Q&A

 

​

​​

Q1: What is Categorical AI and what are its characteristics?

 

A1: Categorical AI is an advanced artificial intelligence technology owned by New York General Group (NYGG). This AI system surpasses conventional AI in many metrics, particularly excelling in technological innovation and research and development. A notable characteristic is its ability to generate large number intellectual properties annually, with their total value estimated to exceed $500 billion.

 

​

Q2: What is the evaluation process for intellectual property generated by Categorical AI?

 

A2: The evaluation process is rigorous and multifaceted, as follows:

 

1. Fair value assessment based on IFRS standards by GPT-4 (GPT-4o)

2. Income approach (revenue projection) as the primary valuation method

3. Audit report based on IFRS by GPT-4o

4. Review and checks by third-party experts, including a Qualified Institutional Investor, Ph.D. holders in some fields, and so on.

5. Submission of comprehensive documents to the State of Delaware, Secretary of State, Division of Corporations, the regulatory authority for the General Corporation Law of the State of Delaware and in-kind investments in the State of Delaware

6. SEC Filings to Securities and Exchange Commission, the regulatory authority for Securities Act of 1933 and securities issuances in the United States of America: currently in preparation​

7. IRS tax forms to Internal Revenue Service, the regulatory authority for Internal Revenue Code and tax operations in the United States of America: currently in preparation​

​

This process forms a cycle where NYGG's AI generates intellectual property, and third-party AI evaluates it. According to research from The University of Chicago and Stanford University, GPT-4 demonstrates superior ability to human experts in predicting corporate earnings and evaluating scientific technologies, and according to a study by University of Wisconsin, GPT-4 has passed all of the major accounting-related qualifications in the United States of America: Certified Public Accountant (CPA), Certified Management Accountant (CMA), Certified Internal Auditor (CIA), and EA (Tax Accountant) and performed as well as or better than many accounting professionals, ensuring a high degree of objectivity and accuracy. The valuation report, regulatory authority certificate and audit report can be seen at:

​

https://www.newyorkgeneralgroup.com/newyorkgeneralgroupstakeholderrelations

​​​​​

​This valuation may be subject to review based on third-party checks and reviews, market and technology trends, and so on.

​

Q3: What is NYGG's intellectual property management strategy?

 

A3: NYGG's intellectual property management strategy consists of the following elements:

 

1. In-kind contribution to the subsidiary Murakami Foundation.

2. Strict management as trade secrets

3. Compliance with Delaware General Corporation Law, Securities Act of 1933 and so on

4. Potential future patent applications (in the United States of America or Japan)

 

This strategy maximizes the value of intellectual property while ensuring legal protection.​

 

​

Q4: Is Categorical AI considered Artificial General Intelligence (AGI) or superintelligence?

 

A4: NYGG positions Categorical AI as an early version of AGI or superintelligence. This view is based on:

 

1. Microsoft Research believe that GPT-4 could reasonably be viewed as an early (yet still incomplete) version of an artificial general intelligence (AGI) system.

2. Categorical AI surpasses GPT-5.1 (Successor to GPT-4) in various metrics.

 

However, it's important to note that there is ongoing debate about the definition and criteria for AGI and superintelligence, with no absolute consensus reached yet.

 

​

Q5: What is NYGG's information disclosure policy?

 

A5: NYGG refrains from active information dissemination in the media to protect our trade secrets and address potential risks associated with Categorical AI and its generative technology. However, the company is not a secret entity. Appropriate information disclosure may be made to trusted stakeholders. This policy reflects NYGG's strategic approach to advancing responsible development of innovative AI technology while maintaining competitive advantage. The company carefully balances the technology's influence and social responsibility while promoting the development and application of Categorical AI.

 

​

Q6: What are the potential risks of Categorical AI and its generative technology that NYGG is concerned about?

 

A6: The potential risks of Categorical AI and its generative technology that NYGG is concerned about include the following specific threats:

 

1. Advanced misinformation generation: The possibility of creating highly sophisticated fake information and deepfakes using AI, causing social confusion and political manipulation.

2. Sophistication of cyber attacks: The risk of increased attacks on critical infrastructure and financial systems due to automated hacking techniques utilizing AI.

