The Age of Unmanaged AI Is Over: Introducing New Tools for Guild Insights

Introducing new tools for Guild Insights:

• A single view of how AI is being used across your organization and platforms — daily usage and spend by user, agent, platform, and provider
• Month-over-month trends, with up to 90 days of historical data
• Downloadable reports for finance, leadership, and operational reviews
• Budget Limits to define and stop unexpected spending
• Optimization tools to identify better models for tasks, streamline prompts, trim tools, and mix deterministic code with non-deterministic LLM calls

AI adoption is no longer happening in one place.

Developers are using Cursor. Teams are calling models from Anthropic and OpenAI. Agents are running through Guild and other platforms. LLM calls are also being embedded directly into products, workflows, and backend systems.

For leaders, that creates a new management problem.

You may know how much the organization is spending on AI overall. But do you know which teams are driving that spend? Which platforms they're using? Whether usage is growing? Or where costs will land at the end of the month?

Today, we're introducing new tools for Guild Insights: a single view of how AI is being used across your organization, what it costs, and where you may need to take action.

Connect your provider keys, and in under five minutes, Guild can begin surfacing up to 90 days of historical usage data. No lengthy implementation. No new instrumentation across every workflow.

Just a clear view of what is happening.

Most AI platforms show you the activity happening inside their own products. That's useful, but incomplete. Organizations are operating across multiple tools, providers, models, workspaces, and agents. Managing that environment requires a view that cuts across all of them.

Guild Insights brings together usage and spend from platforms and providers including Cursor, Anthropic, OpenAI, and Guild. Leaders can see:

• Daily AI usage and spend
• Total spend by platform and provider
• Usage and costs by team and user
• Month-over-month trends
• Projected monthly spend
• Up to 90 days of historical data
• Downloadable reports for finance, leadership, and operational reviews

And for tools running on Guild, Insights can show even more granular data, allowing you to break down your spend by use case (for example, code review, debugging, ticket triage, research). This isn't something available in the high-level numbers you get on other platforms.

Guild Insights gives engineering, platform, and finance leaders the visibility to understand not only what the organization is spending on, but also how teams are actually adopting AI.

You can see where usage is growing, which tools are becoming part of daily work, which teams are experimenting, and where spend is increasing without a clear operational plan behind it.

AI costs can accumulate across dozens of accounts, tools, APIs, and expense lines. By the time invoices reach finance, the people responsible for managing the technology may have little context for what created them.

A single total does not tell you enough.

It doesn't show whether spend came from a successful production agent, an engineering team adopting a new coding tool, an inefficient model choice, or LLM calls running continuously in the background.

Guild Insights makes that usage attributable.

Leaders can break spending down by platform, provider, workspace, agent, team, or user. They can compare current activity with previous periods and see where the organization is heading before the next bill arrives.

This is not about monitoring individual prompts or policing developers. It's about giving teams the information they need to manage AI as a shared organizational resource.

Developers get the freedom to use the tools that help them work. Leaders get the visibility required to understand adoption, manage budgets, and make better investment decisions.

Historical reporting tells you what already happened. Managing AI also requires understanding what is likely to happen next.

Guild Insights uses current usage to project month-end spending across platforms and providers. If consumption begins to rise unexpectedly, teams can identify it while there's still time to respond.

Organizations can also set budgets for LLM activity, including calls happening inside agents, applications, and backend workflows.

Budget Limits provide an additional layer of protection. Rather than discovering an uncontrolled workflow after it has consumed the quarter's budget, organizations can define limits and stop unexpected spending before it becomes a larger problem.

Visibility shows you the issue. Control lets you do something about it.

Set budgets for specific LLM activity

Set budgets with built in check points 

Receive notifications when you reach important budget thresholds 

Budget notifications are delivered via email

AI usage is no longer relevant only to the developers building agents.

Engineering leaders need to understand adoption across their teams. Platform teams need to see how shared infrastructure is being used. Finance needs reliable forecasts and cost attribution. Security and governance teams need to know which providers and tools are operating across the organization.

Guild Insights gives each of those groups a common view.

Downloadable reports can be filtered, reviewed over different periods, and exported for planning, budgeting, or leadership discussions. Everyone works from the same underlying data rather than reconciling separate dashboards and provider invoices.

Data is available in downloadable formats for easier sharing with teammates

The first phase of enterprise AI was defined by experimentation. Teams tried new tools, connected APIs, and built agents wherever they could.

The next phase requires management.

Guild Insights gives organizations the visibility to understand AI usage across teams, tools, providers, and agents. Budget Limits provide control. Optimization helps teams get more from the resources they already consume.

Connect your keys. See up to 90 days of activity in under five minutes. Then start managing AI with the clarity you expect from every other critical part of your infrastructure.