Airship AI Holdings, Inc. (AISP)

ITEM 2. MANAGEMENT’S DISCUSSION AND ANALYSIS OF FINANCIAL CONDITION AND RESULTS OF OPERATIONS.

The following discussion and analysis summarizes the significant factors affecting our operating results, financial condition, liquidity and cash flows as of and for the periods presented below. The following discussion and analysis should be read in conjunction with our financial statements and the related notes thereto included elsewhere in this report. The discussion contains forward-looking statements that are based on the beliefs of management, as well as assumptions made by, and information currently available to, management. Actual results could differ materially from those discussed in or implied by forward-looking statements as a result of various factors, including those discussed below and elsewhere in this report, particularly in the sections titled “Risk Factors” and “Special Note Regarding Forward-Looking Statements.”

Recent Developments

Backlog and Pipeline

Backlog as of August 6, 2026 were approximately $6.9 million, representing firm fixed price contracts awarded in the first and second quarters of 2026 that are expected to be shipped and primarily invoiced in the third quarter of 2026.

Separately, our total validated pipeline as of August 6, 2026 was $206.1 million, consisting of single and multi-year opportunities for AI-driven edge, video, and sensor and data management platform across our customer verticals. Our pipeline includes opportunities at varying stages of progression with expected award timeframes over the next 18-24 months. Backlog reflects awards already executed, whereas pipeline reflects identified and qualified opportunities that have not yet resulted in awards. We can give no assurance as to whether, when, or in what amount pipeline opportunities will convert into backlog or revenue.

Federal Funding Environment

A substantial majority of our revenue is derived from U.S. federal law enforcement and homeland security customers. During 2025, uncertainty surrounding federal budget priorities and the pace of government spending, including federal initiatives to reform government processes and reduce expenditures contributed to variability in the timing of government awards and procurement activity, which affected our government business. More recently, we have observed an improving federal funding environment for the mission areas our platform supports, including border security and public safety, which we believe is reflected in the growth and stage progression of our validated pipeline. The timing and amount of government awards remain subject to factors outside our control, including appropriations, shifting agency priorities and procurement cycles, and we can provide no assurance regarding the conversion of any particular opportunity.

Strategic Initiatives

We are pursuing a deliberate expansion of our commercial business to complement our established federal law enforcement and homeland security customer base. While a substantial majority of our revenue is currently derived from government customers, we believe our edge AI and data management platform is directly applicable to commercial organizations with large-scale sensor, security and data-management requirements, and we are investing to broaden our presence across commercial verticals. The strategic initiatives described below are central to this expansion.

Agentic AI (Ask Airship)

We are developing Ask Airship, an agentic AI capability designed to allow users to query, analyze and act on structured data across our platform using natural language, which we intend to offer as a standalone product complementary to our Outpost AI, Acropolis, Airship Command and Fortress Server offerings. We intend to make Ask Airship available on a software-as-a-service (subscription) basis as well as on a pay-per-use, consumption-based basis. For deployments in federal and other security-sensitive environments, we expect on-premises inference to rely on Western open-weight models to satisfy applicable procurement and model-origin requirements.

Robotics

We are extending our edge AI and data management platform to ingest, structure and act on data from autonomous and robotic sensor platforms, with a particular focus on commercial customers as well as government customers, consistent with our broader commercial expansion. We are evaluating commercial models for these capabilities, including subscription and as-a-service arrangements. We can give no assurance regarding the timing, cost, market acceptance or financial contribution of this initiative, which is subject to development, integration, supply-chain and regulatory risks.

Overview

We are an AI-driven data management platform that solves complex data challenges for large institutions operating in dynamic and mission-critical environments characterized by rapidly increasing volumes of data ingested from a rapidly growing number of data sources. We address these challenges by structuring “dark” or unstructured data at the edge, the location at which the data is generated and collected using purpose-built AI models, enabling real-time decision-making rather than transmitting raw data to a central location for processing.

We specialize in ingesting metadata from edge-based sensors used by government and law enforcement agencies worldwide, including video, audio, telemetry, acoustic, seismic and autonomous devices, as well as by large commercial organizations with similar requirements. Our trained AI models detect and extract identifying characteristics of objects within a video frame (for example, a vehicle’s license plate characters and its make, model and color), a process we refer to as “structuring” the data, and allow customers to act on that data both in real time and through historical search.

Our primary offerings include Outpost AI (edge structuring and analysis), Acropolis (enterprise management software backbone, deployable on-premises or in cloud/multi-cloud architectures), Airship Command (visualization across workstation, web and mobile) and Fortress Server (our on-premises data center server and storage platform). These product offerings allow customers to manage their data across the full data lifecycle through a secure, permission-based, single-pane-of-glass architecture.

