344
Productivity & Workflow355
Automation & Workflow225
Software Development251
Marketing & Growth193
AI Infrastructure & MLOps175
Writing & Content Creation204
Data & Analytics142
Photography & Imaging156
Design & Creative170
Customer Support133
Sales & Outreach125
Voice & Speech135
Education & Learning131
Operations & Admin87
OpenAI has shown a limited preview of its Decisions API, which helps software pick from set options quickly and cheaply, similar to TypeSafe AI’s Jev.
In short: OpenAI says it is testing a new “Decisions API” that helps AI systems quickly choose from a set list of options, similar to a tool called Jev.
OpenAI CEO Sam Altman mentioned a new product, called the Decisions API, during the company’s Dev Day event. An API is a way for different software tools to connect and talk to each other, like a plug that lets one app use another app’s features.
Altman said the Decisions API lets OpenAI’s Luna model pick from predefined choices, such as labeling an image by category or selecting an “agent” behavior. An AI agent is software that can take actions on its own, like clicking buttons or sending messages, after you give it a goal.
TechCrunch reports the Decisions API looks similar to Jev, a model released earlier this month by TypeSafe AI. Jev is designed for automation, and it outputs probabilities for a short list of choices. Think of it like giving the AI a multiple-choice question instead of an open-ended essay question.
The Decisions API is in limited preview, and it is not yet clear how close it is to Jev in real-world use. Developers have not publicly tested it much yet, according to the report.
Companies want AI agents that act reliably and do not do harmful things online. Monitoring every action with a large “frontier” AI model can be slow and expensive. Tools like Jev, and possibly OpenAI’s Decisions API, aim to make that monitoring cheaper, so checks can run on every step, like a spellcheck that watches each sentence as you type.
Source: TechCrunch AI