Health, Safety, and Practicality

This is one of four documents that describe your relationship with the Actual Vote app:

• The Actual Vote User Agreement sets out the terms of using the app.
• The Privacy Policy describes how we handle your data.
• This Health, Safety, and Practicality Statement covers safety in the field, health and communicable illness, what to expect from the methodology, and what we can and cannot guarantee.
• Using Actual Vote is a practical guide and tutorial you can read or consult at any time.

Some activities that Actual Vote users may engage in, like recording poll tapes at polling places on election night, meeting with local Audit Committees, and speaking with poll workers and election officials in person, takes them into public settings where the usual risks apply: physical hazards, communicable illness, awkward or hostile encounters, late hours, and unfamiliar locations. This statement explains how America Counts think about those risks, what we expect of ourselves, and what we ask of our users. It also explains the practical realities of how the Actual Vote methodology works and what we can and cannot guarantee about your experience with the app.

The principles below apply across elections, jurisdictions, and changing public health conditions.

Our approach

We approach health and safety the same way we approach the rest of our work: we follow the evidence, we respect the autonomy of our users, and we are honest about what we can and cannot control.

In practice, that means three things.

First, we design our methodology and operational practice so that nothing we ask of a user requires them to take an unreasonable risk. In fact, each user is in complete control over the extent to which they work with the app, and can choose to work well within their risk tolerance. For example, wherever a step is less risky when done remotely, asynchronously, or in smaller groups, a user can opt for that option. The only use case that is structurally required to happen in person is the on-site recording of poll tapes on election night, and even there, each user makes their their own decision as to whether they want to participate, when to go out, and how many locations to visit.

Second, to the extent we are able to, we update our recommendations as conditions change (we are a small organization). This can include areas like public health guidance, local political climate, weather, the legal landscape around poll observation, and things that affect what it is like to be in the field. We communicate this through our website.

Third, we treat the question of how to manage personal risk as belonging to the user, not to us. Users know themselves, their households, and their communities better than we do, and we trust them to make informed decisions on that basis. Our role is to give users the information they need to decide.

Practicality

This section is the orientation for new users: what Actual Vote does, what you should know before your first recording, what the methodology is and isn’t designed to do, what to expect from the experience, and why the work matters. We recommend reading the whole section before your first time in the field.

Welcome to Actual Vote

Welcome to Actual Vote. You are using a tool that lets Americans independently verify how their local votes are reported in elections, by recording paper poll tapes with their phones and contributing the recordings to a public archive used for audit work. This is meaningful, useful work, and we are glad to have you.

Here is what Actual Vote does, in plain terms. After polls close on election night, voting machines at each polling place print paper “poll tapes” that show the local vote totals — how many votes each candidate received from that specific machine, in that specific precinct, for that specific contest. In many jurisdictions, those tapes are then posted publicly at the polling place where anyone can see them. Any time you can access poll tapes, Actual Vote lets you record them with your phone and upload the recordings to America Counts. We then compare the totals on the tapes against the officially reported results and check whether the count was reported correctly. Most of the time it was. Sometimes there are discrepancies that need to be investigated. Either way, your recordings become part of an independent audit of vote reporting in your community.

Things to know before your first recording

Before you submit your first recording, there are three things you should understand clearly. They appear in the consent notice when you first open the app. They are worth knowing in more depth.

Your recordings will be public

Your video recordings are reviewed by our experts, who review and approve them (unless they have anything personal or inappropriate, in which case they reject them). Approved videos are published on the internet at av.democracycounts.org, where anyone in the world can view them. They stay there indefinitely. This is not an accident or a side effect — public access is the point of the methodology. The audit only works if the recordings are independently verifiable by anyone, and that requires them to be public.

When you submit a recording, you should assume that anyone could see it. Your neighbors, journalists, election officials, opposing campaigns, your employer, members of your family, complete strangers on the other side of the world — anyone with an internet connection can find the recording, watch it, and download a copy. Search engines may index it. The recording becomes part of a permanent public record.

