OpenAI launches Decisions API described as a Jev clone

By Splitfeed EditorialLast updated Sep 30, 2026, 19:10 UTC

On September 29 and 30, 2026, The New Stack and TechCrunch both reported on OpenAI's release of a product called the Decisions API, built on an internal platform named Luna. The New Stack framed the release as OpenAI's direct competitive answer to TypeSafe's Jev product, while TechCrunch described the Decisions API as a Jev clone designed to deliver fast, cheap intelligence for coordinating what the outlet called OpenAI's swarming autonomous agents. Neither outlet published detailed pricing or capability benchmarks for the Luna platform at the time of reporting.

The significance of this release sits at the intersection of two trends that have been building quietly: the emergence of lightweight decision-layer products designed specifically to orchestrate multi-agent systems, and the intensifying competition between major AI labs and smaller specialized firms. If TypeSafe's Jev established a market category around low-latency, low-cost inference for agent control, OpenAI entering that space with a comparable product is a meaningful competitive signal. What remains genuinely in dispute is whether the Decisions API is a straightforward imitation, a direct technical parallel that happens to solve the same problem, or a product that merely resembles Jev in name and framing but differs in meaningful ways under the hood.

None of the three camps, Pro-AI, Anti-AI, or Middle Ground, had published reactions at the time of this report. Pro-AI voices would typically welcome a release like this as evidence that the infrastructure for reliable, scalable autonomous agents is maturing. Anti-AI voices would likely raise concerns about the implications of tools explicitly designed to coordinate swarming agents at scale, focusing on control and accountability questions. Middle Ground observers would probably focus on the competitive dynamics, asking whether OpenAI's entry into this niche crowds out smaller innovators like TypeSafe or simply validates the category.

The clearest thing to watch next is any technical disclosure around the Luna platform, including latency figures, pricing, and how the Decisions API integrates with OpenAI's existing agent products. A detailed response from TypeSafe about the comparison to Jev would also sharpen the picture of whether this is a competitive threat or a parallel development.