{"id":25732,"date":"2026-10-26T13:00:05","date_gmt":"2026-10-26T06:00:05","guid":{"rendered":"https:\/\/fpt-is.com\/en\/?post_type=goc_nhin_so&p=25732"},"modified":"2026-10-26T14:33:38","modified_gmt":"2026-10-26T07:33:38","slug":"jev-ai-classification-model","status":"publish","type":"goc_nhin_so","link":"https:\/\/fpt-is.com\/en\/insights\/jev-ai-classification-model\/","title":{"rendered":"What is Jev? Why has it been called the \u201cChatGPT moment\u201d for classification models?"},"content":{"rendered":"

Abstract<\/strong><\/span><\/h2>\n

In mid-September 2026, the startup TypeSafe AI announced Jev, an AI model designed for fast, low-cost classification and decision-making. The model quickly went viral, attracting considerable attention from the AI research and developer communities. Many developers experimented with Jev and built and shared a range of applications. Jev is believed to have numerous potential applications and may replace large language models (LLMs) in classification and decision-making tasks. This article explains what Jev is, why its release has been described as the \u201cChatGPT moment\u201d for classification, how to use it through an application programming interface (API), representative use cases, and limitations to consider before deployment.<\/span><\/p>\n

1. What is Jev, and why has it attracted attention?<\/strong><\/span><\/h2>\n

1.1. Why are LLMs not optimized for simple decisions?<\/strong><\/span><\/h3>\n

Large language models are fundamentally text-generation models conditioned on an input context. Given a context, a model generates successive tokens based on the probability distribution of the next token.<\/span><\/p>\n

This characteristic gives LLMs several disadvantages. First, they can hallucinate, generating content that does not correspond to reality, particularly when they encounter contexts that are rare in their training data. Second, for tasks that require only a classification decision, using an LLM can be cumbersome and poorly matched to the nature of the problem: the output is free-form text that must be parsed before software can use it.<\/span><\/p>\n

1.2. Jev: an AI model for fast decision-making<\/strong><\/span><\/h3>\n

Jev was developed by TypeSafe AI, a startup headquartered in San Francisco, United States. TypeSafe was founded in 2024 by Diogo Almeida, who previously contributed to the development of instruction-following training methods at OpenAI. According to TypeSafe AI\u2019s introductory blog post, Jev is the first model in the \u201cSystem One\u201d category: a class of AI models designed to make fast, structured decisions that software can consume directly.<\/span><\/p>\n

The term \u201cSystem One\u201d derives from a concept popularized in Daniel Kahneman\u2019s book Thinking, Fast and Slow. Kahneman, an Israeli-American psychologist, received the Nobel Memorial Prize in Economic Sciences in 2002. In Kahneman\u2019s account, System One is fast, automatic, intuitive, and relatively effortless, whereas System Two is slower, deliberate, analytical, and dependent on focused attention.<\/span><\/p>\n

Jev was developed to address the limitations of LLMs in classification tasks. Unlike an LLM, Jev does not generate text; it is optimized for decision-making. Jev can be understood as a function call: the input is unstructured data, while the output is a structured classification result with a predefined data type and a probability for each possible option.<\/span><\/p>\n

\"Image<\/a><\/p>\n

Figure 1:compares how an LLM and Jev approach the same classification problem.<\/span><\/em><\/span><\/p>\n

According to TypeSafe, Jev is particularly useful when embedded at decision points within a workflow. In agentic AI applications, for example, Jev can select which model should execute a task (model routing) or check inputs and outputs (guardrails), at substantially lower cost and latency than an LLM.<\/span><\/p>\n

1.3. How did businesses handle classification before Jev?<\/span><\/h3>\n

Classification is a classic problem in artificial intelligence. An AI model might, for example, classify emails as spam or non-spam, or classify comments under a social media post as abusive or non-abusive. Many problems addressed by AI can be formulated as classification tasks. Even an LLM\u2019s process of predicting the next token from the input context is, in essence, a classification problem: the context is mapped to one token in the vocabulary, with each token treated as a class.<\/span><\/p>\n

Classification problems have typically been addressed in one of two ways:<\/span><\/p>\n