Their Claims vs. The Record

The RTIC was sold with specific, quotable claims. This page holds each one next to the record. Nothing here is paraphrased unfairly: quotes are verbatim from the cited coverage, and where I only have a paraphrase, I say so. Use it at the podium, in letters, and whenever someone repeats the sales pitch.

”It solves crime” / “This is about victims”

The claim. Chief Stepp, May 12: “This is about improving emergency response, investigative efficiency, transparency and officer accountability.” “Time is often the difference between life and death.” “This is not about software — this is about victims.” (828NewsNow · Annie Dance) The flagship proof offered: the March 2025 Hillcrest shooting, where Flock located the fleeing vehicle “almost instantly” (Capt. Silberman, WLOS).

The record. One anecdote is not an evaluation. APD has offered no aggregate statistics (no arrests-attributable count, no response-time data) in any coverage I can find. The independent literature shows plate readers help solve some cases and deter nothing (The National Wave); Austin’s full-program arithmetic was 117 million scans for 228 arrests; Atlanta’s homicide clearance rate fell during a 5,000-camera buildout. If the RTIC’s benefits were measurable, the measurements would be on the table. Ask for them.

”There’s a really robust audit trail” / “100% of the time, we have reasons”

The claim. Stepp, May 12: “There’s a really robust audit trail built into all of this.” Tech manager Wingo, Aug 6: “Our lieutenant does quarterly audits. He just did an audit, and 100% of the time, we have reasons.” (828NewsNow)

The record. “Having a reason” is not the safeguard it sounds like. Mooresville’s officer had a reason logged for all 31 alleged stalking searches: she made them up, and they passed review (The Abuse Record). EFF found 14%+ of 11.4 million Flock searches justified with the single word “investigation.” And Asheville’s audits are internal and unpublished. When a resident formally requested the city’s Flock audit records in July 2025, the city refused, and the promised public ALPR dashboards return a 404 (Documents). An audit trail the public can’t see, scored by the people it audits, reporting 100% compliance, is not accountability. The short version: they tell us to trust the audit trail, and when residents legally asked to see it, the city refused. Both halves are in the city’s own record.

”We own the data and control all sharing” / “We will not be sharing with federal immigration”

The claim. Stepp, on the May 12 recording (4:47:05): “Axon neither has access to the Fusus or to the system, or retains data. Actually, police department owns the information”, and to WLOS: “APD owns the data that we collect, and we have to agree to share that information out, regardless of who requested it.” Deputy Chief Aardema, in the same WLOS coverage: “We control Axon Fusus and we will not be sharing information with federal immigration.” Capt. Moore: “We do not and will not share that information short of a court order.” Axon’s rep, to the Watchdog: “We do not have access. We’ve never had access.” (WLOS · Watchdog)

The record. Three problems, all documented. First, the executed contract has never been released, so “we own the data” is unverifiable, and Hillsborough, NC killed its program after reading the standard contract clause letting the vendor disclose data on its own “good faith belief”; a resident read that same clause into Asheville’s record on May 12 (2026-05-12 City Council). Second, Asheville’s plate data flows to agencies across a 500-mile radius: the documented ICE pathway nationally isn’t a federal request to APD, it’s another agency in the network running the search as a favor (4,000+ logged instances), which no APD approval process touches. Third, vendors have failed identical promises repeatedly: Flock ran a CBP pilot on customers’ data without their knowledge, which is why Evanston terminated for breach. And Aardema himself conceded the national pattern on the record: “in other areas you see law enforcement agencies using data in ways that would not be appropriate for the Asheville community.” The promise, restated accurately: it won’t happen here because we’re different. Every city in The Abuse Record was different too.

”We have blocks on immigration and reproductive search terms”

The claim. Wingo, July 8: “We have blocks on words that have anything to do with immigration, anything to do with reproductive issues.” (WLOS)

The record. The Texas deputy who searched 83,000 cameras for a woman who had an abortion later had the search reclassified as a “missing person” case. Keyword blocks filter honest descriptions, and abusers stopped writing honest descriptions the moment audits drew attention (the ACLU documented officers coaching each other to launder the reason field). A keyword filter only catches officers who describe their search honestly.

”Reactive, not predictive”

The claim. APD’s consistent framing: the RTIC is “a reactive system, it’s not predictive police” (Aardema).