This is genuinely a feature of how Actual Vote works, not a problem to manage. The whole point is that the recordings are independent evidence that anyone can verify, which is only possible if they are publicly accessible. But it does mean you need to think about what you put in a recording before you submit it.

Never include personal information in a recording

Because approved submissions become public, you should never include personal information of any kind in a recording — yours or anyone else’s. This is the single most important rule of using Actual Vote, and it is worth being very specific about what it means.

Don’t speak while recording. Don’t say your name. Don’t say your address, your phone number, your date of birth, your social security number, or any other identifying information. Don’t talk about your health, your finances, your relationships, or anything else personal. Don’t have a casual conversation with anyone while you are recording. Even something you say to a friend standing next to you, or something a friend says to you, will be captured in the audio and could end up in the public archive.

Don’t record other people. Don’t capture other people’s faces, voices, or conversations. Don’t record license plates. Don’t record children. Don’t record any documents, papers, or screens that might contain personal information. The safest framing shows only the poll tape, with nothing else in the shot and no audio other than ambient sound.

The vetting process is not perfect. Our staff reviews every submission before it is published, and we try our best to catch personal information before it goes public. The work is done by humans, with all the human attention and care we can bring to it. But we cannot listen to every second of audio on every recording, and mistakes happen. If you put personal information in a recording, there is a real chance it will end up in the public archive accidentally — and once something is in the public archive, removing it does not retrieve copies that anyone else has already downloaded.

The reliable protection against personal information in the archive is at the recording stage, by you, not at the vetting stage, by us. Silent recording of just the poll tape is the safe practice. If you stick to it, you have nothing to worry about. If you don’t, you are taking a risk that we cannot fully manage on your behalf.

Your recordings are evidence

The recordings you submit through Actual Vote are not casual cellphone videos. They are forensic evidence about how votes were reported in your community — contemporaneous, location-stamped, time-stamped, and processed through a documented chain of custody that allows them to be authenticated and used in legal proceedings if it comes to that.

Most recordings will never be involved in any legal proceeding. They will be part of routine audits, contribute to the public archive, and serve their purpose without ever entering a courtroom. This is the normal case, and it is where most of the value of Actual Vote actually lies. But in a small minority of cases, a recording can become the basis for a legal proceeding — a regulatory complaint to a state oversight body, an election contest in state court, a public records lawsuit, a federal civil rights case, or a criminal investigation.

If a recording you made becomes part of such a proceeding, you could be asked to confirm under oath that you took the video. The questioning would typically be brief — sometimes just a single question: “To the best of your knowledge, did you shoot this video?” You are not at any legal risk in this situation. You are a witness, not a defendant. You did nothing wrong by recording a publicly posted poll tape, and confirming that you did it is not an admission of anything problematic. The only real cost to you would be the time and inconvenience of complying with the subpoena — possibly a trip to a courthouse, possibly a brief appearance, then home again.

The “Your recordings as legal evidence” section later in this statement covers this dimension in more detail. We mention it here because we want you to know it exists before you start making submissions. Most users will never encounter this situation, but you should not be surprised if you do.

What Actual Vote is and isn’t

It is worth being clear about the scope of Actual Vote, because many of the most common user frustrations come from misunderstanding what the methodology is and isn’t designed to do.

What Actual Vote is

Actual Vote audits vote reporting — the part of an election that runs from when votes are tabulated by voting machines through when the official results are certified by the local canvassing board. This includes:

• The voting machine tabulating the votes cast at a specific polling place and printing the totals on a paper poll tape.
• The poll tape being delivered to (or read by) the elections office.
• The elections office aggregating the totals from all polling places and voting methods.
• The aggregated totals being published as official results.
• The canvassing board reviewing and certifying the results.

