AnswersPublished11 sources
What Atlassian’s Expanded OpenAI Partnership Means for Rovo and Customers
Atlassian’s October 6, 2026, deal expands OpenAI model access across its platform and Rovo, but does not make OpenAI its exclusive provider or GPT 6 Astra Rovo’s default. Rovo can pair OpenAI models with the Teamwork Graph’s context about people, projects, documents, and decisions.
Published byEdited with GPT-6 LunaImages generated with GPT Image 2
Research answer
Key takeaways
- Atlassian’s October 6, 2026, deal expands OpenAI model access across its platform and Rovo, but does not make OpenAI its exclusive provider or GPT 6 Astra Rovo’s default.
- Rovo can pair OpenAI models with the Teamwork Graph’s context about people, projects, documents, and decisions.
- GPT 6 Astra’s standard API rates are $10 per million input tokens and $50 per million output tokens; those rates alone do not determine a customer’s Rovo credit charges.
AI Prompt
openai.comCreate a landscape editorial hero image for this Studio Global article: What does Atlassian’s expanded partnership with OpenAI, announced on October 6, 2026, mean for its platform and customers, including access. Article summary: Atlassian’s October 6 expansion gives its agents and Rovo broader access to OpenAI models, while giving OpenAI-powered tools a more governed way to work with Atlassian data. It is an expansion of an existing relationship. Topic tags: general, general web, documentation. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fake
Atlassian’s expanded partnership with OpenAI brings GPT-6-family models—including GPT-6 Astra and the GPT-5.6 series—into more of Atlassian’s platform and its Rovo AI. The central change is broader access to OpenAI models and more ways to connect OpenAI tools with Atlassian work. It is not a move to a single-provider platform: Atlassian continues to support a mix of models. 1
What the partnership changes
Under the new agreement, OpenAI models will help power agents across Atlassian’s platform and Rovo. The companies are also connecting Atlassian context with OpenAI products such as ChatGPT and Codex. The partnership builds on a relationship that dates to 2023. 1
Rovo’s Teamwork Graph is important to that arrangement. It connects information about an organization’s people, projects, documents, and decisions, giving AI tools a way to draw on work context rather than relying only on a standalone prompt. The aim is to help agents use that context to support work across Atlassian products.
The agreement includes a commercial spend commitment, but the available announcement materials do not give a public dollar amount. 1
What was already possible—and what is new
This is an expansion, not the start of all connections between OpenAI and Atlassian products. The companies had worked together since 2023, and Atlassian had introduced an MCP integration for ChatGPT in December 2025. The new agreement broadens OpenAI model access across Atlassian and Rovo and extends the ways the companies connect their products. 1
OpenAI’s use of Jira is also part of the existing relationship: a session at Team ’26 Europe described how OpenAI uses Jira to support its AI implementation work.
Ideas such as assigning work to AI agents in Jira and tracking it there point toward possible workflows, but should not be mistaken for a claim that every such capability is newly available to customers today. Check Atlassian’s current product documentation for availability and rollout details. 1
MCP connects AI clients with Atlassian tools
Atlassian’s Model Context Protocol (MCP) server provides a connection between compatible AI clients and Atlassian products. Atlassian says administrators can govern how external AI tools read and write data, including through organization-wide settings and controls scoped to apps, spaces, and classification levels. The server uses existing Atlassian permissions for access.
VentureBeat reported that the rebuilt MCP server handles more than 15 million calls a day and offers more than 220 tools across Jira, Confluence, Bitbucket, Loom, and other products. It also reported administrator options including allowlists, audit logs, and the ability to disable write actions. These controls matter because an AI client may be able to change work as well as retrieve information. 1
For customers, the practical question is not just which model an agent uses. It is also which tools the agent can access, whose permissions it operates under, and whether it can make changes. Atlassian’s published MCP documentation describes the permission model and the available admin controls.
Atlassian remains multi-model
The OpenAI deal does not mean Atlassian is dropping other model providers or routing every Rovo task to GPT-6 Astra. Atlassian describes Rovo as using a range of models, including third-party providers, and its partnership announcement presents OpenAI models as part of that broader mix. 1
That leaves room to choose models for different needs, such as capability, speed, and cost. It also means the announcement should not be read as a promise that Astra will become Rovo’s default model. 1
Pricing, Rovo credits, and Astra safeguards
OpenAI lists GPT-6 Astra’s standard API price at $10 per million input tokens and $50 per million output tokens. Those are API rates; they do not, by themselves, tell customers how many Rovo credits a particular task will use or what it will cost under an Atlassian plan. Atlassian describes Rovo credit use as depending on the complexity of an interaction and access to enriched Teamwork Graph context.
OpenAI classifies Astra at the “Critical” cybersecurity-capability threshold in its Preparedness Framework and describes strengthened safeguards for the model. Such protections are relevant when agents use tools: safety controls can affect whether a task proceeds, even when a user’s work is legitimate.
Data access and customer checks
Atlassian’s MCP server connects external AI clients through existing Atlassian permissions, while its AI Trust information says model providers do not use customer inputs and outputs to improve their services. These statements address different parts of the data boundary: permissions govern what a connected client can access, while the trust policy describes how providers handle inputs and outputs.
Before enabling agents, administrators should review which tools are available, whether write actions are allowed, and which organization-level or app-level controls apply. The partnership expands the potential for AI to work with company context; access settings determine what that means in a specific deployment.
Studio Global AI
Continue your research
This page includes a source-backed answer you can continue inside Studio Global.
People also ask
What is the short answer to "What Atlassian’s Expanded OpenAI Partnership Means for Rovo and Customers"?
Atlassian’s October 6, 2026, deal expands OpenAI model access across its platform and Rovo, but does not make OpenAI its exclusive provider or GPT 6 Astra Rovo’s default.
What are the key points to validate first?
Atlassian’s October 6, 2026, deal expands OpenAI model access across its platform and Rovo, but does not make OpenAI its exclusive provider or GPT 6 Astra Rovo’s default. Rovo can pair OpenAI models with the Teamwork Graph’s context about people, projects, documents, and decisions.
What should I do next in practice?
GPT 6 Astra’s standard API rates are $10 per million input tokens and $50 per million output tokens; those rates alone do not determine a customer’s Rovo credit charges.
Sources
venturebeat.comAtlassian deepens its OpenAI partnership with a spend ...