AI Agent FinOps: Token Economics, Cost Optimization, and the FOCUS 1.5 Standard — The Complete Guide for Agentic AI Architects (2026)

An Accenture executive stood at the FinOps X 2026 podium in San Diego and shared a story that made 2,500 practitioners wince: a CIO reported spending $250,000 on Claude in a single Wednesday. Four weeks later, the organization was tracking $400,000 per week — and climbing. The agent loops were working exactly as designed. The budget was not.

This is not an outlier. According to EY’s Total Cost of Agents whitepaper (June 2026), a customer-service chat that once cost $0.04 as a simple API call can balloon to $1.20 when it becomes an agentic orchestration involving tool retrieval, planning, and subagent calls — a 30× increase. Gartner projects that more than 40% of agentic AI projects will be canceled by the end of 2027 due to escalating costs, unclear business value, or inadequate risk controls. And McKinsey’s Enterprise AI FinOps Survey (May 2026) found that 93% of enterprises exceed their AI budgets.

Welcome to what J.R. Storment, the FinOps Foundation’s executive director, called “The Great Token Panic” — Era 4 of generative AI, which began around April 2026 when Uber’s CTO publicly disclosed that the company had burned through its entire AI budget in four months.

This guide breaks down the emerging discipline of AI Agent FinOps: the frameworks, cost models, optimization strategies, and open standards that AgenticOps practitioners need to bring agentic AI spend under control before the CFO does it for them.

Why Agentic AI Costs Are Different from Everything Before

Traditional cloud FinOps tracks compute hours, storage GB, and network egress. AI inference billing added a new unit — the token — but single-call inference was at least predictable. Agentic AI breaks that predictability in two fundamental ways.

The Context Window Tax

Every turn in an agent loop re-sends the entire conversation history — system prompt, tool definitions, prior tool outputs, and accumulated reasoning. In a worked example from PandaStack founder Ajay Kumar’s analysis (October 2026), a 20-step tool-calling agent consumes 730,000 input tokens to produce just 12,000 output tokens — a 61:1 input-to-output ratio. That means 92% of the bill is paying the model to re-read context it has already seen.

With context windows now stretching to millions of tokens per call, and agent architectures that chain, retry, self-correct, and spawn subagents, Storment identified two structural cost drivers at FinOps X: expanding context windows and agentic loops. Goldman Sachs projects global token consumption will reach 120 quadrillion tokens per month by 2030 — a 24× increase from current levels.

The Agent Multiplier Effect

Anthropic’s own directional guidance quantifies the multiplier: a single agent with tools consumes roughly 4× the tokens of a standard chat interaction. A multi-agent system consumes approximately 15×. Independent analysis pushes the upper bound even higher — on BrowseComp, token usage alone explained 80% of performance variance, and Princeton HAL’s study of 21,730 rollouts found that agents can be 100× more expensive while delivering only 1% better accuracy.

SAP’s internal data, presented at FinOps X by Frederik Pohl, told the same story from the enterprise side: while cost per token fell steadily since 2023, total AI spend doubled over the same period. Pohl called it Jevons paradox for AI — cheaper tokens lead to exponentially more token consumption, not lower bills.

The Seven Cost Categories of an AI Agent

EY’s Total Cost of Agents framework identifies seven distinct cost layers for agentic AI. Most organizations only account for the first three in their business cases — costs four through seven emerge later, often as budget-breaking surprises.

EY’s central recommendation: “No agent should be approved, scaled, or renewed without a clear measure of what it produces per dollar.” That requires tracking all seven layers, not just the API invoice.

A banking case study analyzed by Beri.net illustrated this precisely: in a per-loan AI processing workflow, token costs accounted for only 22% of total AI spend. The remaining 78% went to tool calls, vector database queries, human review, and compliance overhead — the hidden layers that EY’s framework exposes.

The Tokenomics Foundation and the New Cost Governance Stack

The industry response to the Great Token Panic arrived at FinOps X 2026 in San Diego (June 8–10, 2026), where two major institutional developments reshaped the AI cost governance landscape.

The Tokenomics Foundation

The Linux Foundation announced the intent to launch the Tokenomics Foundation on June 3, 2026, to work alongside the FinOps Foundation. Its mission: provide a vendor-neutral home for the questions Storment outlined — “How do we measure all this? How do we standardize pricing and price changes? How do we price for uncertainty and failure in our AI experiments?”

