Training Data Privacy Policy

This policy explains what data DropHammer™ Technologies LLC (“DropHammer,” “we,” “us,” or “our”) collects to improve its voice technology, how that data is used, and what control you have over it. It is written in plain language for the people who actually use DropHammer — contractors, operators, and crew members working on job sites.

What We Collect (Two Tiers)

DropHammer collects two types of data to improve its voice recognition technology. They are separate, and you control each one independently.

Tier 1: Noise and Environment Data (Default: On)

What it is:

• Ambient noise level measurements
• Audio environment profile data
• Voice clarity metrics

What it is NOT:

This is NOT audio. These are numerical measurements — numbers and statistics about your work environment. They cannot be reversed into speech or used to identify you.

Why we collect it:

To improve voice recognition accuracy in the environments where you work.

How it works:

Measurements are processed on your device. Only the numerical results are sent to our servers. No audio leaves your phone for Tier 1 data.

How long we keep it:

Aggregated data is retained for three (3) years from the date of your last interaction with the Service, after which it is permanently destroyed. No personal identifier is attached — we cannot trace a measurement back to you.

How to turn it off:

Settings > Privacy > toggle off “Share Noise Data.” All app features continue to work the same way.

Tier 2: Voice Recordings (Default: Off)

What it is:

Raw audio recordings from voice commands you give DropHammer. This means the actual sound of your voice when you say things like “schedule a job for Tuesday” or “how much is the Johnson invoice.”

When it is collected:

Only while the toggle is ON. If the toggle is OFF, no voice recordings are collected. Period.

Why we collect it:

To improve DropHammer’s voice recognition accuracy in real-world work environments. Your recordings help our proprietary AI technology better understand your commands in the conditions where you actually use them.

What we do before using recordings for training:

1. Dual-layer PII redaction. Personal information (phone numbers, email addresses, physical addresses, and Social Security numbers) is automatically detected and removed from both the text transcript and the audio recording. PII segments in audio are replaced with silence before the recording is saved to our training infrastructure. No unredacted audio is ever stored.
2. Metadata stripping. Audio files are stripped of metadata that could identify you — device info, location, timestamps.
3. Random ID assignment. Recordings are assigned cryptographically hashed identifiers with no reversible connection to your account.
4. Source deletion. Original recordings are deleted within 30 days after features are extracted for training. A daily automated cleanup job enforces this retention limit.

What we NEVER do:

• Sell your audio to anyone. Not to advertisers, not to data brokers, not to other companies. Never.
• Share raw recordings outside DropHammer. Your audio stays within our training infrastructure.
• Use recordings for anything other than improving voice recognition. No marketing analysis, no behavioral profiling, no surveillance.
• Record when you have not spoken to DropHammer. We only capture audio from active voice commands — when you press the talk button or say the wake word. DropHammer does not listen to ambient conversation, phone calls, or anything else. The microphone is only active during a voice command session.
• Allow humans to listen to your raw audio. Training is automated. No DropHammer employee sits in a room listening to recordings.

How long we keep it:

Processed training features (mathematical representations of speech patterns, not audio) are retained for three (3) years from the date of your last interaction with the Service, after which they are permanently destroyed. Raw audio recordings are deleted within 30 days of collection. If you delete your account, raw recordings are deleted within 30 days and processed features are destroyed within three (3) years of your last interaction or upon account deletion, whichever is sooner.

How to turn it on:

Settings > Privacy > toggle on “Share Voice Recordings.” By enabling this toggle, you acknowledge: “I understand that while my raw recordings can be deleted at any time, anonymized features already incorporated into trained models cannot be individually removed.”

How to turn it off:

Toggle off anytime. Previously contributed data remains in the training set because it has already been anonymized and cannot be traced back to you. No new data is collected after you toggle off.

Call Recordings and Training Data

If a DropHammer customer enables Tier 2 voice data sharing, call recordings from the AI Receptionist may enter the training data pipeline. The following additional requirements apply:

Enhanced Disclosure

When Tier 2 voice data sharing is enabled, the AI Receptionist’s call disclosure is enhanced to include: “This call may also be used to improve our voice recognition technology.” This sentence is appended to the standard recording disclosure regardless of which consent tier (Tier 1, 2, or 3) applies to the call. Omitting this disclosure when call recordings flow into the training pipeline would be a material omission that vitiates the caller’s consent.

Consent Requirements

• Tier 1 and Tier 2 states: The enhanced disclosure (including the training data sentence) is played. Continuation of the call constitutes consent to both recording and training use.
• Tier 3 states (strict all-party consent): The enhanced disclosure is played, AND the caller must provide affirmative consent (press 1 or say “yes”) before any recording or training use occurs. If the caller declines recording, no audio is captured and no data enters the training pipeline.
• If Tier 2 voice data sharing is NOT enabled: The training data sentence is omitted from the disclosure. Call recordings are retained only for quality assurance and business record-keeping, not for training.

