Advanced Biomed Inc. (ADVB)

Advanced Biomed Inc. operates in the enterprise artificial intelligence financial verification market, a segment the company characterizes as remaining at an early stage, where adoption depends on financial institutions' technology budgets, regulatory requirements, model performance, integration costs, data availability and users' willingness to add automated verification to existing workflows. The demand drivers for AI financial verification are the wide use of large language models and AI agents by financial institutions in credit analysis, investment research, risk assessment and compliance review; the risk that AI output contains errors from fabricated or miscalculated figures, confused entities, incorrect dates for event sequences, logical contradictions, incorrect policy or regulation citations, and overly optimistic or pessimistic assessments which in turn can lead to poor rating inaccuracies, poor credit decisions, compliance and legal risk; increasing regulatory focus on the explainability, accountability, and accuracy of AI used by regulated financial institutions; and audit and internal-control requirements for documented review and audit trails of the analysis used in decisions and disclosures. The company has fully exited its historical biomedical operations, having sold Advanced Biomed HK together with its subsidiary Shanghai Sglcell in December 2025 and Advanced Biomed Taiwan in June 2026, and has shifted entirely to developing its AI application through the FinTruth Financial Verification Agent Platform.

FinTruth's competitors include large financial technology companies, AI model providers, enterprise software companies, and financial institutions developing internal verification systems. The company believes its competitive strengths include FinTruth's design as an independent verification layer that evaluates AI output against source evidence, financial constraints and domain-specific rules, rather than relying only on improvements in the underlying generative model; Smith RM, its proprietary financial reasoning model, and the platform's ability to work with models specified by customers; the combination of four verification modules (numerical, logical, ontology and narrative) in a three-layer framework, with a financial rules knowledge base covering PRC, IFRS/HKFRS and U.S. GAAP accounting standards; claim-level, source-level evidence tracing with a three-panel linked review interface and exportable audit trails; flexible deployment through SaaS, API or on-premises, so that customer data can stay within the customer's environment; and its proprietary financial datasets and expert-annotated evaluation datasets. The company acknowledges it is at an early commercial stage, with a limited operating history, a relatively small customer base, a small team compared with large technology and financial-data companies, limited cash resources and limited brand recognition among large financial institutions, and that AI model providers or financial institutions may also build similar verification capabilities into their own products. The company does not currently hold ISO/IEC 27001, SOC 2 or other security certifications or attestations, which some large financial institutions require of their vendors.

Advanced Biomed Inc. is a holding company incorporated in Nevada whose revenue is generated through its AI technology operations, which currently consist of continued development and operation of FinTruth; development of financial verification algorithms, financial rules and related data and knowledge infrastructure; product and software engineering; enterprise product testing, deployment and customer implementation; discussions, demonstrations and commercial engagement with prospective financial-institution and technology customers; and development of commercial models including enterprise subscriptions, usage-based application programming interface fees and integrated solutions. The company plans to generate revenue through three channels: usage-based API fees, including token- or transaction-based metering, charged to financial institutions, financial IT companies and other enterprise customers on a per-report or per-task basis for both web and API use; enterprise subscription revenue through annual or periodic arrangements, with on-premises deployments charged a one-time license fee plus an annual maintenance fee; and integrated or bundled solutions offered in collaboration with foundation-model providers, enterprise technology providers and financial IT platforms, with pricing negotiated with each partner. Target customers include commercial banks, securities firms, insurance companies, asset managers, auditing and professional-services firms, and financial technology companies. During the fiscal year ended June 30, 2026, the company generated approximately $127,637 (HK$1,000,000) ¹ in revenue from its operations related to FinTruth, and as of the date of the report it had not generated any customer revenue from FinTruth, with revenue relating to FinTruth derived solely from agency service fees.