Our typical customer engagement is a multi-year agreement that includes our core offerings together with professional services, technical support and software maintenance, which we expect to result in predictable, long-term recurring revenue. From inception until the Merger in December 2023, we operated as a 100% employee-owned, bootstrapped company with no outside investment. As a U.S.-based company, we operate at the intersection of public safety and AI, in which we address a combined market we estimate at more than $70 billion in 2026 and growing, spanning edge AI hardware and software (approximately $30 billion, according to Grand View Research, 2026, service and professional robotics, including robotics-as-a-service (approximately $31 billion according to Fortune Business Insights, 2026), and agentic AI (approximately $10 billion according to  multiple industry analysts, 2026).

Fair Value Transactions in Connection with Merger

As a result of the Merger, we entered into the following transactions that were measured at fair value and vary quarterly with the share price and other items. Any change is non-cash and is recorded as a gain or loss in other income (expense). See Note 12– Fair Value Measurements for more information.

Private Placement and Public Warrants in Connection with Merger

At the Merger closing, we assumed 515,000 private placement warrants and 16,184,612 public warrants. On June 3, 2024, we reduced the exercise price of such warrants from $11.50 per share to an exercise price of $7.80 per share. On November 20, 2024, we further reduced the exercise price of the outstanding public warrants and private warrants to an exercise price of $4.50 per share. The purpose of this reduced exercise price was to potentially raise proceeds received from the exercise of such warrants, if any, for working capital and general corporate purposes. As of June 30, 2026, there were 515,000 private placement warrants and 16,145,006 public warrants outstanding.

Segment Reporting

The Financial Accounting Standards Board, or FASB, Accounting Standard Codification, or ASC, Topic 280, Segment Reporting, requires that an enterprise report selected information about reportable segments in its financial reports issued to its stockholders. Operating segments are defined as components of an entity for which separate financial information is available and that is regularly reviewed by the Chief Operating Decision Maker (“CODM”) in deciding how to allocate resources to an individual segment and in assessing performance. The Chief Executive Officer, Chief Financial Officer and President are the Company’s CODM. The CODM monitors the revenue and expense components of the various products and services we offer, but operations are managed and financial performance is evaluated on a corporation-wide basis in comparison to a business plan which is developed each year. Accordingly, all operations are considered by the CODM to be one operating segment and one reportable segment as contained in the Consolidated Statements of Operations and Comprehensive Loss to the consolidated financial statements. The CODM uses consolidated net income (loss) as its required measure of segment profit/loss, as such measure is determined in accordance with the measurement principles most consistent with the consolidated financial statements.

Results of Operations

The following table sets forth key components of our results of operations during the three months ended June 30, 2026 and 2025.

(dollars in thousands)

[Excerpt truncated for page length; source filing is linked above.]

ITEM 7. MANAGEMENT’S DISCUSSION AND ANALYSIS OF FINANCIAL CONDITION AND RESULTS OF OPERATIONS.

The following discussion and analysis summarizes the significant factors affecting our operating results, financial condition, liquidity and cash flows as of and for the periods presented below. The following discussion and analysis should be read in conjunction with our financial statements and the related notes thereto included elsewhere in this report. The discussion contains forward-looking statements that are based on the beliefs of management, as well as assumptions made by, and information currently available to, management. Actual results could differ materially from those discussed in or implied by forward-looking statements as a result of various factors, including those discussed below and elsewhere in this report, particularly in the sections titled “Risk Factors” and “Special Note Regarding Forward-Looking Statements.”

Overview

We are a robust AI-driven data management platform that solves complex data challenges for large institutions operating in dynamic and mission-critical environments with rapidly increasing volumes of data being ingested from a similarly rapidly growing number of data sources.

We solve these challenges by structuring “dark” or unstructured data at the edge, the location at which the data is generated and collected, and leveraging purpose-built AI models. Unstructured, or “dark” data, which is typically categorized as qualitative data, cannot be processed and analyzed via conventional data tools and methods. Conversely, structured data, typically categorized as quantitative data, is highly organized and easily decipherable by machine learning algorithms.

Structuring and then analyzing data using AI models at the edge, versus transmitting the data from the edge back to a central processing location for structuring and analysis, enables real-time decision making and data-driven operational efficiency.

We specialize in ingesting all available metadata from edge-based sensors used by government and law enforcement agencies around the world, including surveillance cameras (video), audio, telemetry, acoustic, seismic, and autonomous devices, along with large commercial corporations with fundamentally similar capabilities and requirements.

Data generated by these edge-based sensors, including video, can then be run through our trained AI models to detect objects present within the video frame. Once an object is detected, for example an automobile, additional identifying characteristics of the object can be extracted from the image including the license plate characters and the make, model, and color of the automobile. This process of analyzing, logging and categorizing ingested data is referred to as “structuring” the data.

