Razor Labs launches DataMind AI 5.0 as mining AI moves from failure alerts to maintenance execution

Razor Labs Ltd. (TASE: RZR) has launched DataMind AI 5.0, expanding its mining predictive-maintenance platform from early failure detection into the wider process of investigating faults, coordinating repairs and verifying that equipment has safely returned to operation. The new version combines artificial intelligence diagnostics, engineering analysis, maintenance collaboration and fault-lifecycle management across mobile mining fleets and fixed processing assets. The strategic change is significant because mines frequently possess sensor data and equipment alerts but still depend on fragmented human processes to decide what should be repaired, when the work should occur and whether the intervention solved the underlying problem. Razor Labs enters this product cycle after doubling revenue in 2025, improving gross margins and expanding deployments across Africa, Australia and Latin America. The central tension is whether DataMind AI 5.0 can convert technical progress into larger recurring subscriptions and operating leverage before continued losses and cash consumption constrain the company’s expansion.

How does DataMind AI 5.0 extend Razor Labs beyond conventional predictive-maintenance alerts?

Predictive maintenance traditionally attempts to identify changes in vibration, temperature, pressure, electrical current, oil condition or other equipment signals before a machine fails. The output is usually an alert showing that an abnormal condition may be developing.

That warning is valuable, but it does not complete the maintenance process. Engineers must determine whether the signal represents a genuine fault, identify the affected component, assess its severity, check the operating history, decide whether production should continue and create an appropriate maintenance task.

DataMind AI 5.0 is designed to connect those stages in one environment. Razor Labs said the release brings together artificial intelligence diagnostics, engineering investigation, collaboration tools and fault-lifecycle management. The platform is intended to follow an issue from its first detection through investigation, repair and final verification. Version 5.0 became available to customers worldwide on July 28, 2026.

The product therefore moves closer to becoming a maintenance operating system rather than remaining a specialist condition-monitoring tool. A mining company may already receive alerts from original equipment manufacturers, sensors, control systems and fleet-management platforms. The practical difficulty is deciding which signals matter and ensuring that an actionable fault does not disappear inside another dashboard.

Razor Labs is attempting to reduce that gap by placing the diagnosis, supporting evidence, maintenance discussion and resolution status around the same equipment issue. This could increase the platform’s value to reliability teams because customers are being offered a workflow through which action is managed, not simply another source of warnings.

The commercial implication is equally important. Software embedded in maintenance decisions, work orders and performance verification may become more difficult to replace than a standalone analytics application. Deeper workflow integration can support higher retention, expansion across additional assets and more predictable subscription revenue.

Why is the gap between detecting a fault and completing a repair costly for mining companies?

Mining equipment operates under heavy loads, abrasive conditions, heat, dust and continuous production pressure. A fault involving a haul truck, crusher, conveyor, mill, pump or loading system can affect more than the individual machine.

A stopped conveyor may restrict the flow of material through a processing plant. A failed haul truck can reduce fleet availability and increase pressure on the remaining vehicles. A crusher shutdown may interrupt downstream operations even when the affected component represents only a small part of the wider production system.

Maintenance teams consequently face a difficult balance. Shutting equipment down too early can reduce production unnecessarily. Delaying an intervention may allow a manageable defect to become a more expensive failure.

Razor Labs has published several case studies illustrating the economic argument behind its platform. During the second quarter of 2026, the company said DataMind AI identified a transmission-return-circuit problem on a GHH SLP5 load-haul-dump vehicle, an aftercooler issue on a Komatsu 930E haul truck and brake overheating on a Caterpillar 793D haul truck. Razor Labs attributed avoided downtime and estimated savings of about $20,000, $75,000 and $25,000 respectively to those detections. These figures are company-reported case-study outcomes rather than independently audited financial results, but they show the type of return customers are expected to evaluate.

Earlier deployments also included the detection of structural looseness on an apron-feeder motor, which Razor Labs said prevented five hours of downtime and approximately $180,000 in losses. Another case involving a conveyor motor was reported to have prevented about ten hours of downtime and $191,000 in estimated production losses after the software identified bearing damage associated with an electrical grounding problem.