3. Autonomous weapon systems: The danger of unmanned weapons equipped with AI being used without regard for international law or ethical considerations.

4. Privacy invasion: Promotion of profiling and surveillance society through large-scale collection and analysis of personal data.

5. Economic disruption: Destabilization of financial markets due to misuse of high-frequency trading and algorithmic trading.

6. Vulnerability of biometric systems: Neutralization of security systems by breaching biometric authentication using AI.

7. Acceleration of social division: The possibility that AI-driven personalized content recommendations may reinforce confirmation bias and deepen social divisions.

 

​

Q7: Please explain in detail the countermeasures NYGG is considering against the potential risks of Categorical AI and its generative technology.

 

A7: The countermeasures NYGG is considering against the potential risks of Categorical AI and its generative technology include a multifaceted approach:

 

1. Information Control Strategy:

   - Restraint in active information dissemination in media: Minimizing public information about technical details and progress to reduce the risk of misuse.

   - Selective information disclosure: Releasing only essential information through carefully selected, trusted channels.

 

2. Enhanced Security:

   - Implementation of multi-layered defense systems: Strictly managing access to AI models and data to prevent unauthorized access and information leakage.

   - Utilization of quantum cryptography: Introducing quantum cryptography to counter advanced threats that conventional encryption methods cannot address.

 

3. Ethical Framework Construction:

   - Establishment of an AI ethics committee: Organizing a committee including external experts to formulate and oversee ethical guidelines for AI development and use.

   - Regular ethical audits: Continuously evaluating AI system behavior and impact to identify and address potential issues early.

 

4. Technical Measures:

   - Robustification of AI models: Enhancing resilience against adversarial attacks and inappropriate inputs to prevent unexpected behavior.

   - Adoption of federated learning: Reducing the risks of centralized data management by distributing the learning process.

 

5. Legal and Regulatory Approach:

   - Proactive regulatory compliance: Anticipating trends in AI technology regulations and establishing compliance systems.

   - Participation in industry standard development: Actively engaging in the formulation of industry standards for AI technology safety and reliability.

 

6. Education and Awareness Activities:

   - Strengthening internal education programs: Implementing continuous education on AI ethics and safety for employees.

   - Dialogue with stakeholders: Deepening understanding of AI's social impact through dialogue with policymakers, academic institutions, and civil society.

 

7. Risk Assessment and Management:

   - Regular risk assessments: Continuously evaluating new risks associated with AI technology evolution and updating countermeasures.

   - Scenario planning: Preparing response strategies in advance by envisioning various scenarios, including worst-case scenarios.

 

8. Research and Development Direction Adjustment:

   - Safety-focused R&D: Emphasizing the development of AI technologies that minimize potential risks.

   - Management of dual-use technologies: Paying special attention to the development and management of technologies with potential military applications.

​​​

​

Q8. In which fields, other than technological innovation and R&D, can Categorical AI be particularly applied?

​

A8. Categorical AI shows exceptional promise in two key fields: First, in finance, particularly in investment banking due diligence and valuation processes. This technology enables comprehensive analysis of corporate value assessment, going beyond traditional financial data to encompass complex relationships between companies, market trends, and industry structures. For instance, in M&A transactions, it enables more sophisticated determination of appropriate corporate values and optimization of investment portfolios by considering intricate interrelationships that conventional methods might overlook. Furthermore, it allows for more precise analysis in evaluating market and credit risks. Second, in science, particularly in computer simulation-based analysis. When analyzing complex natural phenomena and experimental data, this approach can reveal structural relationships that traditional statistical methods might miss. For example, in weather forecasting, molecular dynamics, and quantum system simulations, it enables efficient calculations while preserving the essential structure of the systems being studied. This capability leads to more accurate predictions and potentially new scientific insights. What makes Categorical AI particularly effective in these fields is its ability to systematically analyze relationships between complex structured data while maintaining mathematical rigor. Additionally, its results are highly interpretable, facilitating verification and judgment by domain experts.

​​

​

Q9: How was the analysis posted on the NYGG homepage conducted?

 

A9: Many of the analyzes in the NYGG's homepage are based on computer simulations using Categorical AI.

