> For the complete documentation index, see [llms.txt](https://usdr.gitbook.io/digital-service-toolkit/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://usdr.gitbook.io/digital-service-toolkit/digital-service-teams/ai-augmented-team-delivery.md).

# AI-Augmented Team Delivery

How a small digital service team can use AI to increase their delivery capacity

## How AI Fits Into Your Team

AI augmentation is not a separate digital service team model. [Executive, enterprise, and hybrid teams](/digital-service-toolkit/digital-service-teams/digital-service-team-models.md) can all use AI tools as part of how they research, design, build, test, analyze, and document their work.

AI tools can help cross-functional team members contribute across more parts of delivery and reduce time spent on routine tasks. They do not replace the people or expertise needed to understand problems, make decisions, and deliver reliable services.

The [next page](/digital-service-toolkit/digital-service-teams/ai-augmented-team-delivery/structuring-an-ai-augmented-team.md) explains the capabilities the team still needs and how they can be organized within different team structures.

{% hint style="info" icon="head-side-gear" %}
Using AI tools to accelerate the work gives the team more space to focus on user needs, make outcome-oriented decisions, check the quality of its work, and iterate based on feedback.
{% endhint %}

<figure><img src="https://2300667447-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FvrFiXac7VON42MNzAMFq%2Fuploads%2FjXYDZxPqbGB6nUzjFPJW%2Fhow%20ai%20fits%20into%20your%20team%208326.png?alt=media&amp;token=a6d0d401-60b6-4edb-8647-f555c54c4541" alt="Infographic showing that AI-augmented delivery can increase capacity within executive or enterprise team models without changing the team’s placement, priorities, authority, or required safeguards. AI can accelerate research synthesis, prototyping, coding, testing, analysis, and documentation."><figcaption></figcaption></figure>

### Recent shifts enabling AI-augmented delivery

Since 2024, two shifts have expanded what smaller digital service teams can realistically build, launch, and maintain:

1. AI development tools — Claude Code, Gemini CLI, Codex, and similar — can support more routine coding, testing, and analysis work
2. Government cloud platforms increasingly support iterative development, testing, and deployment

{% hint style="info" icon="head-side-gear" %}
Together, these shifts make it more realistic for a small digital service team to build, launch, and maintain focused workflow improvements that previously required more staff or outside support.
{% endhint %}

## Potential benefits and risks

AI tools can help a small digital service team increase its capacity, but they can also introduce new delivery, quality, and sustainability risks. Weigh both before deciding how the team will use them.

| Potential Benefits                                                                                  | Potential Risks                                                                                    |
| --------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------- |
| Helps a small team prototype and deliver more quickly                                               | Delivery may move faster than security, privacy, accessibility, procurement, or approval processes |
| Allows [cross-functional staff](#user-content-fn-1)[^1] to contribute across more parts of the work | A small team may become dependent on one highly technical person                                   |
| Reduces time spent on routine analysis, documentation, and development tasks                        | AI-generated work may introduce errors or quality issues without careful review                    |
| Creates an opportunity to demonstrate value before expanding the team                               | Prototypes may become unsupported production tools without clear ownership and maintenance         |
| Builds internal knowledge about users, workflows, data, and technology                              | Teams may automate a flawed process before fully understanding the problem                         |
| Reduces reliance on vendors for some smaller tools and improvements                                 | Staff may be expected to cover too many capabilities without enough time or expertise              |

## Where AI-Augmented Teams Can Make an Impact

Local government can be a strong environment for an AI-augmented team. Local governments manage many focused workflows that affect residents and staff. They may also have more flexibility when selecting approved tools and shorter approval pathways for some projects.

<figure><img src="https://2300667447-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FvrFiXac7VON42MNzAMFq%2Fuploads%2F3EDwEn4lFZtDD48IkmYo%2Fservices%20that%20can%20benefit%20from%20ai-augmented%20teams%202.png?alt=media&amp;token=5c5d1413-11f0-4f41-94d1-c56b318da866" alt="Infographic showing eight local government services that can benefit from AI-augmented teams: parks and recreation, transit, sanitation, code enforcement, public works, permitting, public records requests, and internal operations. A cityscape with trees, a bench, and a bus appears below the service icons."><figcaption></figcaption></figure>

These services may offer opportunities to reduce manual work, simplify processes, improve access to information, support staff, or test an idea before making a larger investment. The team structure needed will depend on the work’s users, workflows, data, integrations, operational demands, and consequences of failure.

{% hint style="info" icon="head-side-gear" %}
The [next page](/digital-service-toolkit/digital-service-teams/ai-augmented-team-delivery/structuring-an-ai-augmented-team.md) will help you assess those factors and choose a team structure that can responsibly deliver and maintain the work.
{% endhint %}

[^1]: Cross-functional team members contribute across multiple areas, such as a product manager who also conducts research or an engineer who supports data analysis