The record. APD’s own former chief told the Public Safety Committee otherwise. In July 2025, Chief Mike Lamb described the department’s approach to that committee using the words “predictive policing” (twice, on the record) while presenting crime-reduction stats. The “reactive, not predictive” line arrived later, once “predictive policing” had become a public-relations liability. The capability and the intent the department described in 2025 don’t vanish because the 2026 script changed. (And on immigration: APD’s own words to the Public Safety Committee, July 2025, “APD does not utilize the license plate reader systems at all to alert to immigration violations… They’re turned off.” Turned off is a toggle, not a wall: the capability sits in the platform, one settings change away.)

”It’s reactive, violent-crime-only, not surveillance”

The claim. Aardema: “We have been clear from the get-go that this is a reactive system, it’s not predictive police and the primary focus is going to be on violent crime.” Stepp: “We’re not going to just randomly surveil or monitor people.”

The record. Nothing in the approved package structurally limits the system to violent crime: the limits are revisable policy (Promise Tracker). Where these systems drifted, they drifted exactly here: Austin’s audit found 10–20% of searches with no documented purpose at all; San Francisco’s police tapped a private camera network to watch protests live; Memphis surveilled activists in violation of a federal consent decree (The Abuse Record). “Reactive” also quietly concedes the efficacy point: a reactive system doesn’t prevent the violent crime it’s sold on.

”No facial recognition”

The claim. Stepp, repeatedly; Axon’s rep: “We also do not use any form of facial recognition.”

The record. The exclusion exists only in Resolution 26-97, a resolution and not an ordinance: no enforcement mechanism, no penalty, reversible by any future council in one vote, and unverifiable against the still-unreleased contract. The platform’s capabilities are set by the vendor’s roadmap, and both major vendors add AI features mid-contract, which is why the cities that mean it require a council vote before any new capability activates (How Cities Won). A commitment without a mechanism is a preference, not a protection. Notably, even NC Attorney General Jeff Jackson’s letter of support (read into the record by Mayor Manheimer on May 12) hedged its every finding with “Based on what APD has described”: the state’s top law-enforcement officer evaluated the description, not the documents, because the documents weren’t available.

”Drones and real-time intel de-escalate”: the deep dive

The claim. Council member Hess, March 25, 2025, on the city’s video at 32:50: “When you actually responsibly deploy technology like drones and real-time intelligence, what you do is you actually de-escalate the risk for officers and our citizens. You’re removing the need to send an officer blindly into a dangerous situation and that lowers the chance of use of force [incidents].” He cited Durham: “They didn’t throw out public safety, they modernized it.” Wingo’s tour anecdote: a drone revealed a “gun” was a cigarette lighter.

The record. This claim has never been measured true, anywhere:

• No evaluation of any drone-as-first-responder program, including Chula Vista’s, the most-studied on earth (20,000+ flights), has ever measured use-of-force outcomes. The only rigorously documented benefit is response speed. The author of the field’s first randomized controlled trial says the literature is “anecdotal testimonies… hardly any rigorous analyses.” (GovTech data analysis · USC/Mourtgos RCT)
• The RTCC literature is the same. The field’s own research institute at FSU concedes “little empirical work has been conducted on their effectiveness”; the best pro-RTCC study (Miami) measured case clearance; use of force was never studied. (FSU RTCC Institute · Justice Quarterly)
• The measured record runs the opposite direction. LAPD’s Inspector General: 32.3% of Flock stolen-vehicle alerts were false: 161 in two months, each capable of triggering a guns-drawn “high-risk stop”; the IG warned false positives raise the “risk of a confrontation with serious consequences.” The Institute for Justice’s database counts 27 documented wrongful ALPR stops since 2018: in nearly two-thirds, officers had guns drawn before discovering the error. In Aurora, a misread plate put four Black children, ages 6–17, face-down at gunpoint; the city paid $1.9M. Real-time “intelligence” that’s wrong doesn’t de-escalate; it manufactures armed confrontations with innocent people. (NBC LA on the IG audit · IJ database · CNN on Aurora)
• What drones actually do in practice: WIRED’s analysis of ~10,000 Chula Vista flights found them concentrated over the city’s poorest, largely immigrant neighborhoods, ~4,700 residents overflown per flight, one in ten flights with no stated purpose; the city fought four years in court to hide the footage and lost, paying $500K in fees. LAPD launched DFR drones 31+ times over a single ICE-protest. And locally: APD’s drones are already documented in use for homeless-related enforcement, not hostage standoffs. (WIRED via Reason · The Intercept)
• The Durham example is backwards. When Hess praised Durham’s “modernization,” Durham had voted 4–2 to reject ShotSpotter (after 91% of its unprompted alerts proved dead ends), had no real-time crime center at all; and when one was proposed in Feb 2026, council declined to even vote on it. What Durham actually scaled is HEART, its unarmed crisis-response program, and Durham’s violent crime fell 17% in 2025 without any of the tech. (WUNC · ABC11 on the RTCC rejection · ABC11 on the crime drop)
• The premise is also false. 2025 was the safest year for American police officers since 1943: 111 line-of-duty deaths, down 25%; a third are traffic-related, and ambush (“blind entry”) is the rarest assault circumstance in FBI data. No drone or RTCC addresses the actual risk profile. (NLEOMF · FBI LEOKA)
• The proven de-escalation investment exists, and it isn’t this. The only intervention with a peer-reviewed causal evaluation showing reduced enforcement outcomes is unarmed response: Denver’s STAR cut targeted low-level crime 34% at a quarter the cost of police response (Science Advances, 2022); Durham’s HEART diverted 2,000+ calls with responders reporting feeling safe 99% of the time and never once needing police backup. Asheville already runs a Community Responder pilot: 8 positions on grant patchwork, while the surveillance center gets $1.14M. (Science Advances · Durham HEART evaluation)