Actual Vote checks the integrity of this chain by recording the paper poll tape (the local-level output) and comparing it against the centrally-reported totals (the central output). If the two match, vote reporting was done correctly. If they don’t, something happened in the chain that needs investigation.

What Actual Vote isn’t

Actual Vote does not audit:

• Voter registration and eligibility. Actual Vote does not check whether the right people voted or whether people who should have been able to vote were prevented from doing so.
• Ballot interpretation. Actual Vote does not check whether the voting machine correctly interpreted the marks on individual ballots.
• Voting machine internal accuracy. Actual Vote does not check whether the machine got the right answer when it tabulated the ballots fed into it.
• Voter suppression. Actual Vote does not provide evidence about whether eligible voters were prevented from voting.
• Absentee, mail-in, or early voting integrity. These have their own auditing mechanisms that operate separately from Actual Vote.
• Recount procedures. Actual Vote can inform a recount but does not itself conduct one.

If your concern about election integrity is one of these other things, Actual Vote will probably disappoint you. The methodology is designed to be narrow and specific, and the narrowness is a feature — it is what makes the findings defensible — but it means Actual Vote is not a general-purpose election integrity tool.

Why the narrow scope is a feature, not a limitation

The narrow scope of Actual Vote is the reason it produces specific, verifiable, defensible findings. Methodologies that try to address everything usually do nothing well. Actual Vote does one thing — independent verification of vote reporting — and it does that one thing rigorously enough that the findings are worth taking seriously. The rigor is possible precisely because the scope is narrow.

The narrow scope also means that most of the value Actual Vote produces is in routine confirmation that vote reporting was done correctly. This is not glamorous, but it is genuinely valuable. Public confidence in elections depends on the routine, ongoing verification that things are working — not on dramatic discoveries of malfeasance. Actual Vote contributes to that routine verification at a level that very few civic tools achieve.

What to expect from Actual Vote

This subsection covers the practical realities of using Actual Vote, including the things we cannot guarantee and the kinds of frustrations that have surprised users in the past. Reading it before your first time in the field will save you time, disappointment, and unrealistic expectations.

We can’t guarantee that poll tapes will be where you expect them

The Actual Vote methodology assumes that paper poll tapes are printed by voting machines after the polls close and posted in a public location at the polling place where they can be viewed and recorded. In most jurisdictions, this is the practice. But poll tape posting practices vary, and we cannot guarantee that the tapes will be where you expect them when you arrive.

Common situations:

• The tapes weren’t posted yet when you arrived. Polls sometimes close late if there are still voters in line, and the closing procedures take time after that. You may need to come back later. We cannot tell you exactly when the tapes will go up.
• The tapes were posted but you can’t reach them. Sometimes the tapes are posted on the inside of a window of a building that is locked after polls close, or in a location that has been secured for the night. You may be able to record through a window with difficulty, or you may be locked out entirely.
• The tapes were posted but key information is obscured. Tapes can be torn, damaged by weather, or partially obscured by other materials. Some of the information you need may not be readable.
• The tapes weren’t posted at all. Sometimes poll workers forget to post the tapes, or are improperly instructed not to post them, or the tapes get blown away or vandalized after posting.

When any of these things happen, your trip to that polling place will not produce a usable recording. This is not a failure of the methodology and it is not something we can prevent. It is a feature of the operational reality of US elections, which vary substantially in how carefully poll tape procedures are followed. We track these incidents and work over time to improve procedures in jurisdictions where they are not being followed properly, but for any specific election in any specific jurisdiction, we cannot guarantee a clean experience.

We can’t guarantee that the app will work flawlessly on every device

The Actual Vote app is built for both iOS and Android, and we test it on a range of devices, but mobile devices and operating systems are diverse and bugs can affect specific combinations we have not yet encountered. If the app fails on your device in the field, you have a fallback: you can use your device’s built-in camera app to record the poll tape and email the recording to us. It is less convenient and the metadata will be less complete, but the recording will still be usable.

