Odysight.ai Inc. (ODYS) Company Overview

What does Odysight.ai do?

Odysight.ai Inc. is a small, development-stage visual-sensing company listed on Nasdaq and the Tel Aviv Stock Exchange under ODYS. It designs miniature video sensors, embedded software, processing units, and artificial-intelligence algorithms that let operators see and analyze safety-critical components while machinery is running. The practical objective is predictive maintenance and condition-based monitoring: detect wear, misalignment, loosened fasteners, leaks, foreign objects, or other anomalies before they become failures.

Where is the technology used?

The company targets aerospace, defense, transportation, energy, heavy equipment, and other industrial settings where conventional inspection is difficult or where downtime is expensive. Its systems have been associated with NASA, the Israeli Air Force, Israel’s Ministry of Defense, Israel Railways, helicopter programs, unmanned aerial vehicles, elevator monitoring, and mining-truck demonstrations. The official company website presents the platform as a bridge between physical visibility and machine-learning-based maintenance decisions.

Odysight.ai matters less because of its current revenue scale than because it is trying to establish a new maintenance category: continuous visual sensing inside assets that previously relied on periodic manual inspection, vibration sensors, temperature readings, or post-failure diagnosis.

How does Odysight.ai make money?

The present model begins with engineering engagement, proof-of-concept work, pilot deployment, and an initial system sale. A typical kit may include miniature cameras, illumination, a processing unit, embedded software, communication capability, and application-specific algorithms. Revenue can arise from product shipments, integration and development services, project milestones, and support. Management also wants the installed base to create recurring software, analytics, maintenance, and data-service revenue over time, but recurring revenue is not yet the dominant reported stream.

Which revenue stream matters most now?

The answer is transitional. In 2025, total revenue was $3.015 million, including $1.7 million from derecognition of a contract liability tied to a former Fortune 500 medical customer and about $1.2 million from vision-based predictive-maintenance and condition-monitoring platforms. The core platform therefore remained small, although management reported that platform revenue increased about 23% year over year. The 2025 annual results and strategy update are available through the company’s full-year 2025 release.

Why is backlog central to the model?

Backlog represents booked purchase orders or hard commitments not yet recognized as revenue. It was $13.8 million at December 31, 2025 and $14.0 million at March 31, 2026. That is large relative to 2025 revenue, but it is not equivalent to guaranteed sales or profit. Programs can be delayed, rescheduled, reduced, or canceled, and aerospace and defense projects may require years of testing, certification, procurement, and platform integration. For Odysight.ai, conversion speed matters as much as backlog size.

What do the latest results show?

The quarter ended March 31, 2026 illustrates both the opportunity and the financial fragility. Revenue was only $82,000, down from $2.065 million in the prior-year quarter. The comparison is distorted because first-quarter 2025 included $1.7 million of contract-liability derecognition from the legacy medical customer. Even after adjusting for that effect, management cited geopolitical delays affecting orders and deliveries in predictive-maintenance projects.

What changed after the quarter?

Operationally, the company reported the completion of first U.S. test flights of its visual-sensing system on a UH-60 Black Hawk helicopter in June 2026 and a proof-of-concept purchase order from Honeywell Aerospace’s auxiliary-power-unit division later that month. These are strategically relevant validation events, but they should not be confused with scaled production revenue. The official investor-relations page provides the current reporting package and subsequent business updates.

How did Odysight.ai evolve from medical imaging to predictive maintenance?

The company’s strategic history explains its unusual technology base. It did not start as a conventional industrial-software vendor. Its present capabilities came from miniaturized imaging developed for medical applications, where small form factor, image quality, illumination, reliability, and operation in constrained spaces were already critical.

