OPINION
MARK T. PITTMAN, JUSTICEThe' trial ■ court found Appellant Ricky Lynn Elks guilty of driving while intoxicated — felony repetition (DWI) and sentenced him to.serve ten years in prison and to pay a $1,000 fine, but the trial court suspended imposition of the confinement portion of the sentence and placed Appellant on community supervision for eight years. In his sole point, Appellant contends that “[t]he trial court erred by denying [his] motion to . suppress because the [police] officer lacked reasonable suspicion to initiate a traffic stop based solely on the ambiguous notation that insurance was ‘unconfirmed’ and because the evidence was not developed to establish the reliability of the license check computer system.” Given the specific facts of this case, we uphold the trial court’s conclusion that reasonable suspicion justified the stop, and we affirm the trial court’s judgment.
I. Statement of Facts
Appellant-was driving his Jeep northbound on Rio Grande Boulevard in Euless, Texas when Officer David Chaney, a police officer for the City of Euless, stopped Appellant solely because the patrol car’s computer indicated that liability insurance on the Jeep was “unconfirmed.” Ultimately, Officer Chaney determined that Appellant had committed DWI and arrested him. Appellant filed a motion to suppress all evidence gained from the stop, and after a hearing, the trial court denied the motion. The trial court found Appellant guilty after a bench trial, and he appealed.
A. Evidence Regarding the Stop
Officer Chaney testified as follows about . the stop of Appellant:
• At approximately 3:30 p.m. on September 1, 2015, Officer Chaney parked his patrol car in the center median of the 2200 block, of Rio Grande Boulevard in Euless;
• Officer Chaney’s patrol car had a database maintained by the Federal Bureau of Investigation (FBI) and the Texas Department of Public Safety (DPS);
• A 2004 Jeep Grand Cherokee SUV traveled northbound past Officer Chaney;
• Officer Chaney typed the Jeep’s license plate into the computer, “[j]ust checking [the database] for warrants, . registration, the validity of the registration, and confirmation that the ve- ■ hide had insurance or not”;
• After Officer Chaney input the Jeep’s license plate number, the “insurance return” from the database showed that the Jeep’s insurance status was “unconfirmed per the State of Texas,” which meant that the Jeep was “not showing to have current insurance as required by Texas state law”; . ,
• The return from the database would have said “insurance confirmed” if the Jeep had been insured;
• Officer Chaney’s training and experience led him to conclude that the Jeep was uninsured;
• Officer Chaney stopped the Jeep based solely on the “unconfirmed” insurance return from the database;
• Appellant did not provide Officer Chaney with insurance 'information at the time- of the stop;
• Officer Chaney conceded that Appellant’s insurance card admitted at trial indicated that he had insurance coverage on the Jeep on the date of his arrest; and
• It is possible for a driver to have an insurance card but not.actually have insurance.
B. Evidence Regarding the Database Providing the Insurance Information
Officer Chaney testified as follows regarding the database:
• It. was maintained by the FBI and DPS and part of NCIC and TCIC;1
He .had training for the database, which the trial court could have inferred Officer Chaney indicated was Texas Law Enforcement Electronic Telecommunications System (TLETS), early in his career and completed refresher training, a refresher test, and continuing education on the system évery two • years;
• Officer Chaney did not know how the database worked;
• Officer Chaney did not know if the database was • web-, internet- or cloud-based;
• Officer Chaney did not know the database’s requirements for accuracy;
• As far as Officer Chaney knew, the database “could be 99 percent accurate” or “50 percent accurate”;
• Officer Chaney did not 'know how often insurance companies reported ' information or how often they were required to report’ information to the database;
• Officer Chaney did not know how insurance companies reported information to the database;
• To the best of Officér Chaney’s knowledge, insurance companies reported coverage of vehicles to the State of Texas;
• Officer Chaney did not know how often “they check”;
• When questioned if he knew how accurate the database was, Officer Chaney answered, “I have to take it as being accurate”;
• Officer Chaney stated that police officers “take it on good faith that the information that is provided ... [by the database] is valid information”;
• No one had ever told him that the database was unreliable;
• Officer Chaney could input license-plate numbers, vehicle-identification numbers (VINs), driver’s license numbers, and names into the database from his patrol car;
• Officer Chaney could retrieve warrant information, vehicle information (including whether it had been stolen and whether it was insured), and information on the driver’s license or identification card’s validity from the database;
• Officer Chaney estimated that he had used the database “[t]ens of thousands” of times;
• Based on Officer Chaney’s experience, the database was very accurate;
• Officer Chaney estimated that he had received an “unconfirmed” insurance return “probably 20 percent” of the time, “give or take[,]” and “[h]undreds, if not thousands” of times;
• Officer Chaney estimated that he . had discovered that someone had insurance despite receiving an “unconfirmed” return a “handful” of times; “[vjery, very seldom”; and “less than ten percent of the time”; and
• Officer Chaney admitted the possibility that other “unconfirmeds” could be shown to be false later when a ■person cited for not having insurance showed proof of insurance to a judge.
