Evolv Technologies pitched New York City on walk-through weapons-screening systems and, according to reporting, highlighted deployments at Florida Disney venues as evidence that the technology could work in busy public spaces. That example helped make the pitch legible; it did not establish that a theme-park setup would perform the same way in a subway. In the month-long subway pilot reported by the Associated Press, the system generated 118 false-positive alerts and found no firearms. The result raised practical questions about alarms, searches, privacy and site-specific testing—not a simple verdict that the technology works or does not.
What Evolv proposed to New York City
Evolv Technologies sells weapons-screening systems intended to flag potential weapons while allowing people to walk through without the item-by-item divestment associated with airport-style screening. In March 2024, Mayor Eric Adams’ administration announced a 90-day process required under New York City’s POST Act before testing electromagnetic weapons-detection systems. Officials publicly demonstrated Evolv equipment as the city considered possible uses in places including hospitals, schools, Times Square, the Port Authority Bus Terminal, City Hall and subway stations. The announcement and demonstration are documented in the Mayor’s Office transcript.
The process and later trials were evaluations, not proof of a permanent citywide purchase. After the subway pilot, the Associated Press reported that the NYPD was still assessing results and had not entered a contract with Evolv at that time (AP report). That is a time-specific account, not a claim about every subsequent procurement action.
Why Disney featured in the pitch
Tech Times reported that Evolv co-founder Anil Chitkara emailed Adams’ office emphasizing the company’s use at Florida venues associated with Disney, including Disney Springs, and that the connection helped draw the administration’s attention (Tech Times, May 14, 2024). The report makes Disney a persuasive reference point: a widely recognized operator of crowded destinations where entry flow matters.
#1 Best Overall
- 18 Precision Detection Zones: Instantly pinpoints the exact height and intensity of metal on a person - from handguns, knives and box cutters to phones, tools and metal contraband - with fully adjustable sensitivity so you catch real threats while ig
- Built for High-Traffic Security: Screen a high volume of visitors at schools, courthouses, government and office buildings, airports, stadiums, arenas, concerts, nightclubs, bars, warehouses, houses of worship and special events - walk-through fast w
- LED Display & Built-In People Counter: Bright side LED bars show which zone triggered the alarm and how strong it is, while the integrated counter tracks total visitors and total alarms - useful data for busy checkpoints and entry logs.
- Indoor & Outdoor Archway Design: The sturdy, free-standing gate assembles in minutes for lobbies, doorways, entrances and gates, for lobbies, entrances and gates, with anti-interference technology that keeps detection stable even near electronics or
- Professional Grade, Low Maintenance: Rugged panels, clear audible and visual alarms and simple controls let any staff member operate it easily, for fast, touch-free, professional security screening that protects your people and property.
That is an argument about venue operations, not an independent validation of accuracy or an endorsement by Disney of subway deployment. The available reporting does not establish that all Disney properties use the same system configuration, sensitivity settings, staffing or screening procedures. Nor does it answer how many entrances were covered, how bags were handled, how alerts were resolved, or what performance measures were independently audited. Those details matter before a buyer treats one venue as a direct model for another.
How the scanner works
New York City’s algorithmic-tools disclosure describes people walking between two towers while sensors process electromagnetic data. The system is designed to identify an anomaly that may indicate a weapon; cameras then provide visual context, and an alert can send monitoring staff a still image and roughly three-second video. Staff review the alert and conduct any secondary search. The disclosure is available in the city’s agency record and its 2024 algorithmic-tools report.
In other words, the system does not itself establish that a person has a gun or make the human decision about what to do next. It generates a potential-weapon alert for staff to assess. Evolv describes its Express system as using electromagnetic sensing, cameras and software to support natural-flow entry; those are vendor descriptions, not independent performance findings (Evolv Express; Evolv FAQ).
Why a theme park and a subway are different tests
A Disney venue and a subway station may both have crowds, but the operating conditions are not interchangeable. The comparison below is an operational analysis, not the result of a head-to-head study.