FinTruth is the company's primary AI platform and the sole product offered to customers, operating as a rule-based and model-based check platform with research and development and engineering activities conducted by employees based primarily in Hong Kong. Its reasoning core, Smith RM, is a proprietary financial reasoning model developed and owned by Acellent and is not subject to any third-party license restrictions. The platform is model-agnostic, meaning that, at a customer's request, it can be configured to work with third-party or open-source large language models specified by the customer, and it can also verify the output of third-party AI financial reasoning systems. FinTruth consists of five principal components: a data ingestion layer that ingests corporate credit information, financial statements, business registration information, judicial information and industry data, as well as the AI-generated report and source documents uploaded by the user; the Smith RM reasoning core, which provides step-by-step verification through an API and includes a self-repair mechanism so that issues identified during verification can be fed back for correction; a formal verification engine consisting of four verification modules, being numerical, logical, ontology and narrative consistency verifiers; a financial rules knowledge base consisting of accounting standards, banking regulations and industry benchmarks; and an output layer for report rendering, format conversion and audit logs. The knowledge base covers PRC Accounting Standards for Business Enterprises, International Financial Reporting Standards, Hong Kong Financial Reporting Standards, and U.S. GAAP. The platform currently supports verification and user interface in English and Chinese but does not support cross-language verification, and it currently supports, in production, credit and credit-rating analysis reports, research reports, and financial statements and annual reports, with source materials including financial statements, credit reports, business registration information, judicial information and industry data.

FinTruth currently provides three principal capabilities: document and source comparison, allowing AI-generated financial analysis to be checked against underlying source documents; source-level evidence tracing, allowing statements and numerical values to be traced to supporting evidence; and verification and audit reporting, including consistency checks, confidence indicators and audit trails. In document and source comparison, FinTruth provides a side-by-side comparison of the AI-generated report and source files, with automatic highlighting of differences to identify data deviations and factual conflicts, and in the detailed analysis view a claim list, the AI report and the source materials are linked in three panels. In source-level evidence tracing, FinTruth traces claims and numbers back to their specific location in the source files, and by selecting a claim the reviewer is taken to the corresponding paragraph in the source materials. In verification and audit reporting, FinTruth produces a report overview with an overall pass score, counts of passed, suspect and erroneous claims, chapter-level quality scores, review suggestions and operating metrics covering processing time, token usage and retry counts, and audit reports contain conclusions and evidence and can be exported in PDF and Word formats. Confidence is presented at three levels: claim level, where each claim is classified as pass, suspect or error, with error assigned when a claim contradicts the source documents or violates a hard constraint such as an accounting identity that does not hold, and suspect assigned when the source documents do not contain sufficient evidence to support it but no contradiction is found; report level, being a detection pass rate scored out of 100 calculated as the number of passed claims divided by the total number of claims checked, with suspect and erroneous claims not counted as passed; and chapter level, where each chapter receives a separate score across dimensions such as content completeness and data accuracy, with low-scoring chapters and identified numerical-extraction and calculation errors generating review suggestions. The platform is delivered in three ways: a browser-based SaaS workspace hosted on international public cloud infrastructure provided by multiple cloud providers; APIs, including a step-by-step verification API that allows verification checkpoints to be embedded in a customer's or partner's AI financial-analysis workflow; and on-premises deployment within the customer's own environment. FinTruth supports two modes of use, being standalone verification of a completed report and embedded verification checkpoints, and a typical report is processed in a few minutes.