The parties agreed that how the database works is controlled by law.
C. The Trial Court’s Findings of Fact and Conclusions of Law
At the end of the suppression hearing, the trial court delivered oral findings of fact and conclusions of law. The trial court found:
• “[Appellant] was traveling in a motor vehicle on a public road [i]n December of 2015 and drove past Euless Officer Chaney operating a vehicle in a public place”;
• “[Appellant] had every lawful right to do so”;
• “[Appellant] was observed driving by Officer Chaney”;
• “Officer Chaney was a traffic enforcement officer”;
■ • “[Officer Chaney] observed the license plate number on [Appellant’s] 2004 Jeep Grand Cherokee and entered it into a state database on a car computer terminal”;
• “The database contained] FBI, [DPS] and TLETS official data which [included] among other information, records, as required by law, to be maintained concerning compliance with the Financial Responsibility Act, i.e., proof of insurance”;
• “[T]he terminal information came back as, quote, unconfirmed, unquote”;
• “[Unconfirmed, according to the testimony of this officer, and, quite frankly, the cases which the defendant lost the motion to suppress on appeal versus won is confirmed and unconfirmed terminology used of whether insurance is confirmed or is not”;
• “[B]y stipulation there appears to at least have been issued on August 4th of 2015 a Geico Insurance policy which covered [Appellant] and [the Jeep] ... [, and the s]tated expiration date [is] six months later in February of 2016”; and
• “[T]here is no evidence that this information was provided to the officer at the time of the stop, which would have possibly given reason to cease any further temporary detention.”
Within his findings, the trial court stated,
I [would] totally understand the ... proper concerns raised by Defense Counsel of the absolute reliability of the database and the information contained[, i]f this were a trial for failure to maintain financial responsibility and what information and reliability must we trust before a computer entry gets a person convicted of driving without insurance. ... [A]s correctly pointed out by State’s Counsel this is a reasonable susp[ic]ion for detention to determine whether you do or don’t have insurance, and this officer’s experience is less than ten percent mistake. And the issue isn’t whether your client is guilty of driving without any insurance but whether a peace officer enforcing traffic laws had a reasonable suspicion to determine if, in fact, there was insurance. [Emphasis added.]
The trial court concluded,
Officer Chaney had reasonable suspicion to stop the vehicle for a detention for such reasonable period of time as necessary to confirm that there was no insurance or to correct the mistaken information displayed on the public data information terminal provided by state agencies as required by Texas law.
And, therefore, this Court finds the stop to be lawful as a matter of fact and law.
II. Discussion
A. Standard of Review
We review a trial court’s ruling on a motion to suppress evidence under a bifurcated standard of review. Amador v. State, 221 S.W.3d 666, 673 (Tex. Crim. App. 2007); Guzman v. State, 955 S.W.2d 85, 89 (Tex. Crim. App. 1997). We give almost total deference to a trial court’s rulings on questions of historical fact and application-of-law-to-fact. questions that turn on an evaluation of credibility and demeanor, but we review de novo application-of-law-to-fact questions that do not turn on credibility and demeanor. Amador, 221 S.W.3d at 673; Estrada v. State, 154 S.W.3d 604, 607 (Tex. Crim. App. 2005); Johnson v. State, 68 S.W.3d 644, 652-53 (Tex. Crim. App. 2002).
A trial court’s findings on a motion to suppress may be written or oral. State v. Cullen, 195 S.W.3d 696, 699 (Tex. Crim. App. 2006); State v. Varley, 501 S.W.3d 273, 277 (Tex. App.—Fort Worth 2016, pet. ref'd).
B. Reasonable Suspicion
A detention, as opposed to, an arrest, may be justified on less than probable cause if a person is reasonably suspected of criminal activity based on specific, articulable facts. Terry v. Ohio, 392 U.S. 1, 21, 88 S.Ct. 1868, 1880, 20 L.Ed.2d 889 (1968); Carmouche v. State, 10 S.W.3d 323, 328 (Tex. Crim. App. 2000). An officer conducts a lawful temporary detention when he or she has reasonable suspicion to believe that an individual is violating the law. Crain v. State, 315 S.W.3d 43, 52 (Tex. Crim. App. 2010); Ford v. State, 158 S.W.3d 488, 492 (Tex. Crim. App. 2005). Reasonable suspicion exists, when, based on the totality of the circumstances, the officer has specific, articulable facts that when combined with rational inferences from those facts, would lead him to reasonably conclude that a particular person is, has been, or soon will be engaged in criminal activity. Ford, 158 S.W.3d at 492. Reasonable suspicion is a lower level of suspicion than probable cause, and' probable cause “falls far short of a preponderance of the evidence standard.” Baldwin v. State, 278 S.W.3d 367, 371 (Tex. Crim. App. 2009). Reasonable suspicion is also an objective standard that disregards any subjective intent of the officer making the stop and looks solely to whether an objective basis for the stop exists. Ford, 158 S.W.3d at 492.