Airship AI’s software allows customers to view structured data both in real-time as well as to conduct searches on the structured data at a later point in time. Real-time structured data use includes, for example, alarms on a specific license plate or a specific make, model or color of automobile. Non-real-time structured data use includes, for example, searching a database of video data that has been previously ingested and stored to find instances of a particular license plate being visible, along with other logged vehicle characteristics such as make, model and color of an automobile.

Additional edge deployed AI models enable similar object detection and recognition of common and custom trained objects, such as an aircraft, boat, person, animal, bag, or weapon. Airship AI’s models provide similar data points for these object types allowing analysts the ability to be notified in real-time of the detection of a specified object and similarly search for historically detected objects. Examples include detecting aircrafts and boats along with their respective tail numbers and hull registration numbers.

Our AI modelling process starts with pre-trained AI models from our technology ecosystem partners which we then customize using proprietary datasets tailored towards our customers’ unique workflow requirements. Where customers have pre-existing AI models or engines, we integrate those models or engines into our edge platform allowing customers to leverage proprietary models within the Airship AI software ecosystem.

Our primary offerings include Outpost AI, Acropolis, and Airship Command. Our offerings allow customers to manage their data across the full data lifecycle, when and where they need it, using a highly secure permissioned based architecture.

Recent Developments

Expansion into Robotics and Autonomous Systems

We are pursuing the extension of our edge AI platform to support robotic and autonomous system deployments and expect to conduct pilot programs during 2026. We believe the integration of advanced computer vision, sensor fusion, and real-time edge analytics with mobile and semi-autonomous platforms represents a natural extension of our existing software capabilities.

Our robotics-related initiatives are focused on enabling our AI software to operate on, and integrate with, ground-based or mobile robotic platforms for applications such as security, inspection, monitoring, and situational awareness. These platforms are expected to leverage our existing AI models, including object detection, behavior analysis, and anomaly detection, deployed at the edge to support real-time decision-making in dynamic environments.

During 2026, we plan to conduct limited pilot deployments with selected customers and partners to evaluate technical performance, operational integration, and market demand. These pilot programs are not expected to generate material revenue. Any future commercialization of robotics-enabled offerings would be expected to complement our existing software-centric business model. There can be no assurance that these pilot efforts will result in commercially viable products or services.

Backlog and Pipeline

We also began to see movement during the fourth quarter of 2025 and into the first quarter of 2026 on several large projects in the federal and commercial marketplace, including several notable awards:

Our backlog as of December 31, 2025 was $3.3 million.

Our total validated pipeline as of December 31, 2025 was $173.4 million, consisting of single and multi-year opportunities for AI-driven edge, video, and sensor and data management platform across our customer verticals. Our pipeline includes opportunities at varying stages of progression with expected award timeframes over the next 18-24 months.

Warrant Exercise

On October 8, 2025, we entered into a warrant exercise inducement offer letter with the holder of existing common stock warrants exercisable for an aggregate of 2,162,162 shares of common stock to exercise such warrants at the existing exercise price of $4.50 per share, in exchange for our agreement to issue new common stock warrants to purchase 2,702,702 shares of common stock at an exercise price per share of $6.20. The aggregate gross proceeds received from the exercise of the existing warrants were approximately $9,729,729, before deducting financial advisory fees. We intend to use the net proceeds from the exercise of the existing warrants for working capital and general corporate purposes.

The shares of common stock issuable upon exercise of the existing warrants are registered for issuance pursuant to a registration statement on Form S-3 (File No. 333-284462), which was declared effective by the SEC on January 31, 2025.

In consideration for the immediate exercise of the existing warrants for cash, the holder received the inducement warrants to purchase 2,702,702 shares of common stock in a private placement pursuant to Section 4(a)(2) of the Securities Act. The inducement warrants have an exercise price of $6.20 per share, are immediately exercisable and will be exercisable for five and one-half years from the date of issuance.

The inducement warrants and the shares of common stock underlying the inducement warrants offered in the private placement have not been registered under the Securities Act or applicable state securities laws. Accordingly, the securities may not be offered or sold in the United States except pursuant to an effective registration statement or an applicable exemption from the registration requirements of the Securities Act and such applicable state securities laws. As part of the transaction, we have filed a resale registration statement on Form S-3 with the SEC to register the resale of the shares of common stock underlying the inducement warrants, which registration statement went effective on November 16, 2025.

Key Performance Indicators

Historically, a majority of our product revenue has consisted primarily of a bundled hardware and software product and to date we have sold or licensed a minimal amount of standalone software. In the future, we expect to see more delivery of our products using a cloud-based software solution which will allow us to create additional subscription revenue.

We have historically evaluated our business solely based on revenue generated from customers and we have not tracked any other customer-related metrics. As we grow and increase our product offerings and customer base, we intend to modify and develop more advanced performance indicators. We believe the following key performance indicators apply to us in the future:

[Excerpt truncated for page length; the complete text is on the linked full-MD&A page.]