Oracle adds Fusion Claw, a governed runtime for long-running enterprise agents

Wednesday, 7 October 2026By Sridhar Mukkandi1 min read

Oracle introduced Fusion Claw on 29 September, a governed execution runtime for its Fusion Agentic Applications, built for longer-running work such as research, computation, simulation and continuous re-planning. It launches with 25 Claw-powered applications across finance, HR, supply chain and customer experience, bringing Oracle's portfolio to 75 agentic applications.

Customers set out what an agent may do: objectives, procedures, policies, permissions, risk thresholds, decision rights and approval requirements. Each run then records the authority applied, the evidence used, the decisions made and the transactions executed. Automation is graduated, from quick assistance up to full automatic execution within the authority a customer delegates.

So what

The pattern is worth copying with any stack: write down what an agent may decide, set the risk level above which it must ask, and keep a receipt of every run. Those three things are what make an agent auditable.

SourceOraclewww.oracle.com/news/announcement/oracle-extends-fusion-agentic-a Go deeper · BookDecide, Don't Generate

Drafted with AI from the original source, then checked and edited by Sridhar Mukkandi. Spotted a mistake? Write to hello@tarkika.com and we'll correct it here.

More from The Signal

Models Cloudflare releases Clef, open decision models that answer with probabilitiesThe probability is what makes a decision model useful: code can act on a confident answer and hand a borderline one to a person. Check how well those probabilities match reality on your own data before you pick the cut-off. Cloudflare blog Tools AWS open-sources Strands Decider, a 2B model that picks an answer and says how sure it isA small model that runs on a laptop and reports its own confidence is a cheap first gate for agent steps such as routing, tool choice and guardrails. The team's own guidance is the part to copy: below 0.9, confirm the answer or ask a person. Strands Labs on GitHub Models Reflection announces Beam, a 501B open-weight model it says needs 3 to 4 times less computeIf the efficiency claim holds, an open model near the top of the open field at a third of the compute would cut the cost of serving reasoning-heavy agents. Wait for the weights and run your own tests before planning around it. Reflection AIAll stories