We build it ourselves.
We turn industry and business model data, organizational knowledge, work workflows, and agent collaboration into real AI platforms.
Home/ AI Lab
TIL AI LAB
We build AI platforms ourselves,
and design better ways to work and learn.
TIL AI Lab develops and operates NextBM200, AI ToolBox, and AI Company in-house. We turn business model exploration, organizational knowledge work, customer consultation, and collaboration with AI agents into real services, while studying both what AI does well and the judgments that people must retain.
This experience extends beyond platforms. We integrate AI into organizational research, decision-making, and execution workflows, and design education that enables people to define problems first and exercise critical judgment.
Rather than producing power users who know more features, we develop people and organizations that think better with AI and act responsibly.
NextBM200 · AI ToolBox · AI Company · HAMMA
BUILD · OPERATE · LEARN · VERIFY
A team that has built and used AI firsthand understands both its potential and its limitations.
AI delivers fast answers and high productivity. At the same time, it can confidently present incorrect answers or cause people to skip the thinking process. TIL AI Lab applies a single set of principles across platform development, organizational adoption, education, and verification: what to delegate to AI, which problems and hypotheses people should formulate first, what to verify, and who should have final judgment and responsibility.
We operate it in real work.
We use AI in research, business planning, content, proposals, customer consultation, and recurring knowledge work to assess its results and limitations.
We design ways to think better.
Rather than teaching tool features, we focus on problem definition, thinking before prompting, cross-feedback, critical verification, and learning transfer.
We complete it through human judgment.
We separate AI outputs into facts, interpretations, and hypotheses, then review the evidence and feasibility to support responsible decisions.
We do not replace education and judgment with AI. We add the value of AI to effective education and responsible execution.
AI LAB PLATFORMS
AI Platforms We Build and Operate Ourselves
Our three flagship platforms are TIL AI Lab's living laboratories. They directly address distinct challenges, from idea exploration and organizational knowledge work to customer touchpoints and multi-role agent collaboration. Under MORE, you can also explore internal AI projects that are in development or completed.
PLATFORM · 01
NextBM200
AI for Exploring Industry Changes and Business ModelsDrawing on data from more than 5,000 business models and industry trends, it explores ideas and develops them into concrete business models. Through Trend Explorer, Idea Studio, natural-language business model search, BM reports, and API · MCP integrations, it supports everything from individual ideation to corporate research and evaluation of new businesses.
TREND EXPLORER · IDEA STUDIO
BM SEARCH · BM REPORT
nextbm200.com
PLATFORM · 02
BM SEARCH · BM REPORT
AI ToolBox
A Platform That Turns Organizational Knowledge and Workflows into Actionable AI AppsThe platform runs Inside AI, which supports teams' recurring tasks, and Customer AI, which handles customer questions and consultations, in one place. Teams can create and share prompt apps and workflows, applying internal manuals, operational knowledge, and product information to business planning, market and competitive analysis, marketing, proposals, and customer consultation. Corporate knowledge chatbots and customer-facing AI apps turn website visitors' questions and experiences into lead capture and consultations with people.
AI APPS · WORKFLOW
KNOWLEDGE BASE · CUSTOMER AI
aitoolbox.kr
PLATFORM · 03
KNOWLEDGE BASE · CUSTOMER AI
AI Company
An AI-Native Company Operating System for Small Businesses, Powered by Multi-Role Agent TeamsMultiple AI team leads and sub-agent teams responsible for research, knowledge operations, marketing, creative work, and development divide roles and collaborate. By connecting a proprietary memory system, an external research pipeline, and a slide model, they create PPT · PDF · HTML materials and perform recurring knowledge work. They use company knowledge and work standards captured in the SSOT · SOP, while major decisions and external actions proceed with human approval.
MULTI-ROLE AGENTS · COMPANY MEMORY
RESEARCH PIPELINE · KNOWLEDGE WORK
Onboarding Case
RESEARCH PIPELINE · KNOWLEDGE WORK
MORE · View Internal AI ProjectsCLOSE · Hide Internal AI Projects
BMT-BM Analyzer
Identify the business model structure in a business plan and find comparable leading companies.When a team uploads its business description or business plan, the project uses public data from listed companies and companies subject to disclosure requirements to structure the business model's customers, value proposition, revenue structure, and key activities, and diagnose its strengths and areas for improvement. Based on the analysis, it automatically matches leading companies with similar business models, enabling teams to explore comparable cases, directions for differentiation, and the next validation tasks. This is a corporate BM analysis project.
BM ANALYSIS · DIAGNOSIS
PEER COMPANY MATCHING
PEER COMPANY MATCHING
InsightMiner
Discover the insights needed for the next decision in scattered knowledge.The project uses AI and a scalable topic taxonomy to structure materials scattered across management, marketing, sales, product, and business development, then explores relevant information and connections across departments and work domains. It goes beyond storing information: this knowledge and insight discovery project helps organizations find evidence faster and identify new opportunities.
KNOWLEDGE TAXONOMY · DISCOVERY
CROSS-FUNCTION INSIGHT
CROSS-FUNCTION INSIGHT
SEBM Analyzer
Provides BM diagnostics and leading-case matching using social enterprise disclosure data.Based on social enterprise disclosure data, the project structures social value, business structure, revenue model, key stakeholders, and execution capabilities to analyze and diagnose the business model. It automatically matches leading social enterprises with similar BMs, presents comparative cases and benchmarking points, and helps organizations quickly explore areas for improvement and directions for growth. This is an internal analysis project.
SOCIAL BM DIAGNOSIS
LEADING CASE MATCHING
LEADING CASE MATCHING
AI EXECUTION SYSTEM DESIGN
We Bring Our Platform-Building Experience into Organizational Execution Systems.