Rank #2
- PROVEN DETECTION PRECISION: Detects medium pistols from 9" away, large knives from 6", razor blades from 3", and foil-wrapped contraband from 1" - trusted Super Scanner V performance in an upgraded design
- ONE-BUTTON SIMPLICITY WITH MILITARY-GRADE DURABILITY: Intuitive one-button activation and three-button control, plus IP65 waterproofing and mil-spec drop protection with ergonomic microtextured grip
- 7 CUSTOMIZABLE SECURITY LEVELS:From large threat detection to prison-grade screening sensitivity, with instant interference elimination via Offset Button and rebar negation through Reduced Sensitivity Switch
- 125+ HOUR RECHARGEABLE BATTERY LIFE: Eliminate battery waste and costs with included lithium-ion batteries and USB-C charging cable - over 125 hours runtime with just 5-hour charging time
- 8-INCH SELF-CALIBRATIING SURFACE: Large scan area reduces operator fatigue with automatic sensitivity maintenance for consistent, reliable detection without manual adjustments
| Theme-park or destination venue | New York City subway |
|---|---|
| Visitors arrive at destination entrances where screening can be concentrated at selected gates. | Riders enter through a distributed network of stations and access points. |
| Visitors may expect venue security procedures and can encounter a managed entry flow. | Riders are seeking transportation; screening affects access to an essential public service. |
| The venue can shape entrance layout, staffing and bag policies for its own operation. | Existing station geometry, fare control, accessibility, emergency egress and congestion constrain deployment. |
| Operators can plan around a venue’s own traffic patterns and procedures. | Crowd density and movement vary by station and time, under pressure to keep people moving. |
| Screening can be integrated with the venue’s broader admission process. | Any alert or queue can affect riders who may have limited alternative routes or time. |
A buyer would need to know whether Disney-area deployments screened every entrance or selected gates; what staffing ratio and bag rules applied; which sensitivity settings were used; and how alerts were handled. The reporting cited here does not establish those particulars. Without them, Disney’s experience may show that a system can be incorporated into certain high-volume venue operations, but it cannot establish performance in a transit network.
What New York’s testing found
New York’s evaluation unfolded through public demonstrations and testing at different sites before the subway trial. A Bronx hospital test drew separate scrutiny: reporting described false alarms in more than 85% of cases, while the AP reported that Evolv’s CEO had acknowledged subway environments were not an especially suitable use case (Tech Times; AP). That hospital figure should not be compared directly with the subway result: the settings, periods, methods and likely definitions differ.
For its month-long subway pilot, the AP reported that Evolv equipment generated 118 false-positive alerts, a 4.29% false-positive rate in the reported dataset, and detected no firearms. Searches did find knives or other blades, according to the AP’s account (AP results report, October 24, 2024). The rate is meaningful only alongside the dataset and its definition of a false positive; it should not be generalized to every site, configuration or item.
No firearm detection during a limited pilot does not establish that the system missed a firearm: the result may simply reflect that no firearm was presented or encountered. But false alerts still have consequences. They can consume staff time, prompt searches, delay riders and erode trust. To judge threat-detection performance, an evaluation also needs controlled, documented testing of weapons and ordinary objects, with results separated by site and system settings.
PC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11Outdated Drivers Are Slowing You Down
One free scan finds every outdated or missing driver and matches the right update for your exact hardware.Free scan · exact hardware matchRank #3
- High Sensitivity & Easy Operation: This handheld metal detector features high sensitivity and one‑button operation with no complicated setup. It quickly detects nails, screws, weapons, and jewelry, making it ideal for security screening at airports, schools, and large events.
- Adjustable Sensitivity: This metal detector has 2 adjustable sensitivity modes. To exclude small metal objects, simply switch to low sensitivity mode, which reduces sensitivity significantly, allowing detection of only larger metal items.
- 2 Alert Modes: This metal detector wand features an audible sound mode for clear safety checks and a vibration mode for quiet or confidential patrols.
- Stable Performance: Even when the battery voltage drops from 9V to 7V, the detection distance remains unchanged, ensuring reliable operation at all times.
- Power Supply: Powered by a 9V battery (included). Easy to replace and widely available, no charging or cables needed for reliable on-the-go use.
Accuracy claims, criticism and the FTC action
Critics have questioned both nuisance alarms and the possibility of missed weapons, including whether company testing reflects real operating conditions. The Legal Aid Society warned in AP reporting that false alarms could create privacy and civil-liberties harms (AP report). A report by Stop Spying, summarizing IPVM-related testing, also examined alarm rates and operating behavior; it is public-interest and investigative evidence, not a regulator’s certification (Stop Spying report).
Evolv has disputed some reporting and testing interpretations, arguing that context such as system settings, customer procedures and the specific threat configuration can affect what a reported test means. The company’s responses should be read as its position, rather than as independent confirmation of performance (Evolv response to IPVM coverage; Evolv statement on reported claims).
