The government is buying AI by the token. It should buy results.

TL;DR

GSA (General Services Administration) has signed a 27-month OneGov agreement with OpenAI to provide ChatGPT access to roughly 23 million government employees on a consumption-based, per-token pricing model with a 50% discount. This replaces the prior $1 per user monthly approach and establishes a major AI procurement vehicle across federal agencies. The procurement model shifts cost drivers from per-seat/subscription pricing to usage volume, which the source argues conflicts with Executive Order 14402’s stated preference for fixed-price, performance-based contracts. Immediate implications: agencies and contractors will need to reassess acquisition strategy, pricing models, and capture plans; contracting officers may face pressure to reconcile token-based billing with outcome-focused acquisition guidance; and firms that price or structure proposals around performance outcomes may have competitive and advisory opportunities. Expect follow-on solicitations, task orders, and guidance clarifying how agencies will apply performance metrics and budget management to tokenized consumption pricing.

Key Points

• What happened: GSA signed a 27-month OneGov agreement with OpenAI to provide ChatGPT to about 23 million government employees under a consumption-based, per-token pricing model with a 50% discount, replacing a prior $1 per user monthly model.
• Who is affected: NAICS 541512, 541511, 541519, 518210, 541513; agencies listed in segmentation including GSA, DOD, DHS (Department of Homeland Security), VA, HHS, DOJ, DOE, DOT, Treasury, State; contract vehicles including OneGov, OASIS+, 8(a) STARS III, Alliant 3, SEWP; market segments and compliance regimes in segmentation.
• Timeline: 27-month OneGov agreement (per Summary).
• What contractors should do NOW: review capture and pricing strategies to address consumption-based billing; model impact of token pricing on cost, margins, and total cost of ownership for agency customers; prepare performance-based alternatives and value-based pricing narratives aligned with Executive Order 14402; update pipeline and win themes to target OneGov and related vehicles; and stand up monitoring for follow-on solicitations and agency guidance.

Who Is Affected

Affected segments include federal IT/AI suppliers, software and cloud service providers, and professional services firms that support federal AI deployments. Specific NAICS codes, agencies, contract vehicles, market segments, and compliance regimes named in the segmentation are affected:

• NAICS: 541512, 541511, 541519, 518210, 541513
• Agencies: GSA, DOD, DHS, VA, HHS, DOJ, DOE, DOT, Treasury, State
• Contract vehicles: OneGov, OASIS+, 8(a) STARS III, Alliant 3, SEWP
• Market segments: Artificial Intelligence, IT Services, Cloud Services, Software as a Service, Digital Transformation, Enterprise Software, Professional Services
• Compliance surfaces: FedRAMP (Federal Risk and Authorization Management Program), NIST 800-171 (NIST Special Publication 800-171), FISMA, FAR (Federal Acquisition Regulation) Part 12, FAR Part 16, Executive Order 14402

Frequently Asked Questions

Q: Does this procurement create a government-wide AI buying vehicle?

A: The Summary says the OneGov agreement with OpenAI establishes a major AI procurement vehicle across federal agencies. For agency-specific ordering guidance and scope, pending source review.

Q: How does this model compare to prior pricing?

A: The Summary contrasts the consumption-based, per-token pricing model (with a 50% discount) against the previous $1 per user monthly model. For detailed rate tables or contract terms, pending source review.

Q: Is this consistent with Executive Order 14402?

A: The Summary cites an argument that token-based pricing is inconsistent with Executive Order 14402’s preference for fixed-price, performance-based contracts because token pricing incentivizes usage volume rather than measurable mission outcomes. How contracting officers will reconcile the two is pending source review.

Definitions

• OneGov: The named contract vehicle used by GSA in the Summary to deliver the agreement.
• per-token pricing: A consumption-based pricing model where charges are driven by tokens consumed, as described in the Summary.
• ChatGPT: The OpenAI product being provided under the OneGov agreement in the Summary.
• Executive Order 14402: Cited in the Summary as preferring fixed-price, performance-based contracting approaches.

Intelligence Response

• Cabrillo Signals War Room — Already detected this event and delivered this briefing. Use War Room to maintain the event record, collate source materials, and push alerts to capture and BD teams.
• Cabrillo Signals Match Engine — Re-score opportunity pipelines and task-order-fitting models to reflect token-based pricing risk/opportunity and to reprioritize pursuits that can monetize outcomes rather than consumption.
• Cabrillo Signals Intelligence Hub — Track follow-on solicitations, agency clarifications, and task orders tied to OneGov and affected agencies. Save searches on SAM.gov (System for Award Management) and alert when applicable opportunities appear.
• Proposal Studio (Proposal OS) & Proposal Studio Workflow Tracker — Rapidly update proposal content, compliance matrices, and win themes to articulate performance-based alternatives and to produce audit-ready capture documentation through the 9-gate Workflow Tracker.

Who to notify:

• BD Lead — assess pipeline impact and reallocate pursuit resources.
• Capture Manager — revise capture plans and win strategies for OneGov solicitations.
• Pricing Lead / Finance — model margin impact of token pricing vs. fixed-price engagements.
• CTO/CISO — evaluate technical and security implications of large-scale ChatGPT use for agency customers.
• Proposal Manager — update proposal artifacts and compliance matrices.

First 48-hour playbook

• Hour 0–4: Confirm detection in Cabrillo Signals War Room; distribute this briefing to BD, Capture, Pricing, and CTO/CISO. Flag immediate pursuits tied to OneGov.
• Hour 4–12: Run Match Engine rescoring on active pipelines; identify high-risk/high-opportunity pursuits and flag for urgent capture reviews. Create saved searches in Intelligence Hub for OneGov/OpenAI-related notices.
• Hour 12–24: Convene capture + pricing workshop to model token consumption scenarios vs. fixed-price and draft performance-based alternatives. Start updating Proposal Studio templates and win themes.
• Hour 24–48: Finalize prioritized pursuit list, assign Workflow Tracker gates, and prepare outreach playbook for agency liaisons and partners to clarify ordering guidance and performance metrics.

Primary hub: Winning Federal Contracts Guide (/insights/winning-federal-contracts)

Related guides:

• CMMC (Cybersecurity Maturity Model Certification) Compliance Guide (/insights/cmmc-compliance-guide)
• CUI (Controlled Unclassified Information)-Safe CRM Guide (/insights/cui-safe-crm-guide)