“Over 95% of our APD officers are trained in crisis intervention”

The claim. Hess, same meeting, ~37:05 on the video: “Over 95% of our APD officers are trained in crisis intervention for mental health, addiction, and de-escalation. And that’s something we should be proud of.”

The record. The last public figure was 90% (WLOS, Oct 2023); no public document supports “over 95%”; a records request for the roster would settle it. More fundamentally: the DOJ’s own evidence clearinghouse rates CIT “Ineffective” at reducing use of force and arrests (the meta-analytic literature finds benefits in officer attitudes and referrals, not force outcomes), and CIT’s Memphis-model founders themselves insist it’s a specialist-team model, not a training percentage. Even if the number were true, it wouldn’t demonstrate what it’s offered to prove. (WLOS 2023 · DOJ CrimeSolutions · Taheri meta-analysis)

Positioning: the Public Safety Committee, which Hess chairs, meets Sept 24, 11 AM (city page); the LAPD IG audit and the Durham inversion land hardest in his own venue. The CIT number is best tested by records request, not speech: either the document exists (then cite DOJ’s rating) or it doesn’t (then the claim dies quietly).

”Trust will not be assumed. It will be demonstrated.”

The claim. Stepp, May 12 (the best line of the night).

The record, so far. The executed Axon contract: unreleased. The Flock audit logs: refused under a records exemption. The promised ALPR dashboards: 404. Roney’s 29 questions: unanswered since March 2025. The quarterly audits: internal, unpublished. (Promise Tracker · Documents)

“It saves the city money”

The claim. Mayor Manheimer: “we received a grant from the federal government to be able to pay for our own”, and framed Roney as voting “against a grant that would help save the city money.” (BPR)

The record. The arithmetic, from the city’s own May 12 staff reports (Documents):

• Status quo: $30,000/year to the county, all-in. No staff, no facility, no renewal risk.
• “Our own”: the grant covers a room and prepays Fusus (~$467K) for ~7 years. Asheville then owns: the bundled Axon payments ($740,000 in FY27, $777,500 in FY28 per the staff report; trajectory rising), an RTIC staff (dispatcher, supervisor, two operators; plausibly $250K+/year), facility operations, and the post-2033 software renewal at whatever Axon then charges. The staff report’s own documented bundling savings over the entire 7.5-year contract: $46,948.25, about what the city pays the county for 18 months of access today.
• The escalation baseline is Axon’s own investor pitch: net revenue retention of 124–126% for 20+ consecutive quarters: the same customers paying ~25% more each year. A resident literally submitted Axon’s investor deck as public comment on May 12; it’s in the city’s own meeting folder.
• The question that was never answered: Patrick Conant asked council, on the record, “What is the ongoing operational cost after the grant runs out?” No answer has ever been given publicly. The honest framing is “the grant lets us buy a much bigger capability”; the savings framing fails unless total cost of ownership comes in under $30K/year, which nothing in the record supports. The city replaced a $30,000-a-year subscription with a facility that will cost more than that per month to staff, and called it saving money. The grant is the bait on a financial hook: federal money covers the setup and creates the illusion of “free,” while the 7.5-year contract locks in escalating obligations that outlive the grant by design. Ask any councilmember what the FY2034 line item is. Nobody knows, and that’s the deal they signed.

”We’re short-staffed, this is a force multiplier”

The claim. The staffing-shortage justification runs through the whole sales arc: cameras and the RTIC let a depleted APD “do more with less.”