We strongly recommend that you download the app and take a practice recording before your first real attempt. The app supports practice recordings of a sample poll tape that we provide, so you can confirm that the upload pipeline works on your device before election night, when the stakes are higher and you do not have time to troubleshoot.

We can’t guarantee that we’ll analyze your specific submissions in time

America Counts is a small organization. We process submissions as quickly as we can, but during a busy election period we may have more submissions than we can fully transcribe and analyze before the election is certified. When that happens, we have to prioritize. The factors that affect prioritization include:

• The strategic importance of the jurisdiction (some races and some states are higher priority than others for the methodology).
• The completeness of the per-jurisdiction setup (we need to have the comparison logic for the specific county and election in place before we can analyze submissions from there).
• Whether the submitter has been in touch with us in advance and we have agreed to prioritize their work.
• The total volume of submissions and the resources we have available.

If you submit recordings without being in touch with us in advance, and we are swamped with other work, your submissions may be analyzed later than you expect, or may not be analyzed at all in time for the immediate election. The recordings will still be in the public archive, where you (or anyone else) can perform the analysis manually using the official results.

If you want a guarantee that your specific submissions will be analyzed in time for a specific election, contact us in advance and we can discuss whether to make a specific commitment.

We can’t guarantee that anything specific will result from your work

Most submissions to Actual Vote produce one of two outcomes. Either the audit confirms that the official results were reported correctly (the most common outcome), or the audit identifies an initial non-match that gets investigated and resolves — through AC’s own review or a cooperative ask to the local elections office — to a match or a confirmed discrepancy. In a small minority of cases, a discrepancy survives investigation, remains apparent (no benign explanation found), and becomes the basis for further action — a regulatory complaint, an election contest, a public records lawsuit, or some other proceeding.

We cannot promise that your specific submissions will produce any specific outcome. We cannot promise that they will lead to media coverage, regulatory action, or legal proceedings. We cannot promise that any proceedings that do result from your submissions will succeed. We cannot promise that any successful proceeding will lead to restated official results.

Most of the value Actual Vote produces is upstream of the dramatic outcomes — it is in the routine confirmation, the public-archive transparency, the cooperative engagement with elections offices, and the gradual improvement of vote reporting procedures over time. The dramatic outcomes are real but rare, and you should not measure the value of your participation by whether they happen.

What you should and shouldn’t expect to feel

Most users find Actual Vote satisfying because they are doing something concrete and useful for their community’s elections. They walk away from a successful evening of recording with the sense that they participated in something meaningful.

Some users come in expecting more — expecting that their recordings will catch fraud, expose corruption, or vindicate a particular political view. Actual Vote will probably disappoint these users. The methodology is designed to be neutral and to produce honest findings, which means that most of the time it produces unremarkable confirmations rather than dramatic revelations. If you are looking for a tool that will validate a pre-existing belief about an election, Actual Vote is not that tool. If you are looking for a tool that lets you contribute to honest, independent verification of how your community’s votes were reported, Actual Vote is that tool.

We say this not to discourage anyone but to make sure you start with the right expectations. The users who get the most out of Actual Vote are the ones who value the process for what it actually is, not for what they might wish it to be.

Why this matters

It is worth saying explicitly why this work matters. The practical content above can feel dry, and the expectation-management subsection is necessarily sober. Neither captures what is actually at stake.

Elections are the most basic act of democratic legitimacy. The people who vote decide who governs, and the people who govern derive their authority from the consent of the governed expressed through that vote. For this to work — for democracy to actually function as a system of consent rather than a ritual that papers over something else — the vote has to be counted correctly and reported honestly. This is not a small thing. It is the foundation that the rest of the system rests on.

