Rounda Corporation Launches Corporate Support to Resolve 'Operational Confusion' Post-Generative AI Implementation

Rounda Corporation (Headquarters: Shibuya-ku, Tokyo; President and CEO: Takumi Iwata; hereinafter "the Company"), a provider of data consulting and generative AI implementation support, has launched a corporate service that provides end-to-end support for the utilization of generative AI in enterprises, from implementation consideration to operational establishment.

This support covers the comparative verification of multiple generative AI models such as GPT, Claude, and Gemini, the organization of utilization policies for each business operation, the design of usage rules and internal deployment processes, pilot implementations in small teams, and retrospective reviews of utilization status.

Additionally, by utilizing "RoundaChat," a generative AI environment for teams and projects developed and provided by the Company, clients can cross-use multiple LLMs, compare outputs, share within teams, and check usage status, aiming to establish generative AI utilization tailored to the practical operations of each company.

Background: Generative AI moves from "whether to implement" to "how to differentiate usage and establish it"

In recent years, the business utilization of generative AI, particularly ChatGPT, has spread rapidly. On the other hand, the following challenges are increasing in corporate environments:

- Not knowing which model to choose among ChatGPT, Claude, Gemini, etc. - Usage methods varying erratically across departments and personnel. - Wanting to confirm the effectiveness and cost impact per task before signing a company-wide contract. - Inability to decide on generative AI usage rules and internal deployment processes. - Conducting PoCs but failing to lead to routine business establishment. - Lack of internal personnel to consult or staff to drive the initiative.

Simply implementing generative AI as a tool does not lead to business results. It is necessary to identify the suitable model for each task, organize usage rules and evaluation criteria, and integrate it to a state where frontline staff can use it continuously.

The Company has supported corporate business transformation through data infrastructure construction, generative AI implementation, LLM utilization, and project support. In 2025, the Company will officially launch "pj-one," which supports the launch, progression, and management of projects using generative AI and LLMs, initially promoting AI use in project management centered around DX and system construction.

Based on these initiatives, the newly launched corporate support for generative AI implementation and operational establishment is specifically structured around the theme of "how to choose, how to use, and how to establish generative AI," an issue faced by many companies.

Support Details: End-to-end support from model selection to internal establishment

This support provides a combination of the following contents tailored to the company's generative AI utilization status and business challenges:

1. Hearing of AI utilization status and organizing issues We organize the generative AI tools currently used, internal usage status, desired tasks for utilization, concerns, and security/cost/operational challenges. Specifically, we start with status assessment for companies where some employees use generative AI individually, but the organization cannot manage or deploy it systematically.

2. Comparative verification of multiple LLMs including GPT, Claude, and Gemini The required performance varies by task, such as text creation, meeting minutes summarization, internal document search, research, inquiry response, code generation, and data analysis assistance. This support does not assume the implementation of a single model, but compares multiple LLMs to organize the suitability and cost-effectiveness for each task. Using RoundaChat, clients can compare responses from multiple LLMs on the same screen, advancing model selection while confirming quality, speed, and cost parameters.

3. Design of usage rules and guidelines To expand AI utilization internally, rule design such as "what it can be used for," "what information can be inputted," and "how to verify outputs" is essential. The Company organizes generative AI usage rules, prompt design policies, output verification perspectives, and information management precautions tailored to business contents and organizational structure.

4. Pilot implementation in small teams/projects Rather than proceeding with a company-wide rollout suddenly, we first test generative AI utilization in small teams or specific tasks to confirm effectiveness, challenges, and operational load. It is intended to test multiple LLMs on a team basis before a company-wide contract to optimize model selection, usage rules, and cost parameters.

5. Internal study sessions, seminars, and utilization workshops To establish AI utilization, sharing frontline understanding and utilization concepts is as important as preparing tools and rules. The Company offers internal seminars and workshops for management, managers, and frontline staff covering topics like basic understanding of AI, business use cases, common PoC failure points, and integration into practical operations.

About RoundaChat RoundaChat is a generative AI environment provided for teams and projects. It allows cross-utilization of multiple LLMs (GPT, Claude, Gemini), comparison of answers, sharing of prompts and histories, team evaluations, and confirmation of usage status. It features the ability to compare multiple models on practical tasks before implementing a single generative AI service, allowing companies to consider the usage method that suits them best.

About Free Consultations and Seminars Upon the launch of this support, we are accepting free consultations regarding generative AI implementation and operational establishment. Companies can consult based on their unique situations regarding themes like "which AI model should we use," "what should be verified before company-wide implementation," "how to prepare internal rules," and "how to proceed to establish AI utilization in the field."

About Rounda Corporation With the mission of "creating a future without information disparity," Rounda Corporation operates in data infrastructure construction, data utilization support, generative AI/LLM implementation support, and AI product development. They provide end-to-end support from data infrastructure to AI implementation, supporting corporate business transformation based on internally proven methods.