Anthropic launches cost-tracking tools for enterprise AI users

New tools for AI accountability

Anthropic recently released three new features for its Claude Enterprise platform to help organizations manage their artificial intelligence investments. These tools arrive as companies face increasing pressure to demonstrate that their spending on AI actually results in measurable efficiency. The new features provide deeper insights into how employees interact with the software and what those interactions cost.

The update includes Smart reports, In-app surveys, and Pooled group budgets. These features are designed to bridge the gap between simply using a tool and understanding its impact on a business. By providing data on costs and workflows, Anthropic is helping firms move toward a more disciplined approach to AI deployment.

As Cryptopolitan on MSN reported, these tools are specifically aimed at helping administrators understand the return on investment for their AI spend. While the software provides the data, the ultimate goal is to help companies decide where to allocate their resources most effectively. Each organization can generate 10 Smart reports per month for free to begin this analysis.

How the tracking features work

Smart reports are currently in beta and allow administrators to select specific teams and applications to analyze over a period of up to 28 days. The system looks at samples of Claude transcripts to categorize the types of tasks employees are completing. It also calculates the usage costs based on Anthropic’s standard pricing, regardless of any discounts the company might have.

Beyond just costs, Smart reports can identify technical issues like unsuccessful tool calls and suggest recurring workflows. These identified patterns can then be turned into reusable skills for the organization. However, Anthropic has stated that these reports are disabled by default and are intended to guide investment decisions rather than evaluate individual employees.

In-app surveys offer a more qualitative approach by asking employees directly about their experience with the tool. Administrators can customize the questions to find out what Claude helped them achieve or what tasks would be slower without it. The results can then be exported to help leadership understand the human element of AI adoption.

Managing shared team budgets

To prevent unexpected expenses, Anthropic introduced Pooled group budgets. This feature allows a team to share a single monthly spending limit while still maintaining separate limits for each individual employee. When a user sends a prompt, the cost is deducted from both the individual and the shared team totals simultaneously.

This system creates a safety net for administrators who need to ensure that a team does not exceed its allocated budget. If either the individual or the shared limit is reached, usage is automatically put on hold. This prevents the kind of runaway costs that can occur when multiple users interact with high-volume AI models at once.

Administrators are kept informed of their spending status through a series of automated alerts. These notifications are triggered when the team reaches 50%, 75%, 95%, and 100% of their shared budget. This level of transparency is designed to help managers stay on top of operating costs as their AI deployments expand.

The challenge of measuring ROI

Despite these new tools, a significant gap remains between perceived productivity and actual financial results. While Smart reports can show what was produced and what it cost, they cannot prove that the work led to higher revenue or lower expenses. Research suggests that employees often feel more productive even when measured improvements are not yet visible.

Data from a KPMG survey of senior leaders across 20 countries highlights this divide. While 89% of executives said their companies benefited from AI, only 12% consistently compared those benefits to the costs. This number rose to 48% only for companies that were already seeing some clear benefits from the technology.

Similarly, a McKinsey study found that 80% of companies reported improvements in individual productivity, but only 37% saw an impact on enterprise-level earnings. These figures suggest that while AI is helping people do their jobs, proving it makes a company more profitable remains a complex hurdle for many businesses.

Rising costs and future outlook

The need for better reporting is driven by the rapid growth of AI spending and the difficulty of managing it. A consultant recently claimed that an unnamed enterprise ran up a $500 million Claude bill in a single month due to a lack of usage limits. While this specific figure is unverified, it underscores the risks of deploying large-scale AI without proper oversight.

Anthropic’s own share of token spending has seen a notable increase recently. According to Ramp’s AI Index, Anthropic’s share rose from 51.3% in late September to 60.5% by late October 2026. This trend reflects a broader global increase in AI spending, which is projected to reach $2.7 trillion by 2026.

As infrastructure costs continue to dominate the market, these new enterprise tools represent a move toward maturity. By helping companies see exactly where their budgets go, Anthropic aims to help firms identify which workflows deliver the most value. The next step for many organizations will be using this data to refine their AI strategies and ensure their investments are paying off.