Significant operational developments during the period include the acquisition of 100% of the equity interests of Acellent Technologies (Hong Kong) Co. Limited on April 2, 2026 in exchange for 270,000 shares of Common Stock valued at an estimated $4.00 per share, or $1,080,000 in the aggregate ², with the acquisition closing on April 30, 2026. The company sold 100% of the equity interests of Advanced Biomed HK, which wholly owned Shanghai Sglcell, to Wei Ha Hui for a purchase price of US$23,000 ³, and on January 26, 2026 Advanced Biomed HK acknowledged indebtedness to the company of $6,925,549 ⁴ to be repaid in full on or before January 26, 2027. On June 30, 2026, the company agreed to sell 100% of the equity interests of Advanced Biomed Inc. (Taiwan) to Miaozhu Zhao for an aggregate purchase price of US$490,000 ⁵. On June 6, 2025, the company entered into the ELOC Agreement with HELENA GLOBAL INVESTMENT OPPORTUNITIES I LTD., pursuant to which it had the right to issue and sell up to Twenty-Five Million United States Dollars ($25,000,000) ⁶ of Common Stock; on January 30, 2026, the company issued 1,650,710 shares of Common Stock (82,536 shares as adjusted for the Reverse Stock Split) to the Investor as commitment fee shares valued at $500,000 ⁷, and on July 14, 2026 delivered notice of termination effective July 21, 2026, with no advances made under the agreement prior to termination. On January 28, 2026, the company issued and sold an aggregate of 4,000,000 shares of Common Stock (200,000 shares as adjusted for the Reverse Stock Split) at a purchase price of $0.062 per share (pre-Reverse Stock Split), for aggregate gross proceeds of $248,000 ⁸, closing on January 29, 2026. On April 13, 2026, the company entered into a loan agreement with Jie Wang pursuant to which Mr. Wang agreed to lend US$600,000 ⁹ for operating purposes, unsecured, bearing interest at 10% per annum ¹⁰ and maturing six months after disbursement, subject to extension for an additional six months by mutual written agreement. The company effected a 1-for-20 reverse stock split effective February 13, 2026, with trading on a split-adjusted basis beginning February 20, 2026 under the new CUSIP number 00752P203, and on March 12, 2026 received a written decision from the Nasdaq Hearings Panel stating that the company had regained compliance with Nasdaq Listing Rule 5550(a)(2). Management changes included the resignation of Dr. Steven I-Fang Cheng as director and Chief Technology Officer effective March 25, 2026, the appointment of Mingze Yin as a director, the cessation of Dr. Yi Lu's service as Chief Executive Officer, director and Chairman effective April 28, 2026, and the appointment of Dr. Xiaomin Chen as Chief Executive Officer, director and Chairman of the board of directors.

The company's financial performance reflects a business in transition, with net cash outflow in operating activities from continuing operations of $0.96 million ¹¹ for the year ended June 30, 2026 and $2.3 million ¹² for the year ended June 30, 2025, and net cash outflow in operating activities from discontinued operations of $0.99 million ¹³ and $3.5 million ¹⁴ for the years ended June 30, 2026 and 2025, respectively. The company had net current assets of $12.3 million ¹⁵ as of June 30, 2026. As of October 13, 2026, Advanced Biomed Inc. had 1,652,133 shares of outstanding Common Stock ¹⁶, par value $0.001 per share ¹⁷, and the aggregate market value of shares held by non-affiliates as of December 31, 2025 was approximately $2.7 million ¹⁸. The company had 12 employees ¹⁹ as of the date of the report, comprising 2 ²⁰ in Executive & Product Leadership, 3 ²¹ in AI Research & Algorithm Engineering, 2 ²² in Product & Software Engineering, 1 ²³ in Data & Knowledge Engineering, 1 ²⁴ in Client Solutions & Commercialization, 2 ²⁵ in Finance & Accounting, and 1 ²⁶ in Legal & Corporate Operations.

The company has adopted a twelve-month business and marketing plan running from July 1, 2026 to June 30, 2027, with objectives that include plans to commercially deploy FinTruth to twenty institutional customers ²⁷, conversion of ten pilot engagements into commercial contracts ²⁸, growth from twelve to approximately eighteen full-time employees ²⁹, and maintaining sufficient liquidity to fund planned operations. The plan consists of three stages. Stage I, covering months 0-3 and described as completed, involved enhancing document comparison, source-level evidence tracing, numerical and logical consistency verification, ontology and entity verification and audit-report generation; improving enterprise deployment, including on-premises deployment; refining financial datasets and rules; establishing pricing models comprising per-report usage fees, subscriptions, and on-premises license and maintenance fees; and holding demonstrations and starting pilot and proof-of-concept engagements with prospective customers. Stage II, covering months 3-6 and described as in progress, is set to place FinTruth in commercial use by paying customers, including commercial banks and financial IT and fintech companies, with between five and ten pilot or proof-of-concept engagements ³⁰ underway or under discussion and the company working to convert them into paying customers and recurring revenue. Stage III, covering months 6-12 and described as not yet started, is set to begin in approximately January 2027.