C. Stops Based on “Unconfirmed Insurance” Returns
Drivers in Texas must maintain proof of financial responsibility for their vehicles. See Tex. Transp. Code Ann. § 601.051 (West 2011). Operating a vehicle for which financial responsibility has not been established is .a misdemeanor punished by.a fine. See id. § 601.191 (West Supp. 2017). As this court pointed out in Swadley v. State, cases involving stops based on an officer’s database-derived suspicion that the driver may be committing this misdemeanor fall into two categories:
In the first category — the one upon which [Swadley] relies — the courts have, held an officer did not have reasonable suspicion where the evidence was not developed to determine the ambiguous answer’s meaning or reliability. In the second category — -the one upon which the. State relies — courts have held that reasonable suspicion existed when the officer, through experience or training, had additional' information about what the ambiguous answer from the data- . base meant and some idea regarding the data’s reliability.
No. 02-15-00085-CR, 2016 WL 7241564, at *5 (Tex. App. — Fort Worth Dec. 15, 2016, pet. refd) (mem. op., not designated for publication) (citations omitted). This court held that the facts in Swadley “f[e]ll in line with the second category of cases.” Id. at *6. The evidence in Swadley showed:
• To the officer who made the stop, an insurance return of “unconfirmed” from the database meant that the vehicle was not currently insured, based on his training and experience;
• The officer who made the stop had previously relied on the database’s insurance information in making “many traffic stops,” and based on his experience, the database was accurate and reliable; and
• The officer who made the stop knew from his experience with the database and other stops that an “unconfirmed” with a future expiration date “probably meant .,. that the vehicle had insurance” previously but it had lapsed. Id. at *T, *6.
Additionally, a sheriff’s office dispatcher testified that the database was updated weekly. Id. at *6. Finally, an admitted exhibit “verified” that the insurance readout .the officer saw “meant that the vehicle, at least at some point, had 'insurance but that at the time of the stop there was a question about whether the vehicle remained insured.” Id. That document,
entitled “The Texas Financial Responsibility Verification Program[,]” ... identified] four possible responses: (1) “Confirmed,” (2) “Unconfirmed,” (3) “Verify Manually,” and (4) “Multiple.” [The document explained that u]nder “Unconfirmed,” .the database will provide additional information: (1) “Vehicle coverage expired,” which contains the additional footnote that “(t)he system considers ve-hiele coverage expired when the policy has been expired for over 14 days()”;- (2) “No vehicle coverage found”; (3) “Vehicle last match not within 45 days”; (4) “Vehicle coverage expired; Vehicle last match not within 45 days”; or (5) “Vehicle coverage expiration unknown; Vehicle last match not within 45 days.” The third, fourth, and fifth examples carry the same footnote, which provides:.“ ‘Vehicle last match not with 45 days’ indicates the insurance company has not reported insurance for the vehicle and/or person in the past 45 days. This could be [because] the vehicle and/or person is no longer insured, or they, have changed companies and the new company has not reported insurance for them.”
Id. at *3. Our court therefore held that (1) the officer “had specific articulable facts from which he could draw the rational inference that [Swadley’s] vehicle had an insurance issue that [the o]fficer ... was authorized to investigate further” and (2) “the trial court' had enough information upon which to find the database sufficiently reliable” for the purposes of establishing reasonable suspicion. Id. at *6.
Similarly, the United States Court of Appeals for the Fifth Circuit has.joined other federal circuit courts in holding that “a state computer database indication of insurance status may establish reasonable suspicion ... as.long as there is either some evidence suggesting the database is reliable or at least an absence of evidence that it is unreliable.” United States v. Broca-Martinez, 855 F.3d 675, 678-79 (5th Cir. 2017). Specifically, the Broca-Martinez court held:
A state computer database indication of insurance status may establish reasonable suspicion when the officer is familiar with the database and the system itself is reliable. If that is the case, a seemingly inconclusive report such as “unconfirmed” will be a specific and ar-ticulable fact that supports a traffic stop. Id. at 680,
D. Analysis
1. Officer Chaney Had Sufficient Specific, Articulable. Facts from Which He Could Infer that Appellant’s Jeep Was Uninsured.
