JevMade hello@JevMade.com
← Back to getting started videos

JevMade field notes / Video guide

元OpenAI開発者が作った判断しかしないAI「Jev」がエグすぎた

This video introduces TypeSafe AI's decision model Jev, explaining its probability-focused architecture, vendor-claimed low latency and pricing, and community experiments like fast moderation and trading.

Original by 【漫画村】星野ロミの裏知識チャンネルGetting startedBeginner6 min 17 sec Published Source reviewed

Before you press play

What you’ll find in the video

  1. Jev focuses strictly on classification, scoring, and probability outputs rather than generating conversational text, aiming for low latency.
  2. Vendor materials market Jev as dramatically faster and cheaper than general LLMs by charging low input rates and offering zero-cost output generation.
  3. Community experiments demonstrate rapid real-time classification, though high-frequency trading trials produced negative returns despite high self-reported confidence.
Worth knowing

Gemini-assisted video/transcript review. Rapid decision-making and high self-reported confidence scores do not imply accuracy or financial efficacy, as seen in the negative trading experiment.

元OpenAI開発者が作った判断しかしないAI「Jev」がエグすぎた