Early supporters include Accenture, Booking.com, Flexera, Google Cloud, IBM, JPMorganChase, KPMG, Microsoft, Nebius, Oracle, Salesforce, SAP, and ServiceNow. The Foundation has already published several frameworks, including the Seven-Layer Tokenomics Stack, Big-T Notation, and Total Cost of AI (TCA).

Storment defined tokenomics as three layers: production (making tokens — GPU infrastructure, model training), consumption (model routing, caching, context management, prompt design, orchestration overhead, vector DB spend), and value (business return — with software pricing increasingly shifting from seats to token-based usage).

The State of FinOps and Tokenomics: The Numbers

Two surveys anchor the current state of practice:

State of FinOps 2026 (FinOps Foundation, sixth annual, 1,192 practitioners, $83B+ annual cloud spend): 98% of respondents now manage AI spend — up from 63% in 2025 and 31% in 2024. AI cost management is the top skill gap. “FinOps for AI” is the top forward priority.

State of Tokenomics (September 2026, Tokenomics Foundation, 472 responses across 11 industries): Three in four enterprises cannot confidently prove AI business outcomes to the CFO. Organizations that evaluate or use model routers are four times more likely to demonstrate value. Proving ROI was the most-named challenge, cited by 43% of respondents.

The gap is stark: nearly everyone is managing AI spend; almost no one can justify it.

FOCUS 1.5: The Open Standard That Finally Tracks AI Tokens

The FinOps Open Cost and Usage Specification (FOCUS) is the open billing standard that normalizes cost data across cloud providers. Until now, it had no concept of AI-specific billing. That changes with FOCUS 1.5, confirmed for December 2026 ratification.

What FOCUS 1.5 Adds

FOCUS 1.5 introduces native AI token tracking with four new data dimensions: model identity, input and output token labels, cache status, and requester attribution. It also adds a Price Sheet dataset — the first time vendors will be required to expose pricing metadata in a standardized format.

The evolution has been rapid. FOCUS 1.0 launched in June 2024 as a basic cloud billing spec. Versions 1.2 (June 2025) and 1.3 (December 2025) added SaaS, PaaS, and commitment tracking. Version 1.4 (ratified June 4, 2026) brought invoice reconciliation with 47 columns. But none of 1.0 through 1.4 carried a model identity or token count — the most critical dimension for agentic AI cost attribution.

Adopters at the 1.2 level already include AWS, Azure, Google Cloud, Oracle, IBM Cloud, Nebius, and Grafana Cloud. At 1.3: MongoDB, Vercel, Snowflake, and Databricks. The FOCUS Certified program launched with AWS and Nebius earning the first generator badges.

The FOCUS MCP Server

In a development that directly connects to the broader agentic AI ecosystem, the FOCUS project has released an MCP (Model Context Protocol) server — the same protocol that Anthropic donated to the Linux Foundation’s Agentic AI Foundation in December 2025. This means AI agents can now pull the current FOCUS specification on demand, enabling automated cost analysis and compliance checking by the very agents whose costs are being tracked.

What an Agent Loop Actually Costs: Worked Examples

Abstract multipliers do not help practitioners budget. Here are concrete, worked cost figures from published analyses.

The 20-Step Tool-Calling Agent

Ajay Kumar’s worked example (October 2026) models a realistic agent with an 8,000-token system prompt and tool set, adding 3,000 tokens of tool output per step, generating 600 output tokens per step, across 20 steps:

The difference between uncached Sonnet 5.5 ($1.58) and cached-plus-batch Sonnet 5.5 ($0.208) is a 7.6× cost reduction — achieved purely through infrastructure optimization, with zero change to the agent’s capabilities or accuracy.

Customer Support at Scale

Anthropic’s documentation provides a support-agent benchmark: approximately 3,700 tokens per conversation on Haiku 4.5, translating to roughly $37 per 10,000 tickets ($0.0037 per ticket in model cost). Compare this to Intercom’s Fin agent at $0.99 per outcome — the delta illustrates how much of agentic AI cost sits outside the model invoice.

The Coding Session

An Opus 5 one-hour coding session: 50,000 input tokens ($0.25) plus 15,000 output tokens ($0.375) plus runtime ($0.08) equals $0.705. With 40,000 input tokens served as cache reads, the total drops to $0.525 — a 26% reduction for the same work.

