Defining Requirements with AI: ZenTech Launches Closed Beta of 'DefineAI'

ZenTech, based in Shibuya, Tokyo, and led by CEO Akira Sakai, announced on May 11, 2026, the launch of the closed beta for its requirement definition AI service, 'DefineAI'.

DefineAI is a platform designed to enhance the efficiency and quality of requirement definition through features such as automated requirement generation, AI-powered reviews, automatic screen mockup generation, and meeting minutes compilation.

As the adoption of generative and coding AI grows, the importance of accurately defining 'what to build' and 'why' has become increasingly critical in development environments. ZenTech introduces DefineAI as a foundational layer for upstream engineering to support development quality in the AI era.

Background: In the Era of Coding AI, Rework Stems from 'Ambiguous Requirements'

While generative AI is significantly accelerating coding speeds, rework in development projects often stems not from implementation speed itself, but from ambiguous requirements, scattered documentation, misunderstandings among stakeholders, and a lack of acceptance criteria.

Poor quality or missing definitions in the requirements phase often lead to discovered deficiencies later in the process, resulting in project delays and budget overruns. DefineAI was developed to solve this by standardizing requirements definition and providing AI support as a structural mechanism.

About DefineAI

DefineAI standardizes the requirements definition process (templating). For example, teams can establish internal standards for specific project types (e.g., e-commerce systems), ensuring consistent quality regardless of who performs the definition work.

Problems DefineAI Solves

- Inconsistent quality in requirements definition - Over-reliance on individual expertise - Shortage of skilled requirement analysts - Rework caused by ambiguous requirements progressing to development - Unstable output from coding AI due to unclear prerequisite requirements

Key Features

1. Requirement Cards: Structuring requirements to eliminate ambiguity. 2. AI Review: Providing AI feedback on omissions, ambiguity, and contradictions. 3. Mockup Generation: Early visualization to align stakeholders. 4. AI Chat: Assisting in generating cards and refining project requirements. 5. Enterprise-Ready Architecture: Supports air-gapped networks, self-hosting, and auditability.

Statement from CEO 'DefineAI is not intended to replace PMs or SEs. At this stage, requirements should be defined by humans, and DefineAI acts as an upstream infrastructure to support system development in the AI era,' said Akira Sakai, CEO of ZenTech.