​

​

Q10. Is NYGG publishing academic papers or filing patents?

 

A10. NYGG is not currently actively publishing academic papers or filing patent applications. This strategic decision is rooted in a complex set of reasons, primarily because all technologies invented by Categorical AI are core technologies, and maintaining them as trade secrets is deemed optimal. This approach encompasses technologies generated by Categorical AI and owned on its platform across various fields, including AI, quantum computing, robotics, energy, biotechnology, chemistry, and pharmaceuticals/life sciences. Here's a detailed explanation of this strategy:

 

1. Protection of Core Technologies: All technologies invented through Categorical AI are fundamental to the company's competitive edge. These innovations span across multiple cutting-edge fields. Patenting would require public disclosure, potentially enabling competitors to replicate or circumvent these technologies.

 

2. Technological Interdependence: The technologies generated by Categorical AI are intricately interconnected. For instance, AI algorithms might optimize quantum computing processes, which in turn could be applied to novel molecular designs in biotechnology. This interdependence makes comprehensive trade secret protection more effective than individual patents.

 

3. Rapid Technological Evolution: In these advanced technological domains, the pace of innovation often outstrips the patent process (typically 3-5 years from filing to approval). Trade secret protection allows for more agile adaptation to market changes and continuous technological advancement.

 

4. Global Competitive Advantage: Properly managed trade secrets can theoretically be protected indefinitely, unlike patents with limited lifespans. This approach enables NYGG to maintain long-term, global competitive advantages, particularly crucial for platform technologies like Categorical AI.

 

5. Flexible Technology Application: Trade secret management allows for more flexible expansion and modification of technology applications. Patents might limit applications to specific uses or implementations, whereas trade secrets permit broader applications, which is especially valuable for versatile technologies like Categorical AI.

 

6. Financial Considerations: Obtaining and maintaining a vast patent portfolio across multiple technological domains would incur substantial costs. By opting for trade secret protection, these resources can be redirected towards further research, development, and business expansion.

 

7. Mitigation of Talent Exodus Risks: Trade secret management allows for strict control over access to critical technological information, minimizing the risk of technology leakage through employee turnover. This involves implementing a comprehensive confidentiality framework across multiple advanced technology sectors.

 

8. Due Diligence Preparedness: Abstaining from paper publications and patent filings also serves to address due diligence concerns regarding trade secret leakage. This demonstrates to potential investors and business partners the company's stringent intellectual property management and strong commitment to protecting confidential information.

 

However, NYGG is not pursuing a policy of complete secrecy. To balance transparency with technological superiority, the company engages in limited information disclosure:

 

1. White Paper Publications: The company's website features technical reports providing an overview of Categorical AI. These white papers explain the basic principles and potential application areas of Categorical AI without revealing core technological details. This approach showcases the technology's innovativeness and potential value while safeguarding specific implementation details.

 

2. Technology Case Study White Papers: Select application cases of technologies generated by Categorical AI are published as white papers. These case studies demonstrate the practical applicability and market value of the technologies but only present results without delving into the underlying technical processes.

 

3. Limited Technology Demonstrations: Under strict non-disclosure agreements, the company occasionally conducts limited technology demonstrations for specific business partners or potential investors. These demonstrations aim to prove technological efficacy without disclosing critical technical details.

 

4. Industry Conference Presentations: NYGG executives occasionally speak at industry conferences. These presentations discuss technology trends and future outlooks without revealing specific details about the company's technologies.

 

5. Press Releases: The company issues press releases about significant technological milestones or business partnerships. These releases outline achievements or partnership overviews without disclosing specific technical information.

​

In conclusion, NYGG's intellectual property strategy strikes a balance between rigorous trade secret management and strategic, limited information disclosure. This approach aims to maintain technological superiority while gaining external trust. The strategy is optimized to protect and leverage Categorical AI as an innovative platform technology and the diverse core technologies it generates. However, this strategy is subject to continuous re-evaluation and potential adjustment in response to changes in the technological landscape, market conditions, or legal regulations. For instance, if certain technology areas become standardized or if open innovation gains more importance in the future, the company might consider patent filings or academic publications for specific technologies. Through this flexible and comprehensive intellectual property strategy, NYGG aims to secure sustainable competitive advantages at the forefront of rapidly evolving technological innovations.