Where the Money Actually Goes

McKinsey QuantumBlack’s July 2026 report reveals that 60% of total agentic AI spend is consumed by response refinement — checking, correcting, and improving outputs, not the initial inference call. Shutterstock’s engineering team reported at FinOps X that output tokens represent approximately 75% of total token usage, and since output tokens cost 5× input tokens at most providers, the output-heavy workload pattern of agents is inherently more expensive per token than input-heavy batch processing.

The FinOps for AI Framework: Crawl, Walk, Run

The FinOps Foundation’s “FinOps for AI” guidance provides a maturity model that maps directly to how AgenticOps teams should approach cost governance.

Crawl: Visibility

At the Crawl stage, the priority is knowing what you spend and where. The core equation is deceptively simple: cost = price × quantity. But for agentic AI, “quantity” is not a number you set — it is an emergent property of the agent’s reasoning, tool-calling patterns, and error-recovery behavior.

Key actions at this stage include allocating costs by mapping each API key to one team, application, and cost center. The recommended tag keys are Project, Environment, Workload, Team, CostCenter, Purpose, Criticality, and ShutdownEligible. AWS now offers granular cost attribution for Amazon Bedrock at the per-app, per-agent, per-user level (GA at FinOps X 2026).

The Foundation recommends showback before chargeback — make costs visible to teams before billing them internally, so they can understand and adjust behavior first.

Walk: Optimization

At the Walk stage, teams establish two baselines: a cost baseline (from invoices and usage data) and a functional baseline (response time, accuracy, reliability, volume). Optimization happens at the intersection — reducing cost without degrading function.

The critical concept here is model fit: the percentage of requests being handled by overqualified models. If your agent routes every tool-call planning step through Opus when Haiku would produce equivalent decisions, you are paying a premium that buys nothing. The KPI is: requests on overqualified models divided by total requests.

MetLife’s Sonali Niswander presented a three-stage framework at FinOps X: consumption visibility first, then optimization through model routing and design-time evaluation, and finally value realization — tying cost reduction to business outcomes.

Run: Value

At the Run stage, the metric shifts from “cost per inference” to “cost per successful outcome.” This is the framework’s most important insight for agentic AI architects, because agent accuracy is probabilistic — a lower-cost model that fails 20% more often may cost more per successful task than a higher-cost model that rarely fails.

The Foundation defines ten KPIs, including cost per inference, cost per million tokens, cost per successful outcome, cache hit rate, anomaly cost percentage, ROI, time to value, and model fit. For organizations with defined ownership of AI economics, the payoff is measurable: they are 3.7× more likely to demonstrate value to the CFO.

Eight Optimization Strategies, Ranked by Payback

Drawing from the FinOps Foundation guidance, EY’s framework, and practitioner analyses, here are the highest-ROI optimization strategies for agentic AI systems, ordered by typical payback speed.

1. Right-Size the Model Per Step

Not every step in an agent pipeline needs the same model. Planning and reasoning steps may require Opus-class capability. Tool-call parsing, output formatting, and simple classification can often run on Haiku-class models at 5× lower cost. Organizations evaluating or using model routers are four times more likely to show value to the CFO, according to the Tokenomics Foundation survey.

Google Cloud’s FinOps AI Explainability Agent (GA) now breaks AI cost down by model, modality, and token direction — enabling exactly this kind of per-step analysis. Microsoft Foundry’s model router operates across an 11,000+ model catalog. For smaller teams and solo practitioners, open-source gateways like LiteLLM (MIT licensed) offer virtual keys per team with max_budget in dollars and configurable budget_duration.

2. Cache the Stable Prefix

Prompt caching is the single highest-leverage cost reduction for agent loops because of the 61:1 input-to-output ratio. Anthropic’s cache read price is approximately 1/10 of input price, with a 1.25× write cost for a 5-minute TTL and 2× for a 1-hour TTL.

A worked example with a 30,000-token stable prefix on Opus 5 across 100 requests within 5 minutes: uncached cost is $15.00; cached (one write plus 99 reads) drops to approximately $1.67 — a roughly 9× reduction. The break-even is just one cache read on the 5-minute TTL and two reads on the 1-hour TTL.