1. 2013
   The Nevada corporation was formed as Intellisense Solutions, creating the public-company shell that later housed the operating business.
2. 2019
   The company acquired ScoutCam’s miniaturized imaging operations and adopted that activity as its primary business.
3. 2020–2022
   Medical imaging remained the principal commercial base while industrial and aerospace use cases were developed.
4. 2023
   The company changed its name to Odysight.ai, signaling a shift from camera hardware toward visual analytics and predictive maintenance.
5. 2024
   A defense-focused Israeli subsidiary was formed and the company deepened aviation, rail, and industrial programs.
6. 2025
   ODYS uplisted to Nasdaq, raised capital, opened an Italian subsidiary, and moved decisively away from the legacy medical customer.
7. 2026
   The shares began trading on the TASE, U.S. Black Hawk flight testing was completed, and the Honeywell APU proof-of-concept broadened aerospace validation.

What did the strategic pivot change?

The pivot increased the addressable opportunity but also lengthened the path to revenue. Medical-device supply relationships can be concentrated, but once qualified they may produce repeat orders. Aerospace and defense programs offer potentially larger lifetime value, yet they involve certification, export controls, platform-specific engineering, customer funding cycles, and demanding reliability requirements. The company exchanged one kind of concentration risk for a portfolio of long-cycle execution risks.

What gives Odysight.ai a competitive advantage?

The potential moat is a combination of packaging, domain engineering, accumulated image data, algorithms, and customer qualification. A sensor that survives heat, vibration, pressure, contamination, electromagnetic interference, and space constraints is more than a camera. It must be integrated into a platform without compromising safety, weight, power consumption, maintenance procedures, or certification.

Who are the main competitors?

Competition is broader than a list of direct camera vendors. Odysight.ai competes with vibration, acoustic, thermal, pressure, oil-debris, and other sensor technologies; with machine-vision companies; with industrial analytics platforms; with aerospace suppliers that can integrate their own monitoring systems; and with the status quo of scheduled inspection. Large industrial groups such as Honeywell, RTX, Safran, Siemens, Emerson, and SKF possess deeper customer relationships and resources in adjacent condition-monitoring markets, although they do not necessarily offer the same miniature in-situ visual architecture.

How financially strong is Odysight.ai?

Liquidity is currently adequate for near-term operations, but the company is not self-funding. At March 31, 2026, cash was $21.763 million, current assets were $23.282 million, current liabilities were $2.984 million, and total liabilities were $3.478 million. There was no financial debt. That produces a strong current ratio of roughly 7.8 times, but the ratio is less informative than the burn rate because revenue is still small.

What does the annual baseline reveal?

The audited figures appear in the 2025 Form 10-K. The cost structure shows a technology company investing far ahead of revenue. R&D alone was more than three times annual sales in 2025, while sales and marketing nearly doubled. That spending can be rational if programs convert into fleet-level orders, but it creates a narrow margin for execution error.

Who owns Odysight.ai stock, and why does control matter?

Odysight.ai has a single class of common stock, but ownership is concentrated. The 2025 annual report ownership disclosure measured beneficial ownership around March 18, 2026. Moshe “Mori” Arkin was the dominant holder, with 7.944 million beneficially owned shares and 41.83% of the class on the filing’s calculation basis. Directors and officers as a group held 9.761 million shares, or 47.43%.

How should investors interpret the ownership structure?

Concentrated ownership can support patient investment through long certification cycles, but it also reduces the influence of smaller shareholders and may contribute to a limited trading float. The company had 16.773 million shares outstanding at March 31, 2026, up from 12.613 million at the start of 2025. That increase reflects the 2025 public offering and subsequent option and warrant activity. For a loss-making company, future dilution is a core capital-allocation variable rather than a peripheral concern.

Which KPIs best explain Odysight.ai’s progress?

Conventional revenue growth is too noisy to evaluate this business in isolation. The better dashboard combines commercial conversion, deployment depth, margin, cash consumption, and balance-sheet capacity.

How should backlog be interpreted?

