One company in the Israeli Defence-AI Deployment Index announced, in its own words, "over 3,000 battle-proven deployments" of its navigation software. It is graded L0, the lowest band, because no independent public source describes any of them. That gap between phrase and record is the reason battle-tested defence technology is the least useful term a buyer can encounter.
The phrase is now a standard fixture of Israeli defence marketing. The Jerusalem Post has reported growing Asian demand for Israel's "battle-tested" technology. Courthouse News, syndicating Associated Press reporting, described demand rising for Israeli weapons technology marketed on the same basis, and noted that Israel's defence ministry says it uses its equipment to defend the country and denies using battlefields as testing grounds. The term is doing commercial work in a live political argument. Neither of those facts makes it a piece of evidence.
What the phrase asserts and what it omits
Read literally, "battle-tested" asserts three things: that the system was present in combat conditions, that its behaviour was observed, and that the observation was favourable. A buyer needs all three, and the phrase supplies none of them in checkable form.
It never names the operator. It never names the date. It never names the configuration, which matters because an export variant and a domestic variant are different products. Above all it never identifies who conducted the assessment, so there is no way to distinguish a formal post-operation review from a sales engineer's impression.
The index applies one test to every entry and it is deliberately blunt: has a named publisher, other than the company, described the system being used by a military in operations. That is the full grading rule for L3. A vendor's own account of its combat record, however detailed, cannot satisfy it, because the index cannot check it and neither can the reader.
One claim, one grade
Sightec, founded 2018 in Tel Aviv, appears in the index at L0. The strongest publicly available item is an announcement carried on the Starburst accelerator's news page, headlined around "over 3,000 battle-proven deployments" of the company's GPS-free visual navigation technology. The number is precise, the phrasing is confident, and the source is the company.
The grade is not a judgement about Sightec's product, its engineering or the accuracy of its figure. It records that the claim is uncorroborated in the public record, which is the definition of L0. A single report from an identifiable outlet describing a military using the software would move it. The company sits in Israeli ISR and sensor-fusion AI, a category where the same distinction separates the top and bottom of the table.
What an independently reported claim looks like
Compare the sourcing behind the 13 L3 entries. Globes reported that the IDF is using Axon Vision's AI against drones. Janes reported the IDF using UVision's mini loitering munition. Breaking Defense reported that the US Army is expanding its use of SMARTSHOOTER's fire-control system, the sole entry in targeting and fire control.
In each case the sentence that carries the weight was written by an outlet with a masthead, a named reporter and a correction policy, and it can be read in full by anyone who follows the link. That is a much weaker claim than "battle-tested" in one sense, because it does not assert that anything worked. It is a far stronger one in the sense that matters here: somebody other than the seller put their name to it.
Note also what the L3 sources do not do. None of them, as used in the index, ties a product to a named strike, operation or casualty event. The index declines to make that connection even where press reporting has, and the reasoning is set out separately in the analysis of the no-strike-attribution rule.
Why the phrase travels further than the evidence
Three structural facts keep the gap open. Buyers of defence software rarely comment publicly on what they have bought. Operators do not publish assessments of individual subsystems. And in Israel specifically, material touching operational matters is subject to pre-publication review, so the reporting that would corroborate a vendor's account is frequently the reporting least likely to be published.
The result is a market in which the marketing claim is free and the corroborating source is expensive. That asymmetry is visible in the index's shape: 13 entries at L3 against 37 at L2, L1 and L0 combined, a distribution examined in the deployment evidence gap. It is not evidence that 37 companies are exaggerating. It is evidence that the publication mechanism for the claim and the publication mechanism for the proof are not the same mechanism.
For a buyer, the practical move is to stop treating the phrase as information and start treating it as a prompt. The questions it should trigger are listed in how to verify a vendor's deployment claims.
Limits of this reading
The index cannot establish that any "battle-tested" claim is false. A company with an uncorroborated combat record may have an entirely accurate one, held under a customer confidentiality clause or a national disclosure restriction. L0 records the absence of published corroboration and nothing more, and this article should not be read as disputing any specific company's account of its own history.
Equally, an independently reported claim is not a verified one. A reporter may have relied on a briefing that was itself supplied by the vendor. The 50 entries record who published what, not who checked it, and the distinction between a report and a repackaged announcement is its own problem.
Frequently asked questions
What does "battle-tested" mean in a defence procurement context?
There is no standard definition and no certifying body. Vendors use it to indicate a product was present during combat operations. It does not specify the operator, the date, the configuration or who assessed the outcome, so it cannot be independently checked from the phrase alone.
Why is a company claiming thousands of deployments graded L0?
The grade records the strength of independent public sourcing, not the size of the claim. Where the only available account of deployments is the company's own announcement, the entry is L0. A single report from an identifiable outlet describing military use would move it.
Do any index grades rely on a company's combat claims?
No. All 13 L3 grades rest on reporting by named outlets including Globes, Janes and Breaking Defense. Company statements about their own operational record are recorded as claims and never used to establish operational use.
Sources
- idf.ai, The Israeli Defense-AI Deployment Index v1.0, 24 August 2026
- Starburst, Sightec announces over 3,000 battle-proven deployments of its GPS-free visual navigation technology
- Courthouse News Service, Demand soars for Israel's battle-tested weapons tech despite global criticism of its wartime conduct
- The Jerusalem Post, Israel's battle-tested tech sees demand in Asia amid tensions with China
- Globes, IDF using Axon Vision's AI to combat drones
- Janes, IDF using UVision mini loitering munition
- Breaking Defense, Israel's SMARTSHOOTER says US Army expanding use of high-tech fire control system, June 2025
Independent publication of idf.ai. Not affiliated with, endorsed by, or connected to the Israel Defense Forces, the Israeli Ministry of Defense, or any government body. Compiled entirely from publicly published sources. No classified, restricted or non-public information. Listed companies may dispute any entry: send the published source that contradicts it and the entry will be amended or removed.