Critical caveat: minimum cacheable prefix lengths vary by model — 512 tokens for Opus 5 and Fable 5, up to 4,096 tokens for Opus 4.6 and Haiku 4.5. Below these thresholds, caching silently does nothing.

3. Batch Non-Interactive Work

Anthropic and OpenAI both offer batch APIs at 50% off input and output pricing. Crucially, batch and cache discounts stack — so a non-interactive agent workload (nightly report generation, batch document processing, scheduled compliance checks) can combine both for the deepest discount.

4. Set Reasoning Effort Explicitly

Reasoning tokens bill as output, and output costs 5× input at most providers. An agent that reasons more than necessary on routine steps wastes the most expensive token class. Princeton HAL’s study found that higher reasoning effort reduced accuracy in most runs — more thinking does not always equal better results, but it always equals higher cost.

5. Compact Context, Then Isolate Subagents

Context editing (trimming accumulated tool results and intermediate reasoning from the conversation history) can deliver dramatic savings. Anthropic’s documented example cut a request from 70,000 input tokens to 25,000 — a 64% reduction. Subagent isolation is the architectural counterpart: a subagent that performs a search and returns only 1,000–2,000 tokens of summarized results to the lead agent keeps large retrieval context out of every subsequent turn in the main loop.

This maps directly to AI agent memory architecture decisions — how and where you store intermediate results determines how much context re-enters the conversation window and at what cost.

6. Audit Surcharges You Did Not Choose

Many agentic AI costs are invisible until you check. US-only inference carries a 1.1× multiplier on Claude 4.6+. Regional endpoints on AWS Bedrock and Google Cloud add a 10% premium. Web search costs $10 per 1,000 searches. Fast mode on Opus doubles the per-token price. Managed Agents runtime adds $0.08 per session-hour. Review your invoices for surcharges that may have been enabled by default or by a team member who didn’t realize the cost implication.

7. Implement Budget Guardrails

Google Cloud’s Spend Caps (private preview since April 2026) provide automated, non-destructive spend ceilings for AI Studio, Gemini Enterprise Agent Platform, and Cloud Run. For self-hosted or multi-cloud setups, open-source gateways offer programmatic controls:

• LiteLLM (MIT): virtual keys per team, max_budget in dollars, budget_duration; returns HTTP 422 by default when exceeded
• Cloudflare AI Gateway: spend limits added June 5, 2026; free gateway tier
• Portkey (MIT, 13,100 GitHub stars): budget limits on Enterprise and select-Pro plans
• Helicone (Apache-2.0, 6,200 stars): open-source observability with cost tracking

EY recommends “agentic circuit breakers” — spend ceilings, call-volume caps, and automatic shutoffs — coupled with per-task unit-cost benchmarking. Session budgets pause at the limit; task budgets are advisory; max_tokens caps output but not input. Know the difference.

8. Measure Cost Per Successful Task, Not Cost Per Call

The central metric for agentic AI FinOps is cost per successful task, not cost per API call. An agent that costs $0.50 per call but succeeds 95% of the time may be cheaper per successful outcome than one that costs $0.10 per call but requires three retries on average. The recommended practice: maintain a test suite of 20–50 real-failure-mode tasks, graded on outcome, run in CI.

The Vendor Cost Governance Stack at FinOps X 2026

Every major cloud provider announced AI cost governance tooling at FinOps X. Here is a practitioner’s summary of what is now available.

IBM’s Bill Lobig captured the pace: AI spend is growing “several times faster than total IT spend,” compressing roughly two decades of cloud FinOps growth into about five years.

Open-Source Tools for Agent Cost Tracking

For teams running self-hosted or hybrid agent infrastructure, several open-source projects now offer production-grade cost tracking.

OpenCost 1.121.0 (August 2026) introduced AI inference cost tracking that joins vLLM token counters to GPU allocation data, exposing an llm_cost_per_million_tokens metric per model. A proof of concept ran across 109 GPUs and 30 models. This is significant for teams running open-weight models — where the “price” is not a vendor invoice but a GPU-hours calculation that traditional FinOps tools miss entirely.

OpenTelemetry GenAI Attributes are maturing with standardized fields: gen_ai.usage.input_tokens, gen_ai.usage.output_tokens, gen_ai.usage.cache_read.input_tokens, gen_ai.usage.cache_creation.input_tokens, and gen_ai.usage.reasoning.output_tokens. These are still marked as Development status in the OTel registry but are already being adopted by observability platforms.