DataMind AI 5.0 is intended to improve the probability that such detections produce timely action. An accurate alert creates limited value when responsibility is unclear, supporting evidence is difficult to retrieve or the issue is not converted into a maintenance task.

The next competitive stage in predictive maintenance is therefore likely to focus less on generating more alerts and more on shortening the time between detection and resolution. Vendors that can connect analysis with maintenance execution may capture a larger share of mining technology budgets than those providing isolated monitoring tools.

Can Razor Labs turn its sensor-fusion technology into a wider mining workflow platform?

Razor Labs bases DataMind AI on what it calls AI Sensor Fusion, combining information from different equipment and condition-monitoring sources to identify developing faults and their possible causes.

The platform can integrate operational data, condition-monitoring information, maintenance records and sensor inputs. Razor Labs lists compatibility with systems including SAP, asset-management applications, business-intelligence tools, cloud platforms, oil-analysis data, fleet telemetry and industrial systems using protocols such as OPC-UA and MODBUS.

This integration capability matters because mines rarely operate with one standard technology environment. A large mining group may use equipment from several manufacturers, different fleet-management systems, separate plant-control platforms and maintenance software introduced over many years.

Original equipment manufacturers can provide detailed information about their own machines, but mine operators need an asset-health view across mixed fleets and fixed infrastructure. An independent platform can potentially combine information from trucks, crushers, conveyors, mills and pumps without requiring the customer to remain within one equipment ecosystem.

The value of sensor fusion depends on diagnosis quality rather than data volume alone. Combining several weak or poorly calibrated signals can generate more noise. Razor Labs must show that its models consistently distinguish genuine equipment deterioration from operating variations caused by load, speed, environment or production schedules.

Version 5.0 could strengthen that proposition by retaining the engineering context around each detected problem. Investigation notes, evidence, repair decisions and verification outcomes can create a feedback loop that improves future diagnosis.

This also gives Razor Labs an opportunity to build a proprietary industrial dataset. Each confirmed fault and completed repair can help the company understand how particular equipment types behave under different operating conditions. Over time, that accumulated information may become more difficult for a new competitor to reproduce.

How does Razor Labs make money from DataMind AI deployments and customer expansion?

Razor Labs operates as an industrial software and analytics provider rather than a mining-equipment manufacturer. Revenue is generated through DataMind AI licences, subscriptions, deployment services, sensor-related implementation and ongoing software or technical support.

The company’s published contractual terms refer to setup fees and yearly licence fees, indicating a model that can combine initial deployment revenue with recurring software payments. Those terms also make clear that implementation timing depends partly on customer-team availability and the data supplied by the mine operator.

A typical commercial journey can begin with a pilot covering a selected group of critical assets. The customer then evaluates detection accuracy, operational savings and integration before deciding whether to sign a broader subscription.

Razor Labs followed this route with SIMEC Mining in South Australia. After an initial pilot, the companies entered a subscription agreement in February 2025 to continue deploying DataMind AI across critical assets. SIMEC Mining indicated that the system complemented its existing asset-management, downtime-detection and condition-monitoring platforms.

The key growth opportunity is expansion after the first successful deployment. A mine may initially install DataMind AI on a small number of pumps, conveyors or haul trucks. If the platform demonstrates value, the contract can potentially expand across additional equipment, processing areas and sites.

DataMind AI 5.0 may support that expansion by giving maintenance teams a more complete operational workflow. Customers are more likely to extend a system across their asset base when it improves daily decision-making rather than functioning as an occasional diagnostic tool.

However, large mining deployments can require local engineers, hardware installation, integration and customer-specific configuration. Razor Labs must increase recurring revenue faster than delivery costs if growth is to produce sustainable operating leverage.

What do Razor Labs’ 2025 financial results reveal about its commercial progress?

Razor Labs generated revenue of ILS 34.62 million in 2025, an increase of approximately 101% from ILS 17.22 million in 2024. Gross profit rose to ILS 15.72 million from ILS 7.15 million, while gross margin improved to 45.4% from 41.5%.