 

​

Q11. What is NYGG's accounting policy?

​

A11. NYGG uses AI to perform all accounting and auditing, including asset valuation. The AI used is from a third-party organization that has proven to have expertise equal to or greater than that of human experts and is publicly available, and the inputs to and outputs from the AI are disclosed on NYGG's website so that third parties can verify the accounting and auditing processes. We also obtain checks and reviews from third-party experts and professional organizations as necessary, as well as filing and disclosures to regulatory authorities.

 

​

Q12. Is NYGG hiring new graduates?

A12. As of 2025, NYGG is not hiring new graduates.

 

​

Q13. Where can we use NYGG's Categorical AI?

​​

A13. You can access and use it in the following link.​ A Poe registration account (free) is required for use. Note that these models are prototypes with only the minimum viable function. Please use them as the trial version. These bots also use in part Claude Opus 4.8 model run by Anthropic.

​

Our AI Models

​​​​

​

​Q14. How does NYGG implement Categorical AI based on category theory?

​​

A14. NYGG instantiates and implements the highly abstract model called Categorical AI using concrete models such as BERT (Bidirectional Encoder Representations from Transformers) (while BERT is mentioned here, any Transformer-based large language model such as GPT or Claude would suffice) or PVCNN (Point-Voxel Convolutional Neural Network), thereby minimizing implementation barriers. This implementation approach bridges the gap between theory and practice by mapping abstract structures to existing deep learning architectures. By doing so, it reduces the complexity involved in transforming advanced theoretical concepts into practical systems, resulting in a framework that is accessible to both researchers and engineers.​​

​​​

​

Q15. How can third parties verify NYGG’s claims and technical achievements?

 

A15. NYGG seeks to enhance verifiability wherever reasonably possible. Claims regarding technological advantages, intellectual property value, and business potential may be supported by valuation reports, audit materials, regulatory filings, AI input/output records, third-party reviews, and other documentation, to the extent permitted by confidentiality obligations and applicable law.

Although the core technology of Categorical AI is protected as a trade secret, NYGG maintains records regarding its evaluation processes, analytical methods, supporting documents, and external review procedures. This approach is intended to balance the protection of confidential technology with corporate transparency.

 

 

Q16. Does NYGG use external auditors, attorneys, accountants, or other experts?

​

A16. NYGG may engage third-party professionals in law, accounting, tax, intellectual property, artificial intelligence, finance, science, and technology when appropriate. Matters involving intellectual property valuation, in-kind contributions, regulatory filings, investor communications, material contracts, and corporate governance may require review beyond internal judgment.

When AI is used for analysis or evaluation, NYGG does not treat AI outputs as automatically conclusive. Instead, such outputs may be cross-checked against human expert review, other AI systems, public information, applicable laws, accounting standards, and contractual practices.

​

​

Q17. How does NYGG manage exaggerated statements or uncertain information in its disclosures?

 

A17. NYGG recognizes the importance of avoiding misleading statements, especially as a company involved in advanced technologies. When discussing research results, business plans, valuation figures, technical capabilities, or regulatory matters, NYGG aims to distinguish clearly between confirmed facts, reasonable opinions, assumptions, and future possibilities. For example, valuation figures or technical descriptions may depend on specific assumptions, methodologies, datasets, review procedures, and market conditions. NYGG may therefore include explanatory notes or qualifications where appropriate and may revise such information if circumstances materially change.

​

​

Q18. What due diligence materials can NYGG provide to investors or business partners?

​

A18. NYGG may provide due diligence materials to qualified investors, business partners, research institutions, financial institutions, professional advisors, and other credible stakeholders when appropriate. These materials may include corporate profiles, business plans, intellectual property management policies, valuation reports, audit-related materials, legal and compliance documents, technical summaries, and risk management policies. However, NYGG carefully limits access to the core algorithms, source code, prompt designs, unpublished inventions, proprietary know-how, and other sensitive trade secrets related to Categorical AI. Even under a non-disclosure agreement, the scope of disclosure may be restricted to protect NYGG’s competitive position and intellectual property.

