Why Taiwan's military drone tenders keep failing

Taiwan's Ministry of National Defense has penalized counter-drone startup Tron Future NT$140.3 million (about US$4.3 million) for failing to deliver 26 counter-drone systems for Kinmen, Matsu and other bases. That is a forfeited NT$49.4 million performance bond plus NT$90.9 million in late-delivery penalties, according to a ministry report to the legislature covered by the Liberty Times in late October. The Army plans to start the project over.

Most coverage has focused on Tron. The bigger problem is how the military sets prices. In several recent drone tenders, the ministry either asked for more than its budget could buy, or set a budget too low to pay for the non-Chinese parts it requires. The first drives good suppliers away. The second rewards suppliers who cheat. So far, the government's fixes mostly target suppliers, not prices.

The Tron contract

In August 2022, people in Xiamen flew consumer drones over Taiwanese military posts on Kinmen and posted the videos online. The ministry promised a permanent system to detect and jam drones around Kinmen and Matsu. Four years later, troops there are getting handheld jammers instead.

The ministry says Tron received no money under the second contract. Under the first, Tron got more than NT$22 million from a defense foundation's research fund, by Domino Theory's estimate. Tron says it built and tested the system at a loss, in good faith.

Problem 1: asking for more than the budget can buy

The 2024 tender asked for more than the 2022 one. According to the tender document, as reviewed by Domino Theory, the system had to detect very small drones (a radar cross-section of 0.01 m²) at 6 km instead of 5 km, and jam them at 4 km. The tender's technical specs aren't public; only registered bidders can download them. Foreign industry experts told Domino Theory a drone that small couldn't carry a real explosive. A drone carrying one or two kilograms would look at least three times bigger on radar.

The budget came to about NT$38 million (roughly US$1.2 million) per system. A July 2026 US Congressional Budget Office report, cited by Domino Theory, put a full counter-drone setup for one site at US$74 million, plus US$5 million a year to maintain. The two aren't directly comparable, but the gap is large.

Here is how the tender went:

1. Round one (August 2024) failed under a rule requiring three bidders. Local media reported that only Tron bid. Most companies stayed out. One sales rep said the specs couldn't be met at that price.
2. Round two dropped the three-bidder rule. Tron and U&U Engineering bid, and both failed the proof-of-concept test.
3. Round three went to Tron. Tron only started advertising radars that matched the tender's specs after it won.

The 15-member selection committee included four working civilian experts. None of them attended the meeting that gave Tron the contract. Domino Theory reviewed 271 similar ministry awards from 2025 and found no other case like it.

When a tender asks for more than it pays for, many experienced suppliers stay out. The field narrows to the few willing to take on the risk, which raises the odds that the winner can't deliver. In July, the ministry's procurement chief, Lt. Gen. Chao Ya-ping, told legislators the ministry finds it hard to judge what the market can supply when it plans a purchase.

The military isn't the only buyer with this problem. In June, the Coast Guard's NT$1.2 billion surveillance-drone tender failed after at least three of four qualified companies didn't bid. People in the industry said the specs were close to military grade, which ruled out proven commercial products.

The Army now plans to start over, with a new tender in its 2027 and 2028 budgets. It says it will look at proven domestic and foreign technology, including vendors with combat experience. It also plans to seek a supplementary budget so it can pay more per system.

Taiwanese outlet Up Media reported in July that price is now the main obstacle. The National Police Agency's counter-drone tender for critical infrastructure budgets NT$85 million for its top-tier system, which has similar specs plus handheld jammers. The Army paid about NT$38 million per system. According to the report, prices have risen since the original estimate, and at the old price no company might bid.

Problem 2: budgets too low to pay for compliant parts

When the budget is low and the cheapest bid wins, the tender usually doesn't fail. It goes to the lowest bidder, and what happens next varies. A capable supplier may take the loss and deliver anyway. An inexperienced one may find it underpriced the job. And some may cut corners to make the numbers work.

In 2025, Hang Jian Technology won an 8th Army Corps drone contract with a bid of NT$2.95 million. The military had estimated the job at more than NT$4 million. Prosecutors say the company's head and a technical director at Real Plus Technology built the drones with Chinese chips and flight controllers, removed the labels, and submitted fake certificates of origin. A source told the Taipei Times the parts were bought on Taobao. The Army found the problem during inspection and didn't pay. The case is still under investigation, and we covered it when it broke. In October, prosecutors also detained two executives of listed drone maker Jiin Ming over similar allegations. Defense Minister Koo has said the problems came up in small purchases that weren't torn down for inspection, and that those will now be inspected too. We looked at both cases in Why Taiwan's origin prosecutions matter to foreign drone buyers.