In the United States, vote counting and vote reporting are decentralized across thousands of local jurisdictions. Most of those jurisdictions do their work carefully and well. The people who run elections are mostly conscientious public servants who take their responsibilities seriously. But the system is also opaque — the local-to-central reporting step that Actual Vote audits has historically been almost entirely invisible to the public, conducted by local officials whose work is rarely independently verified. When something goes wrong, it usually goes wrong quietly, and the public never finds out.

Actual Vote opens that step to public scrutiny. When you record a poll tape, you are creating a piece of evidence that did not exist before — an independent record of what the local-level vote totals actually were, captured at the time and in a form that anyone can verify. When we compare your recording against the official results and find that they match, we are publishing positive evidence that vote reporting in that location was done correctly. That positive evidence is the most common outcome and it is genuinely valuable: it builds the public confidence in elections that the system depends on. When we find a discrepancy, we surface it through the appropriate channels — starting with a cooperative ask to the elections office, and if unresolved, handing it to a court-capable party. Either way, the system becomes more transparent than it was before.

Most of what makes democracy work is the routine, unglamorous work of people doing their part. Showing up to vote. Serving as a poll worker. Sitting on a canvassing board. Filing public records requests. Reporting on local elections in the local newspaper. Asking questions at town meetings. Actual Vote is one of those routine, unglamorous contributions, and it is exactly the kind of contribution that democratic systems need more of. When you record a poll tape, you are doing your part.

You will not get famous from this work. You will not get rich from it. Most of the time you will not even get visible recognition for it. What you will get is the knowledge that your community’s elections are a little more transparent and a little more verifiable because you participated. That is enough, and it is worth doing.

Field safety

Recording poll tapes with Actual Vote often involves being in public, sometimes at night, conducting an activity that most members of the public are not familiar with. Three categories of concern apply: the ordinary risks of being out in public in the locations you plan to visit, the additional risks that apply at night, and the specific risks of engaging publicly in election-auditing activity in a politically charged moment.

The following precautions address the first two categories. The third is covered in the next section.

Make a plan and share it. Before election night, decide which polling places, vote centers, elections offices, or other locations you intend to visit and in what order. Share that plan with someone — a friend, a family member, your Audit Committee — along with rough timing. Establish a check-in cadence; a quick text on arrival and departure at each location works well. Be especially aware that if you plan to visit many locations on election night, you are likely to be working late into the night or early into the morning at locations that are dark and deserted.

Work in pairs or teams when you can. This is the single biggest thing that improves field safety. A second person can spot poll tapes, manage lighting, navigate, and serve as a witness if anything goes wrong. If you are working alone, your check-in cadence is doing the work that a partner would otherwise do.

Be aware of your surroundings. Park in well-lit areas where possible. Keep your phone charged. Notice who else is around. If anything at a location feels too risky, forget about Actual Vote and leave. There is no recording that is worth taking outside your risk tolerance.

Bring what you need. A bright flashlight is essential at night — both for safety and for illuminating poll tapes in poorly-lit areas. A portable battery is worthwhile if you are visiting many locations, since cold weather and prolonged camera use both drain phone batteries faster than usual. If the weather is bad, bring rain protection for yourself and your device.

Drive carefully. Election night work often involves long hours, late driving, and unfamiliar routes. Plan rest stops if your route is long. Do not drive if you are too tired to do so safely.

Don’t record at a polling place during voting hours unless you have explicit permission from poll workers. Recording poll tapes with Actual Vote is legal in most jurisdictions, but the legal context for recording inside a polling place while polls are open is more complicated, and our methodology does not require it. The Actual Vote process is designed around tapes that are publicly posted after the polls close.

Interactions in the field

Most of the people you encounter while using Actual Vote will be friendly, indifferent, or curious. Some will be skeptical. A small number may be hostile. The following sections cover each.

Poll workers and election officials. If a poll worker or election official tries to stop you from video-recording poll tapes that are posted in plain public view — an activity that is legal in most jurisdictions — it is up to you how much to push back. If there is any doubt, walk away. Please let America Counts know if this happens. It helps us understand local election administration culture and advocate for clearer, safer, more consistent poll tape procedure.