The record. Three problems, each documented:

1. The multiplier has no measured product. The technology’s own record is the efficacy record above: 117 million scans for 228 arrests; Atlanta’s clearance rate fell after its rollout; the only randomized trial of Axon’s efficiency product (Draft One) found no time savings at all.
2. It competes with staffing rather than relieving it. The RTIC needs its own staff (dispatcher, supervisor, two operators per the city’s staffing plan) pulled from the same budget and, for sworn roles, the same shrunken pool. The $740K+/year Axon line is money that cannot fund officers, co-responders, or the community programs residents actually asked for.
3. The tool doesn’t do the job that’s short-handed. Cameras don’t answer calls, de-escalate crises, or patrol. What they generate is more work: alert queues and unverified leads (1-in-3 false stolen-vehicle hits in LAPD’s audit, The Abuse Record) that someone still has to chase, at gunpoint risk to whoever’s car got misread. Understaffing is an argument for spending on people. It’s the case against a subscription that consumes the people budget.

The shifting numbers: 37% → 36% → 24%, and a “100%” that wasn’t

The claims. To the Public Safety Committee, on the record: Chief Lamb, July 2025: violent crime down “37%” year-to-date. Bo Hess, November 2025, in his tribute to the retiring chief: “an over 36% reduction in violent crime.” Interim Chief Stepp, January 2026, reporting the actual annual figure: “a 24% reduction in violent crime last year.” Same city, same year, same room.

The record. A 24% drop is real and worth having, which is exactly why the inflation matters: the sales window for the RTIC ran on the mid-year 37%, the durable number is 24%, and nobody on the committee reconciled them. It’s the same pattern as the camera counts (177,000 in March, 19,000 in July): the numbers move with the argument, and the venue where they should be checked doesn’t check. And in November 2025 Lamb claimed “a multi-year 100% clearance rate for homicides”, minutes before his own deputy chief’s figures showed two 2024 homicides still open. No member flagged the contradiction.

The sheriff’s crime-drop claims: the deep dive

The claim. Sheriff Miller credits the county RTIC for consecutive annual crime reductions; headline: a “15% countywide crime drop” in 2022. (WLOS)

The record. The state’s own data, SBI annual summaries:

• The 15% wasn’t countywide. SBI Table 5 shows the −14.7% applied only to the Sheriff’s unincorporated-county slice (~28% of county crime). Countywide, the 2022 drop was −5.6%, and countywide violent crime rose 12.7% the same year. This critique was first made publicly by then-Rep. Chuck Edwards, the man who later funded Asheville’s RTIC, in a 2023 op-ed titled “No, crime is not dropping in Buncombe County.”
• The trend predates the cameras. Countywide rates fell in 2019→2020 and 2020→2021, before the RTIC was operational; whatever drove the trend wasn’t the thing that didn’t exist yet.
• Then the claim expired. In 2024, with ~1,800 cameras running, the Sheriff’s jurisdiction went +1% while the state fell 4.2% and national property crime hit its lowest level since 1961. BCSO burglary: +42%. BCSO violent crime had already spiked +52% in 2023. If the RTIC caused the 2022 drop, what caused the 2024 rise?
• The comparison that ends the argument: countywide index rates 2021→2024: Buncombe (RTIC) −24.6%; Catawba County (no crime center at all) −42.2%. And within Buncombe, the biggest drops came from Asheville PD, which had no RTIC of its own, just a $30K access subscription.
• No independent evaluation of the BCSO RTIC has ever been conducted. Every claim traces to the Sheriff’s own releases. The academic literature’s verdict on RTCCs generally: “a dearth of evaluation research,” with the favorable findings limited to case clearance, not crime rates. (Justice Quarterly)

Data hygiene for anyone using this: always name the jurisdiction (Sheriff’s slice ≠ countywide ≠ Asheville), cite the SBI table and year, and prefer sheriff-to-sheriff comparisons; this is the version that survives a hostile fact-check.

The vendors’ own claims, already fact-checked

The six claims Axon and Flock reps made at the April 28 work session (immigration filter, complete APD data control, mandatory search reasons, quarterly-audit accountability, city data ownership, no facial recognition) were fact-checked point-by-point in the CAMS/Sunshine Labs one-pager, and Flock’s rep dismissed the hacking coverage as device misconfiguration in the same season researchers rooted a camera in 30 seconds, ~70 cameras were found streaming openly, and Congress asked the FTC to investigate. Vendor credibility is itself a documented issue: Sedona (“misled and lied to”), Oshkosh (heat-map denial refuted by the police chief in 24 hours), Evanston (unauthorized reinstall). See How Cities Won: vendor dishonesty is the single fastest flipper of council votes. Keep this page current; the claims will keep coming.

Related: Promise Tracker · The Abuse Record · Claims Library · Documents