The company's primary growth vector is the commercial scaling and product expansion of FinTruth, which is currently its sole product. As part of the twelve-month plan, the company intends to possibly extend FinTruth's functionality with additional verification modules, including additional financial-report verification functions and increased compatibility with additional financial documents and AI-output verification applications, and to improve its automated customer onboarding experience and integration capabilities. The company is actively engaging in discussions regarding product deployment, business models and pricing with prospective customers and strategic partners, including financial IT companies, MaaS and AI distribution platforms, and fintech subsidiaries of large financial institutions, and it plans to distribute the product indirectly through financial IT companies, MaaS, AI distribution platforms, foundation-model providers and other strategic technology partners through integrated or bundled offerings. As part of the indirect approach, the company recently entered into an agreement with Quantum Capital (Hong Kong) Ltd as its marketing partner to promote and market FinTruth to prospective customers, and based on future customer demand and available financial resources it intends to expand distribution through additional strategic technology partners and by adding business-development personnel.

A second growth vector is the potential acquisition of a legal AI agent designed to assist with the preparation, review and analysis of disclosure documents relating to U.S. capital markets transactions, covering due diligence, document analysis, disclosure drafting, document categorization and information collection, which could potentially expand FinTruth's customer base. Its target users include law firms, public companies, investment banks, financial advisors, accounting firms and other participants in U.S. securities offerings and capital markets transactions. The company is currently in discussions regarding this potential acquisition but has not yet entered into any binding or definitive agreement with respect to it, and there is no assurance that any transaction will be completed. The company also expects to finance the continued operation and commercialization of FinTruth through a combination of operating revenues, proceeds from the disposition of Advanced Biomed Inc. (Taiwan), and future capital-raising activities, and it has received term sheets and engagement letters relating to potential financing transactions, although none of these transactions may be completed.

The company's cost structure is built around a lean, technology-focused operating structure following the disposition of its biomedical operations, with personnel primarily engaged in AI research and algorithm engineering, product and software engineering, data and knowledge engineering, customer implementation and commercialization, supported by the finance, legal and corporate functions necessary for a Nasdaq-listed company. The company does not own any data centers, and the SaaS version of FinTruth runs on rented infrastructure from multiple international public cloud providers, while on-premises deployments run on the customer's own infrastructure. The company's objective is to develop recurring and usage-based enterprise revenue from its AI technology operations while maintaining an operating structure in which product development, hiring and commercial expansion are aligned with customer adoption, revenue generation and available financial resources. The company requires substantial capital to increase its sales and marketing efforts to drive market adoption of its products and services and address competitive developments; maintain, expand and protect its intellectual property portfolio; hire and retain additional personnel such as AI development and research personnel; develop, acquire and improve operational, financial and management information systems; finance general and administrative expenses; and operate as a public company.

The company's capital allocation priorities center on funding the continued operation and commercialization of FinTruth, with the twelve-month plan contemplating growth from twelve to approximately eighteen full-time employees and the maintenance of sufficient liquidity to fund planned operations. The company expects to fund FinTruth's continued operation and commercialization through a combination of operating revenues, proceeds from the sale of Advanced Biomed Inc. (Taiwan), and future capital-raising activities. The company has no dividend policy disclosed and no share repurchase authorization disclosed. The company's AI research and algorithm engineering team currently consists of three full-time employees , and the company intends to grow its overall headcount under the twelve-month plan. The company's datasets are derived from data generated internally by Acellent and are not purchased or licensed from third parties, and the company's data pipelines, datasets, ontology construction, evidence-chain processing, knowledge-base maintenance and data-quality control are managed by its data and knowledge engineering function.

Management has flagged several structural headwinds and constraints to the growth plan. The market for AI financial verification is new and evolving and may not develop as expected, and if financial institutions slow or limit their adoption of AI, if foundation-model providers substantially reduce hallucination and other errors in their models, or if customers prefer to rely on manual review or internal tools, demand for FinTruth may be lower than anticipated. Commercial banks, insurance companies and other regulated institutions typically have lengthy procurement, security review and approval processes, which may delay or prevent sales. The AI industry is intensely competitive and rapidly changing, with competition from foundation-model providers that may incorporate verification, citation and fact-checking functions directly into their models, from large technology, data and financial information companies, from financial IT companies, and from other AI start-ups, many of which have substantially greater financial, technical, data and marketing resources, larger customer bases and broader distribution. Some of the same parties with whom the company seeks to collaborate, including foundation-model providers and financial IT platforms, may also compete with it, and the rapid pace of innovation in AI may render its technology obsolete or require significant investments to remain competitive.