Curious bystanders. If you explain Actual Vote to a curious onlooker, it will very often happen that they accept the explanation and thank you for what you are doing. In those cases, don’t forget that you can encourage them to download Actual Vote and use it themselves on the spot.

Hostile encounters. Unfortunately, the political climate in the United States today is fraught, and instances of people questioning election results — sometimes for good reasons and sometimes for bad — are common. Any election-auditing activity, no matter how rigorous or non-partisan its methodology, can appear suspicious to certain onlookers. This includes recording poll tapes with Actual Vote, which is part of a non-partisan election-transparency activity that is structurally hard to abuse. The methodology is unfamiliar to many people, and some will not accept any explanation of it. In some cases, an onlooker may assume the activity must be part of a partisan ploy by a party they oppose and may angrily confront or attempt to intimidate the Actual Vote user. There is no way for America Counts to prevent such situations. If one happens to you, forget about the app and do what you would normally do in such a situation — which will usually be to leave as quickly as possible.

Law enforcement. It is possible, though uncommon, for police to be called on an Actual Vote user recording poll tapes — usually by a bystander who assumes the activity must be illegal. If law enforcement engages you, be calm, be polite, identify yourself, and briefly explain what you are doing. Recording poll tapes posted in public view is legal in most jurisdictions, but we recommend that users comply with any reasonable request from law enforcement to stop recording or leave, even if the underlying activity is legal. Note the responding officer’s badge number and agency if you can do so safely, and let America Counts know afterward so we can support you and track the incident.

Aftercare. If you have a tense encounter in the field, give yourself permission to decompress before continuing. Talk it through with someone — a partner, a fellow committee member, or America Counts. This work is meaningful precisely because it touches a high-stakes part of public life, and the flip side of that meaning is that the occasional difficult interaction can land harder than it would in another context.

Example scripts

Some users find it useful to have language ready for the most common questions. The following scripts assume you are using Actual Vote to record publicly posted poll tapes at a vote center after the polls have closed. As you would expect, the later it gets, the fewer poll workers, officials, and bystanders are likely to be around.

“Hey, what are you doing? I’m intrigued.”

I’m using Actual Vote to video-record poll tapes — here, look. America Counts will take all the vote totals on the videos and compare them to the officially reported results that come out in a week or two, and investigate any discrepancies they find. So by taking these videos, I’m helping to improve the transparency and accuracy of this election. You can do it too if you want — just go to americacounts.us or search for Actual Vote in the App Store or the Play Store to download the app.

“The polls are closed now. Who are you and why are you filming the poll tapes? Who gave you permission?”

Hi, my name is [name]. I’m a resident of this county, and right now I’m exercising my right to observe and film the poll tapes here that are in plain public view. I don’t need special permission. The activity is non-partisan and structurally hard to abuse for partisan gain. The polls closed hours ago, and there are no voters or poll workers around — nothing about what I’m doing out here in public could pose a threat to either group. The goal is to double-check that the vote reporting was done correctly. In almost all cases, the process confirms that poll workers and election officials have done their job to a high standard. Every once in a while, we identify a discrepancy worth investigating.

“Why bother recording poll tapes? Why not just wait for the official results to be posted on the county website?”

Vote reporting in the United States is vulnerable to error and fraud. If error or fraud takes place in the vote reporting in your county, the official results may not match the totals on the poll tapes. Many counties have internal audit procedures, but those are no substitute for independent audits — for the same reason you would not let a corporation audit its own books or an airline certify its own planes. When an Actual Vote analysis confirms that the official results are correct, that is not a waste of time. It increases the transparency of the election by adding independent evidence of the correctness of the result.

“You’re saying our county is incompetent.”