Cast AI’s 2026 Kubernetes Optimization Report (covering tens of thousands of clusters on AWS, Azure, and GCP during calendar 2025) found that average GPU utilization sits at just 5%, with CPU at 8% and memory at 20%. CPU overprovisioning rose from 40% to 69% in a year. For teams running inference infrastructure, this represents an enormous optimization opportunity before even touching token-level economics.

Pinterest’s “Tokenomic Layer Cake”: A Production Architecture

Pinterest engineer Ambud Sharma presented a five-layer cost optimization architecture at FinOps X that serves as a practical reference for enterprise automation architects:

1. Silicon layer — GPU hardware selection and procurement
2. Infrastructure layer — cluster management, scheduling, utilization
3. Inference stack layer — KV caching, prefill/decode separation, batching
4. Model and quantization layer — model selection, quantization trade-offs
5. Routing and governance layer — model routing, budget enforcement, policy

The key insight: optimization at any single layer hits diminishing returns quickly. The compounding effect across all five layers is where the real savings live. A team that implements KV caching (layer 3), routes to smaller models where possible (layer 4), and enforces budget guardrails (layer 5) can achieve cost reductions that no single technique delivers alone.

Building Your Agent FinOps Practice: A Practical Roadmap

Based on the FinOps Foundation’s guidance, EY’s framework, and practitioner reports from FinOps X, here is a phased implementation plan.

Month 1–2: Instrument and Allocate

• Map every API key to a team, application, and cost center
• Deploy cost attribution tags across all agent workloads
• Set up showback dashboards — make spend visible before billing internally
• Establish a cost baseline and a functional baseline (response time, accuracy, volume)
• If running on AWS Bedrock, enable granular cost attribution (now GA)

Month 3–4: Optimize the Big Levers

• Implement prompt caching across all agent loops with stable system prompts
• Evaluate model routing — identify steps that can run on smaller models
• Move non-interactive workloads to batch APIs
• Audit surcharges (regional premiums, fast mode, web search costs)
• Deploy budget guardrails with circuit breakers

Month 5–6: Measure Value, Not Just Cost

• Define “successful outcome” for each agent workflow
• Implement cost-per-successful-task tracking in CI
• Appoint an Agent FinOps Lead or Head of Agent Economics (EY’s recommendation)
• Begin preparing for FOCUS 1.5 adoption (December 2026)
• Report value metrics to the CFO — organizations that do this are 3.7× more likely to retain funding

What Is Coming Next

Several developments will reshape agent FinOps through 2027:

FOCUS 1.5 ratification (December 2026) will, for the first time, give organizations a vendor-neutral standard for tracking AI token costs across providers. Organizations that adopt it early will have a structural advantage in multi-cloud agent deployments.

Tokenomicon, the successor event to FinOps X, debuts in San Diego in June 2027 (June 7–10), with earlier events planned for Amsterdam (September 2026) and London (February 2027). The rebranding signals that token economics has become the center of gravity for the FinOps community.

Google Cloud’s 95% problem: Sarah McMullin from Google shared at FinOps X that approximately 95% of organizations see no measurable AI value yet. Only about 5% have moved past proof of concept into production. As that ratio shifts, cost pressure will intensify — the organizations that build FinOps discipline now will be the ones that scale successfully.

Prudential Financial’s Pooja Kumar summarized the challenge with three “lies” that FinOps teams tell themselves: we have visibility (you don’t — token billing differs from user-typed tokens because applications add system prompts, history, and retrieved content); we are optimizing (you are not — anomalies in AI spend are more frequent and riskier than cloud anomalies); and AI is just another workload (it is not — enterprise automation platforms are fundamentally different cost animals).

FAQs

What is AI Agent FinOps?

AI Agent FinOps is the practice of managing, optimizing, and governing the costs of agentic AI systems in production. It extends traditional cloud FinOps to cover token economics, model routing, agent loop costs, and the total cost of autonomous AI workflows — including governance, failure recovery, and organizational change costs that sit beyond the API invoice.

How much more expensive are AI agents compared to single API calls?