The company faces geographic, regulatory and macro constraints on its growth plan. Substantially all of the company's operations, personnel and research and development activities are located in Hong Kong, and changes in the political, economic and legal environment in Hong Kong or the PRC, including the possibility that the PRC government may intervene in or influence its operations, could materially and adversely affect its business. The PRC has implemented a series of AI-specific regulations, including the Provisions on the Administration of Algorithm Recommendation for Internet Information Services, the Provisions on the Administration of Deep Synthesis of Internet Information Services, the Interim Measures for the Management of Generative Artificial Intelligence Services and the Measures for Labeling Artificial Intelligence-Generated Synthetic Content, which impose algorithm filing and security assessment requirements, content management and labeling obligations, requirements relating to the legality and quality of training data and protection of intellectual property and personal information, and obligations to prevent discrimination and the generation of false or harmful information. Although the company currently conducts its operations in Hong Kong, FinTruth processes Chinese-language reports, applies PRC Accounting Standards for Business Enterprises, and relies on datasets that include business registration and judicial information, and if the company provides FinTruth, its APIs or Smith RM to customers or users in mainland China, deploys FinTruth on-premises for customers in mainland China, or enters into integrated or bundled arrangements with foundation-model providers or financial IT platforms operating in mainland China, it or its partners may be required to complete algorithm filings, security assessments or other registrations. The United States has imposed, and may continue to expand, export controls on advanced semiconductors, AI model weights and related technology, and has adopted rules restricting certain U.S. investments in AI and other technologies involving countries of concern, which include the PRC, Hong Kong and Macau, and the PRC has adopted countermeasures and its own export control regime, which could limit the company's access to computing capacity, models or software, increase its costs, restrict its ability to serve customers in certain jurisdictions or to enter into partnerships, and adversely affect its ability to raise capital from U.S. investors.

Management's message to shareholders centers on a completed strategic pivot: the company discontinued its biotechnology operations, sold Advanced Biomed HK together with Shanghai Sglcell for US$23,000 and Advanced Biomed Taiwan for US$490,000 , and redirected its resources entirely toward the development and commercialization of FinTruth, which is now its sole product. Management frames the opportunity around the wide use of large language models and AI agents by financial institutions in credit analysis, investment research, risk assessment and compliance review, the risk that AI output contains fabricated or miscalculated figures and other errors, increasing regulatory focus on the explainability, accountability, and accuracy of AI used by regulated financial institutions, and audit and internal-control requirements for documented review and audit trails. The company generated approximately $127,637 (HK$1,000,000) in revenue from its operations related to FinTruth during the fiscal year ended June 30, 2026, and management states that as of the date of the report the company had not generated any customer revenue from FinTruth, with revenue relating to FinTruth derived solely from agency service fees. Management's forward-looking objectives under its twelve-month business and marketing plan running from July 1, 2026 to June 30, 2027 include plans to commercially deploy FinTruth to twenty institutional customers , conversion of ten pilot engagements into commercial contracts , growth from twelve to approximately eighteen full-time employees , and maintaining sufficient liquidity to fund planned operations, with the caveat that there can be no assurance these objectives will be achieved within the anticipated timeframe or at all. The three strategic priorities management has chosen to emphasize are, first, commercial deployment and conversion of the pilot and proof-of-concept pipeline, with between five and ten pilot or proof-of-concept engagements underway or under discussion and Stage II of the plan set to place FinTruth in commercial use by paying customers including commercial banks and financial IT and fintech companies; second, product expansion and platform enhancement, including possible extension of FinTruth's functionality with additional verification modules, additional financial-report verification functions, increased compatibility with additional financial documents and AI-output verification applications, and improved automated customer onboarding and integration capabilities; and third, expansion of distribution through strategic technology partners, building on the recent agreement with Quantum Capital (Hong Kong) Ltd as marketing partner and the potential acquisition of a legal AI agent for U.S. capital markets disclosure documents, for which no binding or definitive agreement has been entered into.