The improvement indicates that Razor Labs is moving beyond the very small revenue base reported earlier in its development. Revenue had been only ILS 5.29 million in 2023, meaning the company expanded sales more than sixfold over two years.

Profitability has not yet followed revenue at the same pace. Razor Labs recorded an operating loss of ILS 15.03 million and a net loss of ILS 16.89 million in 2025. The operating loss narrowed from ILS 17.05 million in 2024, although the net loss increased from ILS 15.48 million.

Operating cash outflow reached ILS 16.9 million, while free cash outflow was approximately ILS 17.07 million. Cash and investments stood at ILS 35.23 million at the end of 2025, compared with total debt of ILS 6.17 million and net cash of around ILS 29.05 million.

These numbers frame the central investor issue. Razor Labs has demonstrated rapid revenue growth, but the business remains loss-making and continues to consume cash. The company must show that customer expansion, subscriptions and software margins can outpace research, deployment and international sales costs.

DataMind AI 5.0 could help if the new workflow functions support higher contract values without requiring an equivalent increase in implementation labour. It could also place additional pressure on spending if customers require extensive support, integration and training to use the broader platform.

The next financial report will need to show whether the 2025 revenue acceleration is continuing and whether gross-profit growth is beginning to cover more of the operating cost base.

Does Razor Labs face customer and geographic concentration despite expanding internationally?

Razor Labs operates across South Africa, the Democratic Republic of Congo, Colombia, Australia, the United States and Israel. Its 2025 revenue nevertheless remained heavily concentrated in a small number of mining regions.

South Africa generated ILS 12.36 million of revenue, Colombia contributed ILS 11.58 million and the Democratic Republic of Congo accounted for ILS 7.25 million. Together, those three markets represented about 90% of total 2025 revenue. Australia contributed ILS 2.21 million, while Israel and other markets made up the remainder.

Concentration can be beneficial during an early growth stage because successful mining deployments often lead to references, neighbouring-site opportunities and relationships with other operators in the same region. It can also create volatility when a major project is delayed or one customer reduces spending.

Razor Labs is using partnerships to extend its reach without building every local implementation capability internally. Process Automation became its official partner and system integrator in Africa in 2025, bringing a regional workforce, local offices and experience supplying instrumentation to mining companies.

The company also reported continued deployment activity across Australia, Africa and Latin America during the second quarter of 2026. DataMind AI was being expanded across mobile fleets and fixed assets, although Razor Labs did not disclose new contract values or the number of revenue-generating sites in its quarterly summary.

The partnership strategy can lower customer-acquisition and deployment costs, but Razor Labs must preserve control over product quality and the customer relationship. Industrial AI systems can lose credibility quickly when sensors are installed incorrectly, alerts are poorly explained or local maintenance teams do not trust the output.

Why has Razor Labs stock rallied in July while remaining lower for 2026?

Razor Labs shares closed at 556.7 agorot, equivalent to ILS 5.567, on July 27, 2026. The official Tel Aviv Stock Exchange data showed a market capitalisation of approximately ILS 219.6 million.

The stock had gained approximately 38.3% during July through July 27, although it remained 4.8% lower for the year. Over the five trading sessions beginning July 20, the share price increased from 524.1 agorot to 556.7 agorot, a rise of about 6.2%.

Delayed market data showed the shares trading at 564.9 agorot on the morning of July 28, up 1.47% from the previous close. The price remained within a reported 52-week range of approximately 296 to 769 agorot.

The July recovery suggests improving attention toward Razor Labs’ commercial progress and exposure to industrial artificial intelligence. However, the stock’s continued negative year-to-date performance indicates that investors are not treating product announcements as sufficient evidence of a durable financial inflection.

At a market value of roughly ILS 220 million compared with 2025 revenue of ILS 34.62 million, Razor Labs was trading at approximately six times trailing revenue. That valuation may appear reasonable under a scenario of continued high growth and improving margins, but it leaves limited room for a material slowdown because the company is still loss-making and cash-flow negative.