Anthropic’s directional guidance puts single-agent-with-tools at roughly 4× a standard chat call and multi-agent systems at approximately 15×. EY’s worked example shows a 30× increase from a $0.04 chat call to a $1.20 agentic orchestration. The actual multiplier depends on the number of tool-calling steps, context window size, and whether caching is implemented.

What is the FOCUS 1.5 standard and when does it launch?

FOCUS (FinOps Open Cost and Usage Specification) is the open billing standard that normalizes cost data across cloud providers. Version 1.5, targeted for December 2026, adds native AI token tracking — model identity, input/output labels, cache status, and requester attribution — plus a Price Sheet dataset. It will be the first version of FOCUS with built-in AI cost support.

What is the most effective way to reduce AI agent costs?

Prompt caching delivers the highest single-technique ROI for agent loops because agents re-read context on every turn (92% of uncached cost is re-reading). Combining caching with model routing (using smaller models for simpler steps) and batch processing for non-interactive work can reduce costs by 7–10× without affecting accuracy.

What is the Tokenomics Foundation?

The Tokenomics Foundation is a Linux Foundation initiative announced at FinOps X 2026 (June 2026) that works alongside the FinOps Foundation to develop best practices, frameworks, and standards for managing enterprise AI token costs at scale. Supporters include Accenture, Google Cloud, IBM, JPMorganChase, Microsoft, Oracle, Salesforce, SAP, and ServiceNow.

Key Takeaways

• The Great Token Panic is real. 93% of enterprises exceed AI budgets (McKinsey), and Gartner projects 40%+ of agentic AI projects will be canceled by 2027 due to costs.
• Agent loops are structurally expensive. A 20-step agent consumes a 61:1 input-to-output ratio — 92% of the bill is re-reading context. Caching alone can cut this by 7–10×.
• Token cost is only 22% of total agent cost. Tool calls, vector queries, governance, and human review account for the rest. Track all seven of EY’s cost categories.
• FOCUS 1.5 arrives December 2026 with native AI token tracking — the first vendor-neutral standard for agent cost attribution across providers.
• Measure cost per successful task, not cost per call. A cheap model that fails often costs more than an expensive one that succeeds consistently.
• Model routing is a 4× value multiplier. Organizations using model routers are four times more likely to prove AI value to the CFO.
• Start with showback, not chargeback. Make costs visible before billing teams internally — behavioral change precedes budget enforcement.
• GPU utilization is 5%. Infrastructure-level waste often dwarfs token-level optimization opportunities.

References

1. EY, “Unlocking Agentic Value: A New Investment Discipline for the Agentic Era” — Total Cost of Agents Series, Whitepaper 1 (June 2026). https://www.ey.com/en_us/insights/ai/agentic-ai-token-costs
2. FinOps Foundation, “FinOps for AI Overview” (Updated October 2026). https://www.finops.org/wg/finops-for-ai-overview/
3. CIO Dive, “Foundation to Tackle AI Token Cost Management” — Makenzie Holland (June 2026). https://www.ciodive.com/news/foundation-tackle-ai-token-cost-management/822839/
4. Flexera, “FinOps X 2026 Recap: 20+ Key Announcements” (June 2026). https://www.flexera.com/blog/perspectives/finops-x-2026-recap/
5. Ajay Kumar / PandaStack, “FinOps for AI in 2026” (October 2026). https://www.ajayk.sh/blog/finops-for-ai-token-costs-focus-2026/
6. Revefi, “Key Takeaways From FinOps X 2026” (June 2026). https://www.revefi.com/blog/key-takeaways-finops-x-2026
7. Gartner, Press Release on Agentic AI Project Cancellations (June 2025). Referenced via EY.
8. McKinsey QuantumBlack, Enterprise AI FinOps Survey (May 2026) and AI Report (July 2026). Referenced via Forkast News.
9. Princeton HAL, Agent Cost-Performance Study — 21,730 Rollouts (2026). Referenced via Frenchy Digital.
10. Cast AI, 2026 Kubernetes Optimization Report. Referenced via PandaStack analysis.
11. Menlo Ventures, Enterprise AI Spend Analysis (December 2025). Referenced via Forkast News.
12. Goldman Sachs, Token Consumption Projection — 120 Quadrillion Tokens/Month by 2030. Referenced via Flexera.

This article is part of the AgenticOps series on RPABOTS.WORLD — practical guides for running agentic AI systems in production.