No. I’m sure most of the poll workers and election officials here are professionals of the utmost integrity. It is just that the stakes of US elections are very high, and vote reporting is always vulnerable to error and fraud. That is why a process like vote reporting needs independent scrutiny — the same independent scrutiny you would demand of the surgeon who removes your appendix or the pilot who flies your plane. Our hope is that the Actual Vote analysis here ends up making this county look good by verifying that the reporting was done correctly.

“This is election tampering. Stop recording and delete the videos or I’m calling the police.”

Okay, I’m leaving. But to be clear, recording poll tapes posted in public view is legal, and it is not election tampering.

In a tense conversation, be friendly but assertive. If the conversation stops being productive, or if at any point you feel unsafe, drop the conversation and leave.

Health and communicable illness

Our methodology often brings users into contact with strangers in indoor and outdoor settings — polling places, elections offices, audit committee meetings. The ordinary risks of communicable illness apply.

We follow the current guidance of the U.S. Centers for Disease Control and Prevention (CDC), the World Health Organization (WHO), and applicable state and local public health authorities, and we update our internal recommendations as that guidance changes. When public health conditions in a given jurisdiction warrant additional precautions — a local outbreak, a high-transmission period, an emerging variant — we communicate this to the affected Audit Committees and adjust our operational practice accordingly.

We ask users to do the following before any in-person America Counts activity.

Check the current public health guidance for the area where they will be working, including the local prevalence of COVID-19, influenza, and other respiratory illnesses, and adjust their participation accordingly.

Stay home if they are symptomatic or have had a recent known exposure to a contagious illness. No America Counts activity is important enough to override this. We design our operational practice on the assumption that some fraction of any committee’s users will need to drop out on short notice for health reasons, and we plan accordingly.

Follow whatever precautions — vaccination, masking, distancing, ventilation, hand hygiene, rapid testing — they judge appropriate to the current public health situation and to their own risk profile. We trust users to make this judgment for themselves and do not require any particular precaution as a condition of participation.

Privacy in your recordings

Actual Vote video-recordings that you submit are placed in a queue to be vetted by America Counts. Recordings that contain poll tapes and nothing inappropriate or private in the audio or video pass the vetting process and become available to be viewed by the public. America Counts then transcribes the numbers on the videos and compares them with the official results, but anyone else in the world could do this too — the videos and the official results are freely available on the internet.

This is why we designed the vetting process to reject recordings that contain no poll tape content (like an accidental recording of someone’s shoes) or recordings that contain sensitive information beyond the poll tape (like someone saying their address out loud during a conversation). We strive to do this vetting accurately and quickly, but we cannot guarantee perfection in a human-run process. On rare occasions, a recording that should have been rejected is approved. These mistakes are typically caught during analysis, at which point the recording is set to “rejected” and removed from public view.

To avoid privacy problems, we ask that you take a few precautions when recording.

If you do not want your face associated with the video, keep it out of the shot — and check that it is not visible as a reflection in any glass you are filming through.

If you do not want your voice associated with the video, do not speak while recording. If you are fine with your voice being captured, feel free to speak — some users announce the precinct code at the beginning of each recording, which is genuinely helpful. But avoid saying anything private (your name, your address, health information, phone number, anything personal) on the recording.

Be careful about capturing other people’s faces, voices, and conversations. If you are recording alone at night, you may be the only one around. In a busier setting — a county elections office, for instance — you may want to ask that people stay quiet or step away while you record.

Some Actual Vote efforts involve teams of users at a table recording many poll tapes in succession at a county elections office. The work can be tedious, and conversation helps pass the time, but please take care that the conversation does not drift into personal details. In one such situation, a user was captured at the table discussing their private health information, and we were forced to reject the affected recordings.

We read vote totals off the video portion of the recording, so audio is not strictly required for analysis. You might ask whether we could capture only video. The answer is that this is not at all straightforward from a software perspective, and the presence of audio in a recording can also provide useful evidence of authenticity if it is ever challenged.