 

 

Q19. How does NYGG ensure legal and regulatory compliance?

 

A19. As a U.S. company, NYGG’s basic policy is to comply with applicable federal and state laws, including corporate law, securities law, tax law, intellectual property law, privacy law, electronic transaction laws, consumer protection laws, and other relevant regulations. NYGG also considers compliance in connection with AI development and use, including privacy, cybersecurity, ethics, dual-use risks, data protection, and social impact. If new regulations, legal requirements, or industry standards emerge, NYGG may update its internal policies, contracts, disclosures, and technical management practices accordingly.

 

 

Q20. How does NYGG ensure the reliability of AI-assisted accounting or valuation work?

​

A20. When NYGG uses AI for accounting support, asset valuation, audit assistance, or analytical documentation, it treats AI as part of a verifiable process rather than as an automatic decision-maker. NYGG may record inputs, assumptions, evaluation methods, outputs, revisions, and review history so that relevant stakeholders can assess the reasonableness of the process where appropriate. Because AI-generated evaluations have limitations, important conclusions may be checked against accounting standards, applicable laws, professional review, market data, regulatory filings, and other supporting materials. This helps NYGG combine the efficiency of AI with appropriate human judgment and governance.

 

 

Q21. Can NYGG’s intellectual property valuation change in the future?

 

A21. Yes. Any valuation of NYGG’s intellectual property may change over time. Such valuations depend on assumptions existing at the time of evaluation, including technological maturity, market conditions, competitive landscape, commercialization potential, regulatory environment, third-party review, and applicable accounting standards. NYGG does not treat valuation figures as fixed or absolute. Instead, they are managed as estimates based on reasonable assumptions. If there are material changes in market conditions, technological development, legal requirements, or business prospects, NYGG may conduct a reassessment and provide appropriate explanations to relevant stakeholders.

 

​

Q22. How does NYGG respond to inquiries from customers, investors, and partners?

 

A22. NYGG responds to inquiries from customers, investors, researchers, business partners, media representatives, regulators, and other stakeholders in a manner that is professional, accurate, and appropriate to the nature of the request. General inquiries may be answered using publicly available information. Inquiries involving investment matters, technical details, unpublished intellectual property, contract terms, financial information, or regulatory matters may require identity verification, confirmation of affiliation, a non-disclosure agreement, or review by qualified professionals before NYGG provides a response.

​

​

Q23. How does NYGG address cybersecurity and information leakage risks?

​

A23. NYGG treats Categorical AI, intellectual property, research data, contracts, stakeholder information, and confidential business records as important assets. NYGG may use layered security measures, including access controls, logging, encryption, backups, confidentiality agreements, internal training, and approval procedures for external information sharing.

For information managed as trade secrets, access is generally limited to those with a legitimate need to know. If NYGG identifies or reasonably suspects unauthorized access, information leakage, or misuse, it will investigate the scope of impact and take actions required under applicable laws, contracts, and internal policies.

 

 

Q24. What is NYGG’s position on AI safety and ethics?

​

A24. NYGG recognizes AI as a foundational technology with significant social impact, not merely as a tool for efficiency. Accordingly, NYGG regards AI safety, explainability, fairness, privacy protection, misuse prevention, and dual-use risk management as important corporate responsibilities. For advanced AI technologies such as Categorical AI, NYGG seeks to control disclosure carefully in order to reduce misuse risks while promoting beneficial applications in areas such as research, medicine, science, finance, energy, education, and other socially valuable fields. NYGG aims to balance technological innovation with social responsibility.

​

​

Q25. Does NYGG plan to disclose more detailed financial information or business progress in the future?

 

A25. NYGG may expand its disclosures regarding financial information, business progress, research and development, intellectual property management, and related matters depending on its stage of growth, regulatory requirements, relationships with investors and partners, and business needs. However, as a privately held company, NYGG must also consider flexibility, trade secret protection, competitive advantage, and contractual confidentiality obligations. NYGG therefore distinguishes between information suitable for public disclosure, information that may be disclosed on a limited basis, and information that must remain confidential.

​​

​

​

For inquiries regarding these matters, please contact:

E-Mail: info@newyorkgeneralgroup.com

​

​

​