Appellant contends that his case belongs in that category of cases in which courts have held that the police lacked reasonable suspicion to stop based on the word “unconfirmed” and no other information. See State v. Daniel, 446 S.W.3d 809, 815-16 (Tex. App.—San Antonio 2014, no pet.); Contraras v. State, 309 S.W.3d 168, 173 (Tex. App.—Amarillo 2010, pet. ref'd), Gonzalez-Gilando v. State, 306 S.W.3d 893, 896-97 (Tex. App,—Amarillo 2010, pet. ref'd). Appellant concedes that Officer Chaney explained that an unconfirmed result from the database will also indicate whether insurance has expired and whether a policy was in effect in the previous 45 days and specify information on any past policy. But because Officer Chaney stated that he looks only for the words “unconfirmed” and “confirmed” and because the appellate record does not provide what other details the insurance return at issue here contained, Appellant contends that the record does not ¿stablish “specific, ar-ticulable facts upon which Officer Chaney could rationally infer” that Appellant’s Jeep was uninsured.
As the State points out, however, besides the “unconfirmed” return from the database, Officer Chaney had a great deal of experience with the database — he had used it “[t]ens of thousands” of times— enough experience to know that except for “a handful” of the “[h]un.dredsi if not thousands” of times he had received an “unconfirmed” return from the database, the return meant that the vehicle in question was not currently insured. We therefore hold that Officer Chaney had sufficient specific, articulable facts upon which to base his inference that Appellant’s Jeep was uninsured. See Oliva-Arita v. State, No. 01-15-00140-CR, 2015 WL 7300202, at *4 (Tex. App.—Houston [1st Dist.] Nov. 19, 2015, no pet.) (mem. op., not designated for publication).
2. The Trial Court Had Sufficient Information to Conclude the Database Was Reliable for Purposes of Reasonable Suspicion.
Appellant also argues that Officer Chaney’s testimony alone was insufficient to establish the database’s reliability, focusing on all the objective information about the database that Officer Chaney did not know. Appellant contends that this absence of information about the database also places his case in the category with Gonzalez-Gilando, 306 S.W.3d at 896-97, not in the category of Swadley, 2016 WL 7241564, at *6.
In Gonzalez-Gilando, the database returned “unavailable,” “not available,” or “undocumented” regarding liability insurance coverage when troopers and a deputy sheriff typed in the information for Gonzalez-Gilando’s vehicle. 306 S.W.3d at 894-95. The deputy testified with no explanation that the return led him to believe that the vehicle did not have insurance coverage, id. at 896-97, but one of the troopers testified that the return could have meant either that the vehicle was insured or that it was not insured, id. at 897 n.2. There was no
other evidence developing the source of the information comprising the database, explaining what was meant when insurance information was unavailable, explaining why such information would be unavailable, illustrating the accuracy of the database, establishing the timeliness of the information within the database, depicting how often those using the database were told that insurance information was unavailable, proving that the program through which the database was accessed was even operating at the time, and the like.
Id. at 897. Our sister court in Amarillo therefore held that the deputy’s inference was not reasonable and that he lacked reasonable suspicion for the stop. Id.
Relying on the holding in Gonzalez-Gi-lando, Appellant states tliat Officer Chaney’s “only knowledge of the database’s accuracy was with his own experience as a user” and contends that knowledge is insufficient to establish reasonable suspicion. We disagree, however, and hold that Gonzalez-Gilando is distinguishable. Officer Chaney explicitly testified:
• He had used the database “[t]ens of thousands” of times;
• Only “a handful” of the “[hjundreds, if not thousands” returns of “uncon-l firmed” he received from the database were in error;2 and
• The database was “very” accurate based on his experience.
Additionally, the parties stipulated that how the database works is controlled by law.
We note that it would be helpful to have objective information about the database— how it worked, the timeliness of the information placed in it, and the error rate, for example — and essential if we were reviewing a conviction based on information yielded from the database rather than whether a police officer with a great deal of experience using the database had reasonable suspicion to stop a motorist based on a return of “unconfirmed” from that database. See id. However, in this case and under these limited and specific facts, we uphold the trial court’s implicit finding that Officer Chaney’s knowledge was sufficient to establish the database’s reliability for the purposes of establishing reasonable suspicion. We consequently hold that under these narrow facts, the trial court did not err by concluding that Officer Chaney had reasonable suspicion for the stop or by denying Appellant’s motion to suppress. We overrule Appellant’s sole point.
III. Conclusion
Having overruled Appellant’s sole point, we affirm the trial court’s judgment.
. NCIC stands for the National Crime Information Center system; TCIC stands. for the Texas Crime Information Center systém.
. Appellant argues that the accuracy rate does not meet the 98% rate prescribed by the applicable regulation. See 28 Tex. Admin. Code § 5.605(b) (2017) (Tex, Dep't of Ins., Data Error Correction Requirement for Insurers Using the Database Program). Our reading of Officer Chaney’s testimony indicates that of the estimated 20% of returns showing ''unconfirmed,” less than 10%, or less than 2% of the overall returns, was in error. Thus, this accuracy rate satisfies the correction requirement for insurers found in the regulation.