Researchers should also track the number of customer platforms moving through the funnel: concept, engineering contract, ground test, flight or field test, certification, initial production, and fleet expansion. Odysight.ai does not yet disclose a standardized count for each stage, so qualitative program milestones remain important but must be interpreted conservatively.

What opportunities could change the company’s scale?

The largest opportunity is to turn one-off engineering work into repeatable modules that can be installed across entire fleets or asset classes. Aerospace auxiliary power units, helicopter drivetrains, UAV components, rail undercarriages, elevator belts, and heavy mining equipment are attractive because failures are costly and visual confirmation can complement existing sensor data.

Why could partnerships matter more than direct selling?

A company with 52 employees cannot independently cover every global aerospace and industrial account. Partnerships with original-equipment manufacturers, maintenance providers, defense contractors, and local distributors can shorten market access and embed Odysight.ai inside established procurement channels. The trade-off is lower control over timing, pricing, customer ownership, and intellectual-property boundaries.

What risks could weaken Odysight.ai’s outlook?

The company’s risk profile is dominated by commercialization, financing, concentration, and execution rather than by debt. Its official Q1 2026 Form 10-Q emphasizes long sales cycles, dependence on a limited number of customers, uncertain backlog conversion, reliance on suppliers and partners, product-quality demands, export controls, cybersecurity, and geopolitical exposure.

Which risk is most immediate?

Cash consumption is the most measurable near-term risk. The company used $13.703 million in operations during 2025 and another $4.254 million in the first quarter of 2026. Management stated that existing resources should fund at least 12 months from the filing date, but also acknowledged that additional funding will be needed until profitability. Because revenue recognition can be lumpy, financing may occur before investors see a smooth commercial ramp.

Why does customer concentration matter?

Two customers represented $2.783 million of the company’s $3.015 million FY2025 revenue. Customer A contributed $1.856 million and Customer B contributed $0.927 million. This means a delayed milestone, renegotiated program, or completed legacy contract can materially change year-over-year results. Diversification across aerospace, defense, rail, elevators, and heavy equipment is strategically important, but it is not yet visible in a stable revenue mix.

Why does Odysight.ai matter for valuation?

A conventional DCF based on near-term earnings is difficult because current revenue is small, losses are large, and commercialization timing is uncertain. The model must instead be built from operating assumptions: number of platforms won, units per platform, hardware price, engineering revenue, recurring software or analytics per installed unit, gross margin at scale, R&D and sales intensity, and the probability that backlog converts.

What should students and investors watch next?

• Quarterly conversion of the $14.0 million backlog into recognized revenue.
• Whether Black Hawk, Honeywell APU, UAV, rail, and elevator programs progress from tests to production.
• Core platform revenue growth after the legacy medical contract has rolled off.
• Gross-margin movement as product mix and manufacturing volume change.
• Operating cash use relative to the $21.763 million March 2026 cash balance.
• New equity issuance, warrant exercise, and the resulting fully diluted share count.
• Customer concentration and the addition of repeat customers in multiple industries.
• Evidence that software, analytics, or service revenue becomes recurring.

What is the key takeaway from Odysight.ai analysis?

Odysight.ai is best understood as a commercialization-stage industrial technology company, not as a mature software or aerospace supplier. Its core proposition is credible and differentiated: put rugged miniature visual sensors inside hard-to-reach machinery, combine the images with embedded analytics, and replace periodic inspection with continuous condition awareness. Customer programs, aerospace demonstrations, a substantial backlog relative to revenue, and a broad patent portfolio provide evidence that the technology addresses real operating problems.

For a student or researcher, Odysight.ai is a useful case study in strategic pivoting, technology transfer, and the tension between customization and scale. For an investor, the analytical task is to connect backlog and technical milestones to realistic revenue timing, margins, cash burn, and dilution. The company may have a valuable sensing architecture, but its valuation ultimately depends on whether that architecture becomes a standardized, repeatable maintenance platform across large fleets and industrial asset bases.