bias and hallucination: AI models can reflect the biases present in their training data, and they can also generate convincing but incorrect information. Always verify AI outputs, especially when they’re being used to inform policy or public communications.
Sensitivity of data: Never input sensitive or personal data into AI systems, especially those that are cloud-based and may not fully comply with data protection regulations such as GDPR.
Human oversight: AI should augment, not replace, human judgment. Use AI to support your work, but always review its suggestions critically.

How to Start

Getting started with AI doesn’t have to be overwhelming. Begin by identifying tasks in your workflow where AI could make a difference — whether by speeding up routine tasks or offering new insights. Experiment with different tools, keeping the following tips in mind:

Experimentation is key: Start small, perhaps by using AI for mundane tasks, and gradually explore its capabilities in more complex areas. The cost of experimentation is low, especially when done privately or within a controlled environment.
Use contextual prompts: When interacting with AI, be clear and precise in your prompts. For example, when asking an AI to classify data, provide as much context as possible to get the best results.
Leverage community knowledge: Engage with colleagues and AI-focused communities to share experiences, tips and best practices. Learning from others can accelerate your understanding and application of AI tools.

Conclusion

AI offers tremendous potential to revolutionise the way government employees work, providing tools that can make processes faster, more efficient and more insightful. However, with great power comes great responsibility. By approaching AI with a blend of curiosity and caution, and by staying informed about the latest developments, you can unlock new levels of productivity while ensuring that your use of AI remains ethical and compliant with governmental standards.

In summary, start small, experiment wisely, and always critically examine AI outputs. The future of work in government is here, and it’s time to embrace it — with care.

This article was written with the help of ChatGPT! How? First, we cleaned up the transcript from the event to remove any times or identifying information. Then we used our knowledge of how to prompt generative AI to turn a transcript into a practical guide.

Do you have any other practical tips to share on the use of AI in government? Join the AI in Government Community of Practice to continue learning with your peers.