For the fiscal year ended June 30, 2026, the company generated approximately $127,637 (HK$1,000,000) in revenue from its operations related to FinTruth, with no comparable prior-year FinTruth revenue disclosed, reflecting the fact that the company acquired Acellent in April 2026 and had not generated any customer revenue from FinTruth as of the date of the report, with revenue relating to FinTruth derived solely from agency service fees. Net cash outflow in operating activities from continuing operations was $0.96 million for the year ended June 30, 2026, compared with $2.3 million for the year ended June 30, 2025, while net cash outflow in operating activities from discontinued operations was $0.99 million for the year ended June 30, 2026, compared with $3.5 million for the year ended June 30, 2025. The company had net current assets of $12.3 million as of June 30, 2026. The company's balance sheet reflects the ELOC Agreement with HELENA GLOBAL INVESTMENT OPPORTUNITIES I LTD. for up to Twenty-Five Million United States Dollars ($25,000,000) of Common Stock, under which 1,650,710 shares (82,536 shares as adjusted for the Reverse Stock Split) valued at $500,000 were issued as commitment fee shares on January 30, 2026, and which was terminated effective July 21, 2026 with no advances made. The January 2026 private placement of 4,000,000 shares (200,000 shares as adjusted for the Reverse Stock Split) at $0.062 per share (pre-Reverse Stock Split) generated aggregate gross proceeds of $248,000 . The April 13, 2026 loan agreement with Jie Wang provided US$600,000 of unsecured borrowing bearing interest at 10% per annum and maturing six months after disbursement, subject to extension for an additional six months by mutual written agreement. The acquisition of Acellent was effected through the issuance of 270,000 shares of Common Stock valued at an estimated $4.00 per share, or $1,080,000 in the aggregate . The disposition of Advanced Biomed HK and Shanghai Sglcell was effected at a purchase price of US$23,000 , with Advanced Biomed HK acknowledging indebtedness to the company of $6,925,549 to be repaid in full on or before January 26, 2027, and the disposition of Advanced Biomed Taiwan was agreed at an aggregate purchase price of US$490,000 . As of October 13, 2026, the company had 1,652,133 shares of outstanding Common Stock , par value $0.001 per share , and the aggregate market value of shares held by non-affiliates as of December 31, 2025 was approximately $2.7 million . The company's workforce stood at 12 employees as of the date of the report, with 2 in Executive & Product Leadership, 3 in AI Research & Algorithm Engineering, 2 in Product & Software Engineering, 1 in Data & Knowledge Engineering, 1 in Client Solutions & Commercialization, 2 in Finance & Accounting, and 1 in Legal & Corporate Operations.

The most material risks to Advanced Biomed Inc. are concentrated in its recent, unproven business transition and its dependence on a single early-stage product. The company discontinued its biotechnology operations and redirected resources toward FinTruth, and as of the date of the report it had not generated any customer revenue from FinTruth, with revenue relating to FinTruth derived solely from agency service fees, and approximately $127,637 (HK$1,000,000) of revenue from FinTruth operations in the fiscal year ended June 30, 2026. The company has recorded negative operating cash flows historically, with net cash outflow in operating activities from continuing operations of $0.96 million and $2.3 million for the years ended June 30, 2026 and 2025, respectively, and net cash outflow in operating activities from discontinued operations of $0.99 million and $3.5 million for the same periods, and it had net current assets of $12.3 million as of June 30, 2026, with no guarantee it will remain in a net current assets position. The company will need substantial additional capital, and although it has received term sheets and engagement letters relating to potential financing transactions, none of these transactions may be completed. The company depends on third-party cloud providers, models and distribution partners, does not own any data centers, and runs the SaaS version of FinTruth on rented infrastructure from multiple international public cloud providers, so interruptions, capacity constraints, price increases or changes in terms by cloud providers, or termination or underperformance by distribution and marketing partners, could disrupt services and impair growth. The company's AI research and algorithm engineering team consists of three full-time employees within a total workforce of 12 employees , and it intends to grow to approximately eighteen full-time employees , so the loss of any key personnel or inability to hire qualified personnel could delay development of Smith RM and FinTruth. The company does not currently hold ISO/IEC 27001, SOC 2 or other security certifications or attestations that some large financial institutions require of their vendors, and it does not currently maintain any commercial insurance, leaving it susceptible to losses including property damage, accidents, or liabilities.