Oracle launches Fusion Claw for governed AI execution

Oracle has launched Fusion Claw, a new execution runtime for its Fusion Agentic Applications. It is available through 25 new applications across finance, human resources, supply chain and sales.

The launch expands Oracle's agentic software portfolio to 75 applications, as it seeks to move customers from AI assistants that support individual tasks to software that can complete more of a workflow with limited human intervention.

Fusion Claw combines AI reasoning with deterministic enterprise computation, allowing models to handle planning and decision-making while conventional software carries out large volumes of repeatable execution. The setup is intended for longer-running work that may require continuous replanning and auditable controls.

Among the 25 new applications are tools for ledger work in finance, workforce staffing in HR, shipping consolidation in supply chain operations and account territory planning in sales. These applications can model options, evaluate constraints and execute actions within predefined governance settings when authorised.

Governance is central to the product design. Organisations can set objectives, policies, permissions, risk thresholds, approval requirements and escalation boundaries through what Oracle calls an Enterprise Operating Envelope, while an Outcome Trust Harness applies those controls to each run.

Each completed process generates an Outcome Receipt that records the authority applied, the evidence used, the decisions made, the actions taken and the final result. Customers can choose different levels of automation, from assisted work to full automatic execution within delegated authority.

The software runs on Oracle Cloud Infrastructure and uses frontier AI models including Gemini and OpenAI models. Support for additional models is planned over time.

At the centre of Oracle's pitch is a business case for measuring AI by operational results rather than model usage or token consumption. The separation between AI reasoning and deterministic computation could reduce the cost of running complex AI workloads by limiting model use to the parts of a process where judgement is needed.

That approach reflects a broader shift in enterprise software, as suppliers try to show that generative AI can do more than help employees with drafting, search or summarisation. Vendors are increasingly building systems that connect AI tools to transactional software used in finance, workforce management, logistics and customer operations, while adding guardrails intended to satisfy audit, compliance and management oversight requirements.

Mike Sicilia, Chief Executive Officer, Oracle, outlined that position in announcing the product.

"Every organisation is looking at AI and asking the same question: how can we use it to create measurable value for our business?" said Mike Sicilia, Chief Executive Officer, Oracle. "Fusion Claw helps answer that question by moving from AI assistance to execution, enabling customers to automate increasingly complex work across their most important business processes, with enterprise-grade governance. By letting AI take on more of the work, while people set the objectives and guardrails, organisations can unlock tremendous capacity for their teams to focus more on driving growth, innovation, and serving customers."

Oracle is also linking Fusion Claw to its broader Fusion AI software stack. The new applications sit alongside Oracle AI Agent Studio for Fusion Applications, which lets customers and partners build and manage agentic applications with no-code and pro-code tools.

Partner response

Large consulting and advisory firms backed the release, highlighting demand among companies looking to move AI projects beyond trials and isolated use cases.

"Together with Oracle, we're helping organisations move from isolated AI pilots to scaled adoption by connecting technology with the operating models, governance, and change capabilities needed to realise value," said Joe Ucuzoglu, Global CEO, Deloitte. "We look forward to advancing that journey alongside Oracle, supporting more complex, outcome-oriented work while keeping accountability and control at the centre."

Google Cloud also pointed to the role of foundation models in business software.

"The next phase of enterprise AI will require models that can reason through complex problems while working within the systems businesses use every day," said Kevin Ichhpurani, President, Global Partner Ecosystem, Google Cloud. "By bringing Gemini models to Oracle Fusion Claw, Oracle is combining frontier AI with the context, controls, and execution capabilities of its market-leading enterprise applications. This creates an important foundation for AI that can take on more sophisticated work while operating within the boundaries organisations define."

Industry analysts framed the launch as part of a progression from task-based copilots to software aimed at complete outcomes.

"Fusion Claw is an important and logical next step in the evolution of enterprise agentic AI," said Aly Pinder, Research Vice President, Aftermarket Services Strategies, IDC. "Oracle is moving the conversation from agents that assist with individual tasks to a next generation of applications that can take on increasingly complex work around a business outcome. With an execution architecture that can combine AI reasoning with precise enterprise computation and governed action, Oracle Fusion Applications customers can gain significant value from enterprise AI."