The release of DataMind AI 5.0 strengthens the product narrative. A sustained rerating would require financial evidence showing that platform expansion is producing subscription growth, higher gross profit and lower cash consumption relative to revenue.

What execution risks could prevent DataMind AI 5.0 from delivering operating leverage?

The first risk is implementation complexity. Mining sites contain varied equipment, legacy software and operational practices. Integrating DataMind AI into maintenance systems and work-order processes may take longer than installing the underlying sensors.

The second risk is trust. Maintenance teams are responsible for safety and production continuity. They may ignore an artificial intelligence recommendation when the diagnosis is not supported by understandable evidence or when previous alerts produced false positives.

The third risk is measurable attribution. Razor Labs can estimate the production losses avoided after an early detection, but it may be difficult to prove exactly what would have happened without the system. Customers will increasingly demand credible, repeatable return-on-investment evidence before extending contracts across large asset portfolios.

The fourth risk is concentration. Revenue remains dependent on a limited number of markets and potentially a relatively small group of mining customers. One postponed deployment could affect a company with annual revenue of less than ILS 35 million more significantly than it would affect a diversified software provider.

The fifth risk is funding discipline. Razor Labs retains a net-cash position, but the 2025 free cash outflow was material relative to its available cash. Continued investment must produce enough recurring revenue and gross profit to extend financial flexibility.

The sixth risk is competition. Original equipment manufacturers, industrial automation companies, enterprise software providers and specialist condition-monitoring vendors are all adding artificial intelligence capabilities. Razor Labs needs to demonstrate that its cross-equipment sensor fusion and mining-specific workflow produce results that broader platforms cannot match easily.

What measurable proof points will determine whether DataMind AI 5.0 creates shareholder value?

The first proof point will be new subscription agreements. Razor Labs needs to disclose additional customers moving from pilots into commercial deployments and existing customers extending the platform across more equipment or mining sites.

The second will be recurring revenue. A growing proportion of licence and subscription revenue would improve visibility and reduce dependence on one-time implementation income.

The third will be gross-margin progression. DataMind AI 5.0 should eventually allow Razor Labs to support more assets and customers without increasing delivery costs at the same rate.

The fourth will be operating cash flow. Revenue growth will carry greater value when cash consumption begins declining as a percentage of sales.

The fifth will be documented workflow adoption. Customers should use the platform not only to detect faults but also to manage investigations, collaborate on maintenance decisions, close work orders and verify repairs.

DataMind AI 5.0 moves Razor Labs toward the part of predictive maintenance where operational and financial value is actually realised. Finding an abnormal signal is useful, but mining companies ultimately pay for fewer production interruptions, faster repairs and more reliable equipment.

What has improved is the breadth of the product, the company’s revenue base and its ability to demonstrate deployments across several major mining regions. What remains unresolved is whether those capabilities can produce repeatable subscriptions and a path toward positive cash generation.

The strongest confirmation would be another period of rapid revenue growth accompanied by higher gross margins, broader customer adoption and lower operating losses. The thesis would weaken if product releases continue to accelerate while contract disclosures, recurring revenue and cash-flow improvement remain limited.

What are the key takeaways from Razor Labs’ DataMind AI 5.0 mining technology launch?

• Razor Labs launched DataMind AI 5.0 globally on July 28, 2026.
• The release extends predictive maintenance from fault detection into investigation, collaboration, repair execution and verification.
• DataMind AI supports both mobile mining fleets and fixed processing assets.
• Razor Labs doubled 2025 revenue to ILS 34.62 million and improved gross margin to 45.4%.
• The company remained loss-making, with a 2025 net loss of ILS 16.89 million.
• Free cash outflow reached approximately ILS 17.07 million, compared with ILS 35.23 million of cash and investments.
• South Africa, Colombia and the Democratic Republic of Congo generated about 90% of 2025 revenue.
• Razor Labs shares gained 38.3% during July through July 27 but remained 4.8% lower for 2026.
• New subscriptions, customer expansion and gross-margin progression will be the clearest commercial tests.
• DataMind AI 5.0 must convert technical alerts into completed maintenance outcomes to create lasting value.