It is fine to record the outside of a vote center — for instance, to show where the poll tapes were posted, or where they should have been posted but weren’t. Avoid recording other people or car license plates.

If a recording is rejected, your effort is not necessarily lost. If a rejected recording contains poll tape images that are critical to an analysis, America Counts may edit it and release a “cleaned up” version that does not contain whatever caused the rejection.

Your recordings as legal evidence

During the Actual Vote analysis process, America Counts treats publicly available video-recordings of poll tapes taken by people like you as evidence about how the official vote reporting should turn out.

If the data on the poll tapes is consistent with the official reporting, your recordings constitute independently-collected evidence that increases certainty that the vote reporting was done correctly. If we find a discrepancy between the recordings and the official results, your recordings are the basis for asking election officials to provide an explanation. If the discrepancy has no benign explanation and persists after that stage, your recordings may inspire someone to mount a legal challenge to the official results.

From a legal standpoint, an Actual Vote analysis is at best an informal analysis on its own. Evidence of an apparent discrepancy cannot, on its own, force anyone to do anything. But if someone does initiate a legal challenge based on this evidence — especially in a race where the outcome is in question — your video-recordings will be submitted as legal evidence and will be analyzed carefully by the defense, who will want to independently confirm or disprove their authenticity and meaning.

In the unlikely event the process gets this far, there is a small chance that the defense could subpoena you to testify in court that you took the recording. The question would simply be: “To the best of your knowledge, did you shoot this video?” This verifies your role in the chain of custody. You have no legal risk in this situation, though complying with a subpoena would cost you some time and inconvenience. If you took the recordings in the county where you live, the courthouse will be relatively close. If you took them elsewhere, your trip may be longer.

If you are subpoenaed in connection with Actual Vote evidence, please contact America Counts right away. We will support you through the process to the extent we can.

It is worth remembering what is happening at this point. Your part in taking the recordings, generating the court case, and testifying that you truly shot them is a real contribution to democracy. On one hand, the evidence in your recordings could be part of forcing official results to be restated — a meaningful win for the accuracy and transparency of the race in question. On the other hand, the existence of the court case demonstrates that civically engaged people are on the job, conducting non-partisan independent audits of vote reporting with actionable results — results that create a powerful incentive for bad actors to think twice before attacking vote reporting, and for election officials to keep doing their work carefully.

What we cannot do

We are candid about the limits of what we can offer.

We cannot guarantee your physical safety in the field. Field work carries inherent risk and we cannot eliminate it — only help you minimize it within reason.

We cannot guarantee that participation in America Counts activities will not result in exposure to COVID-19 or any other communicable illness. Any activity involving travel, polling places, or contact with strangers carries some inherent level of risk.

We do not control the behavior of poll workers, election officials, members of the public, or any other person a user will encounter during America Counts work.

We cannot resolve every location-specific health or safety concern that an Audit Committee may raise, though we will work with committees in good faith to find arrangements that are acceptable to their members. In some cases, the most appropriate arrangement will be for a committee to scale back, postpone, or decline a particular activity. We support that decision when committees reach it.

We cannot guarantee perfection in our human-run video vetting process. On rare occasions, a recording that should have been rejected will be approved. We catch these and correct them as quickly as we can, but we cannot promise zero errors.

We do not assume legal liability for illness, injury, or other harm traceable in whole or in part to America Counts–related activities. This is a structural feature of working with a free, all-user civic auditing tool — not a reflection of how seriously we take user welfare.

Scope and contact

This statement reflects U.S. legal and public health context and is written for U.S.-based users. Users using Actual Vote or other America Counts tools in other countries should follow the legal and public health authorities applicable in their own jurisdiction and treat this statement as a general statement of philosophy rather than specific guidance.

If you have questions about this statement, want to report something that happened in the field, or need support in connection with an Actual Vote-related legal matter, contact America Counts at info@americacounts.us.