— Empowering researchers and practitioners to build, evaluate, and deploy AI models from their own image data — without writing code —
TOKYO, Japan, April 10, 2026 — CarbGeM Inc. (Headquarters: Shibuya-ku, Tokyo; CEO: Masakazu Nakajima), a developer of AI-powered life science solutions, today announced the launch of ‘AI Studio,’ a new capability within its CarbConnect® AI development and deployment platform. AI Studio enables researchers, engineers, and frontline practitioners to build, evaluate, and operationalize AI models from image data—all without specialized programming or AI development expertise.
With AI Studio, users can seamlessly perform the entire workflow—from image data upload and annotation to AI model training, evaluation, and deployment—within the CarbConnect® environment. By removing the dependency on dedicated AI specialists or external vendors, AI Studio transforms the point of data generation into the point of AI creation and application, enabling a fundamental shift in research and operational workflows.
Research institutions and testing laboratories accumulate vast quantities of high-quality experimental and observational data, including microscopy images. However, much of this data remains siloed and underutilized.
Adopting AI in these settings has traditionally required programming skills, substantial computing resources, and specialized development teams—barriers that have made practitioner-led AI adoption difficult. Fragmented data management has further impeded knowledge sharing, standardization of assessment criteria, and research reproducibility.
Against this backdrop, national science and technology policy has increasingly prioritized the integration of experimental data and AI to transform research processes, alongside efforts to build environments where researchers across all disciplines can leverage AI. AI Studio has been developed as a practical platform that addresses these needs by unifying high-quality research data management with practitioner-driven AI development.
AI Studio is a suite of capabilities that enables AI model development, evaluation, and deployment using various types of data—without requiring specialized AI expertise. It integrates not only data storage and sharing, but also streamlined AI model development and implementation into a single platform.
■ Data upload and cloud-based management for diverse data types
■ Cross-team data sharing, commenting, and discussion
■ Image annotation (labeling)
■ No-code AI model training using labeled datasets (via AI Studio)
■ AI model evaluation and accuracy assessment
■ Deployment of trained AI models for research and operational use
■ Multi-site, multi-user collaborative access to data and AI models
■ Continuous data accumulation for model refinement and retraining (AI-ready dataset formation)
Conventional AI development relies on specialized personnel or external contractors, with data collection, processing, and model training operating as disconnected steps. AI Studio fundamentally reimagines this process.
| Conventional Approach | CarbConnect® AI Studio |
| Data siloed by individual or facility | Centralized cloud management, sharing, and search |
| AI development dependent on specialists or outsourcing | Practitioners themselves build and evaluate models |
| Data collection and AI development disconnected | End-to-end workflow from collection to deployment |
| Accumulated data rarely reused | Iterative design: models improve as data grows |
■ Unified Data and AI Platform: CarbConnect® consolidates data collection, sharing, analysis, and AI development in a single environment—from annotation through model training and evaluation.
■ Integrated Communication and Data Utilization: Built-in commenting and discussion features connect team knowledge sharing directly to data analysis workflows.
■ Research Data Infrastructure: Consolidates previously fragmented data into a unified system for management, preservation, sharing, and utilization—supporting the transformation of research data into an institutional knowledge base.
■ Continuous Model Evolution: An iterative design in which AI model accuracy improves as data accumulates, naturally fostering the formation of AI-ready datasets.
■ Operational Connectivity: Integration with existing applications enables immediate deployment of trained AI models into research and operational workflows.
■ Collaborative multi-site image analysis and discussion
■ Development of organization-specific AI models leveraging historical data
■ Standardization of assessment criteria and improved reproducibility for quality management
■ Data accumulation and knowledge base formation for educational purposes
■ Continuous improvement cycles: data accumulation → model refinement → redeployment
■ Exploratory research through data-driven hypothesis generation
CarbGeM positions CarbConnect® not merely as an AI development tool, but as a foundational platform for implementing data-driven research and operational processes at the point of practice.
Our vision is to build a framework in which data generated through daily research and operations is continuously accumulated and channeled through a cycle of management, sharing, analysis, modeling, and reapplication—connecting directly to the next discovery or improvement.
By enabling practitioners to iteratively refine AI models without reliance on specialized organizations, CarbConnect® creates a virtuous cycle among data, models, and the field—driving productivity gains across research and operations.
Looking ahead, CarbGeM will continue to leverage CarbConnect® to advance scientific research and deepen data utilization in applied settings. In particular, amid the growing global movement to transform research through the convergence of experimental data and AI, we will strengthen collaboration with universities and research institutions to promote environments where researchers can translate their own data into actionable insights.
Service Name: CarbConnect® ‘AI Studio’
Launch Date: April 10, 2026
Service Page: https://www.carb-connect.com/aistudio