Taiwan's drone tenders require local production or assembly and ban Chinese parts and Chinese-funded bidders. That makes compliant drones more expensive to build. If the cheapest bid wins, a company that breaks the rules has an advantage until it gets caught. A Liberty Times commentary warned that this lets cheaters win contracts and pushes honest companies out.

Opposition legislator Ma Wen-chun says checking where a drone was assembled isn't enough. Inspectors need to trace chips, flight controllers, radios and storage back to the companies that made them. Otherwise, relabeled parts, or parts shipped through other countries, will get through. Our component risk map covers which parts are hardest to source outside China.

What the government is changing, and the counterarguments

The ministry's fixes so far focus on suppliers. It told the Legislative Yuan that bidders will have to agree to checks on their funding and staff. Companies that fail inspection or acceptance testing will be publicly listed and banned from future bids, including as subcontractors. The Public Construction Commission also issued drone procurement guidelines in February 2026 covering budgets, standard specs, local production and non-Chinese supply chains.

These rules make it harder to cheat or overpromise but they don't fix tenders that ask for more than they pay for. The only change so far that touches price is the higher budget for the counter-drone re-tender.

There are fair arguments against blaming price:

• The spec may have been reasonable. A source quoted by Storm Media said the 6 km detection and 4 km jamming targets were demanding but realistic, based on lessons from Ukraine. Up Media reported that the Army surveyed foreign suppliers, found their radars all reach 6 km, and will keep that requirement.
• It may have been a communication problem. Tron's Korean partner, Huneed, said the tests failed because of how the ministry explained its requirements.
• The technology is still new. An industry insider told Domino Theory they'd be shocked if any counter-drone company hit its promised performance on the first delivery. Most short-range systems in Ukraine still rely on human operators.
• Officials say the process was sound. A former deputy defense minister told legislators the contract was awarded after proof-of-concept testing and scoring by an outside committee.

None of these explains why most qualified companies didn't bid, why the Coast Guard tender failed the same way, or why the ministry's own fix is to raise the price.

What this means for suppliers and buyers

Taiwanese suppliers: Domestic military demand looks big on paper, but it's currently unreliable. Tenders fail, budgets get cut, and price caps can make it impossible to follow the rules and still make money. In May, the legislature cut domestic drones from the special defense budget. Since then, the legislature has passed a NT$240 billion drone act, and the Cabinet has proposed a NT$55.9 billion supplementary budget for coastal attack drones. More money doesn't fix tenders that ask for more than they pay for. Meanwhile, export demand has kept growing. Smaller suppliers shouldn't rely on domestic military contracts alone.

In the US, there are claims some Drone Dominance vendors are selling below cost to stay in the program. That can be a rational bet: a vendor that completes a prototype order can win a follow-on production contract without another competition, and a much larger budget has been requested, though Congress hasn't funded it. Taiwan offers no equivalent. Each tender is competed from scratch, a failed contract can get a supplier barred, and funding is reviewed every year. A Taiwanese supplier that takes a loss to get in has no guaranteed next contract to earn it back.

Foreign buyers: The Hang Jian case shows a certificate of origin can be faked. Taiwan's own army only caught it at final inspection. Buyers who need drones without Chinese parts should write audit rights, batch-level tracing, notice of parts changes, and inspection of multiple batches into their contracts. We made this case in "Non-red is a requirement, not a label".

The re-tender is the test. The Army's second attempt will show whether the lesson has been learned. If the new budget closes the gap with the police agency's price, radar makers that stayed out last time have a reason to bid. If the price stays near the old figure, expect the same short list of bidders and the same risk. The details matter too: whether the 0.01 m² requirement stays, whether the three-bidder rule comes back, and whether the selection committee's experts show up.

Taiwan now has more money for drones than it did a year ago. The Tron case shows money alone isn't enough. When a tender's price doesn't match what it asks for, the strongest suppliers stay out and the risk lands on whoever is left. Until budgets are set from what the market charges, not an estimate made years earlier, Taiwan will keep losing years to contracts that fall through, and its best suppliers will keep looking to export markets instead.