Organizations do not achieve better results simply by adopting AI tools. Starting with the problems an organization needs to solve, TIL AI Lab designs the priority tasks for application, the roles of people and AI, the required data and knowledge, verification criteria, approval points, and operating rules as a single execution flow.
Define Problems and Priorities.
We first select problems that affect customer value, decision quality, knowledge reuse, and organizational performance—not just time savings.
Divide the Roles of People and AI.
We distinguish the exploration, organization, drafting, and recurring tasks that AI will support from the problem definition, verification, approval, and responsibility that people will handle.
Connect Data · Knowledge · Tools.
We place organizational documents, work standards, external information, and the necessary AI tools within actual workflows.
Establish Verification and Approval Criteria.
We clarify the criteria for checking facts, sources, quality, security, bias, and execution risks, as well as the points where people must intervene.
Scale from Small-Scale Execution to an Operating System.
We review pilot results and organize them into SOPs, roles, metrics, and a phased rollout roadmap.
Key Deliverables· AI Application Priorities · Human–AI Role Structure · Data · Knowledge Integration Plan · Verification · Approval Criteria · Operating Rules · Phased Execution Roadmap
AI-ENABLED EDUCATION & WORKSHOPS
HAMMA
Human-AI Mentoring Motif ApproachLearn How to Think Better with AI, Not Just How to Use AI Tools.
HAMMA is TIL's approach to AI education in which people think first, exchange cross-feedback with AI, and complete the process through human judgment. Rather than redefining the essence of education around how to use AI, it adds the exploratory and feedback value of AI to self-directed learning, creativity, problem definition, and problem-solving.
Explore HAMMA Education · Workshops01 · Human FirstPeople interpret the problem and first develop questions, criteria, hypotheses, and drafts.
02 · AI Cross-feedbackAI suggests counterarguments, blind spots, missing variables, and alternative interpretations.
03 · Human Critical LoopPeople separate AI's suggestions into facts, interpretations, and hypotheses, then decide why to accept, revise, or reject them.
04 · Human DecisionPeople make the final decision after considering the organization's context, values, responsibilities, and feasibility.
05 · AI AssistAI structures the agreed conclusions into documents, strategic proposals, canvases, and follow-up action items.
AI SECOND OPINION
Organizations That Use AI Well Do Not Take Its Outputs at Face Value.
If an AI execution system creates the flow of work, AI Second Opinion independently reviews whether the resulting outputs are robust enough to be used in actual decision-making.
In business plans, proposals, research, strategic proposals, and executive reports, it distinguishes claims supported by evidence from assumptions that require further verification, identifies missing perspectives, leaps in logic, business viability issues, and execution risks, and organizes them into revision priorities.
Explore AI Second OpinionFacts · Sources
Logic · Assumptions
Customer · Market Perspective
Business Viability
Execution Risk
Revision Priorities
Logic · Assumptions
Customer · Market Perspective
Business Viability
Execution Risk
Revision Priorities
THE CHANGE WE BUILD
We Turn Individual Proficiency into Organizational Execution Capability.
From Individual Know-How to Organizational Standards
Instead of relying on the prompts of one skilled individual, we establish problem definition, verification, approval, and execution procedures as shared organizational standards.
From Fast Outputs to Better Judgment
We look beyond the speed of outputs. We improve the quality of evidence, the breadth of alternatives, the reasoning behind judgments, and feasibility together.
From AI Dependence to Human-Led Collaboration
We establish a collaborative approach in which people think first, AI broadens their perspective, and people retain final judgment and responsibility.
From One-Off Exercises to Lasting Capability
We build organizational learning capacity that enables people to explain, evaluate, and address new problems even after the education program or project ends.
BUILD · OPERATE · LEARN · VERIFY
Design What AI Should Do in Your Organization
and the Judgments People Must Retain.
TIL AI Lab combines firsthand experience developing and operating platforms with principles for human-led education. Starting with organizational problems, we connect the role of AI, data and knowledge, verification criteria, and approaches to learning and execution. Before making a list of AI tools, tell us which problems your organization needs to solve better.
ANSWER ENGINE
Frequently Asked Questions
Quickly explore The Innovation Lab’s role and methodologies from the perspective of your challenge.
What Does TIL AI Lab Do?
We directly develop and operate AI platforms such as NextBM200, AI ToolBox, and AI Company, and research internal AI projects such as BMT-BM Analyzer, InsightMiner, and SEBM Analyzer. Based on this experience, we provide organizational AI execution systems, human-led AI education, and AI output verification services.
How Is This Different from Typical AI Tool Training?
Rather than training power users who know many features and prompts, we focus on problem definition, thinking before prompting, cross-feedback with AI, critical verification, and learning transfer. We use the HAMMA approach, in which people think first and retain final judgment.
Which Organizations Need AI Execution System Design?
It is suitable for companies and institutions that already use generative AI but have different practices across teams, lack standards for output verification and approval, or want to turn individual productivity into repeatable organizational performance.
What Are the Deliverables of AI Execution System Design?
We develop AI application priorities, a role structure for people and AI, an integration plan for the required data, knowledge, and tools, verification and approval criteria, operating rules, and a phased execution roadmap.
Can TIL's Platforms Be Used in Education and Projects?
Depending on the project, we draw on our experience operating NextBM200, AI ToolBox, and AI Company. Rather than making platform adoption an end in itself, we select tools and methods suited to the organization's problems and learning objectives.
How Is AI Second Opinion Different from AI Execution System Design?
AI execution system design is a project that establishes AI's role and operating rules within an organization's work and decision-making. AI Second Opinion is a service that independently reviews specific documents written or enhanced with AI to assess their evidence, logic, business viability, and execution risks.