In November 2024, the Federal Trade Commission announced action alleging that Evolv misrepresented the system’s ability to detect weapons and ignore harmless items. The agency’s announcement describes its allegations and proposed remedy; it should not be confused with a site-specific accuracy test or treated as a final adjudication by itself (FTC announcement; FTC case page). A procurement decision should distinguish vendor marketing, customer-pilot results, independent testing and legal findings rather than treating them as interchangeable evidence.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What riders’ privacy and civil liberties depend on
The city disclosure makes one consequence concrete: when the system flags a potential firearm, it can transmit a still image and short video to monitoring staff, and the city identifies the general public as the affected population. The disclosure does not establish facial recognition, nor does it resolve every question about data handling. Riders and procurement officials need clear answers on retention, access, deletion, storage location, use for model training and vendor reuse.
Free tools Windows power users keep installed
One-click scans. No signup required.
Rank #4
- ❤【HIGH SENSITIVITYFOR FAST DETECTING】NATOTINORCH Metal Detector Safety Bar is made of high quality components, it can detect hidden small metal objects Detect weapons,knives and other metal objects.Ideal for venues requiring security checks.
- ❤【SOUND AND VIBRATIONG TWO ALARM MODES】The security scanner has a mode switch, which can be freely switched between sound and vibration alarm modes according to various detection and use environments.
- ❤【ADJUSTABLE SENSITIVITY LEVEL】If the detection distance does not meet the requirements, the sensitivity can be adjusted. Press the low-sensitivity switch during the detection process, the detector will only respond to larger metal objects, such as knives, pistols, etc., and will not respond to small metals, such as rings, keys, coins, etc.
- ❤【POWER OPERATED】Powered by one 9V battery standard ( included) or rechargeable battery (Not included). With one external charging port, the battery charger is not included. Please avoid charging it when you have no rechargeable battery, otherwise it can't work normally. And please turn off the detector when charging it.
- ❤【EASY TO CARRY AND WIDELY USED】This metal detector wand comes with a holster clip making it easy to carry. It can be widely used in airports, stations, docks, school examination rooms, exhibitions, museums and other places where safe metal detection is required. Can be used indoors and outdoors.
There are also procedural questions. What happens when an alert concerns a permitted knife, work tool, medical device or other non-threat item? Can a rider decline screening, and what happens then? What standard governs a physical search, and how can someone challenge an erroneous one? These are deployment and governance questions; the available city disclosure does not establish the answers. A system that produces a targeted alert can still lead to a consequential human search.
Regulatory and corporate developments after the pitch
The story also widened beyond the city pilot. Evolv disclosed in SEC filings that it received an SEC subpoena in February 2024 concerning aspects of its marketing practices and a DOJ document request in November 2024. The company said the DOJ informed it on August 7, 2025, that it was no longer the subject of that investigation. The same filings describe the SEC matter as ongoing at the time of filing (Evolv third-quarter 2025 filing; Evolv 2025 annual filing).
The company also faced a corporate investigation into sales practices and financial reporting. The AP reported that Evolv removed its CEO amid a sales-misconduct investigation and warned investors that some recent financial statements were unreliable (CEO report; financial-reporting report). These developments do not settle whether a particular scanner configuration performs well at a particular site; they make careful scrutiny of claims, documentation and contractual accountability more important.
What a responsible public-sector test should measure
For a transit agency, hospital, school or venue, the relevant question is not whether a system has a prominent customer. It is whether a defined configuration meets the buyer’s needs under representative conditions, with manageable consequences when it alerts. A serious evaluation should include:
- Independent or jointly observed acceptance testing with representative weapons, bags, clothing, mobility aids, tools and everyday objects.
- Separate sensitivity and specificity results, with false alarms reported per 1,000 or 10,000 people and a clear denominator, test population and definition.
- Documented missed-threat protocols and performance by setting, including crowded entrances and the actual site environment.
- Realistic throughput and staffing assumptions, including operators, secondary-search staff, supervisors and response personnel.
- Audit access to logs, alert images, software versions, sensitivity settings and incident records.
- Written controls for image retention, access, deletion, cloud hosting, model training and vendor reuse.
- Accessibility and equal-treatment review, an escalation process for alerts on permitted objects, and contractual remedies such as service levels, warranty and cancellation rights.
Natural-flow screening may reduce the need for some forms of divestment, but it does not eliminate staffing, secondary searches or emergency response. Raising sensitivity may improve detection of some threats while increasing nuisance alerts. A visible scanner may deter some people from carrying weapons, but deterrence is not the same as measured detection accuracy. In a subway, a technically capable system can still be a poor operational fit if entrances bottleneck, riders bypass it, nuisance alerts become routine or staff cannot promptly resolve each alert.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