Processing

Call recordings that enter the training pipeline are subject to the same protections as all Tier 2 voice data:

1. Dual-layer PII redaction — personal information is removed from both transcript and audio before training use.
2. Metadata stripping — device info, location, timestamps, and caller phone numbers are removed.
3. Random ID assignment — recordings are assigned cryptographically hashed identifiers with no reversible connection to the caller or account.
4. 30-day source deletion — original call recordings used for training are deleted within 30 days after feature extraction.

Customer Responsibility

The customer (business account holder) controls whether Tier 2 voice data sharing is enabled. By enabling Tier 2, the customer acknowledges that call recordings from their AI Receptionist will enter the training pipeline and that the enhanced disclosure will be played to callers. The customer is responsible for ensuring this is consistent with any representations they have made to their own customers about call recording practices.

Why This Matters

DropHammer is built for real work environments, not quiet offices. Better voice recognition means fewer missed commands, faster response times, and smoother operations on the job site.

Every contribution helps DropHammer work better for you and for every contractor who uses it.

Your Rights

You have full control over your data at all times.

• View what we have collected. Settings > Privacy > “View My Data.” This shows you a count of recordings contributed, hours of voice data, and noise profiles shared.
• Delete all your data. Settings > Privacy > “Delete My Data.” When you request deletion: (1) Your raw voice recordings are permanently deleted within 30 days. (2) Your account-level training preferences and consent records are deleted. (3) Anonymized, aggregated features that have already been incorporated into trained models cannot be individually extracted or deleted — this is a technical limitation of machine learning, not a policy choice. We disclose this BEFORE you enable Tier 2 so you can make an informed decision.
• Change your mind anytime. Both toggles can be turned on or off at any time with no waiting period.
• No penalty for opting out. All app features work identically regardless of your privacy settings. You do not lose functionality, you do not get slower responses, and you are not shown ads or nag screens.
• Data portability and export. When you contribute voice data under Tier 2, your raw recordings remain YOUR data and can be deleted on request. However, the trained models, aggregated features, and analytical improvements derived from processing your data (combined with data from other contributors) are DropHammer’s proprietary intellectual property. Data portability rights apply to your raw recordings and account data, not to DropHammer’s proprietary models or training datasets.
• Contact us directly. [email protected] — a real person will respond.

Who Sees Your Data

• DropHammer engineering team. They work with anonymized, aggregated data for model training. They do not have access to raw recordings tied to individual users.
• Our AI training infrastructure. Automated processing pipelines handle audio. No human listens to raw recordings.
• Nobody else. We do not share data with advertisers, data brokers, analytics companies, social media platforms, or any other third party. We do not sell data. We do not trade data. We do not give it away.

Crew Member Privacy

If your employer uses DropHammer and you are a crew member (not the account owner), these protections apply to you specifically:

• Your employer CANNOT see your individual voice recordings. Training data settings are per-device, controlled by you, not by the account owner.
• Your employer CANNOT require you to opt in to voice recording sharing. This is your personal choice. If anyone tells you it is required, that is not true — contact us at [email protected].
• Your employer CAN see aggregated noise data for their job sites (average noise levels, equipment profiles), but they cannot see your individual contributions or identify which measurements came from your device.
• Your personal voice profile belongs to you. If you leave your employer, your voice profile improvements travel with your DropHammer account. Your former employer does not retain any data specific to you.

Incentive Program

After you have been using DropHammer for a few months, we may offer you the option to turn on Voice Recording sharing in exchange for benefits:

• 1 free month of Tier 3 Intelligence (our advanced analytics and predictive features)
• 25% discount on Tier 3 for as long as Voice Recording sharing remains on
• Better personal voice recognition — the more data DropHammer has from your voice specifically, the better it understands you in your work environment
• Improved noise handling — optimized for the types of work environments you use most

This offer is entirely optional. Declining it has no effect on your app experience. You can accept the offer, use the benefits, and turn off sharing later — you keep any free months already earned, but the ongoing discount stops.

Notice of Financial Incentive (CCPA)

This incentive program constitutes a “financial incentive” under the California Consumer Privacy Act (CCPA). The following disclosures apply:

• Good-faith estimate of data value: We estimate the value of contributed voice data based on the commercial cost of acquiring equivalent licensed training data for domain-specific voice applications, where high-quality labeled audio corpora are scarce. The value of the financial incentive is reasonably related to the value of the data provided by participants.
• Methodology: This estimate is derived from the market rate for commercially licensed voice training datasets of comparable quality and specificity.
• How to opt in: Enable the “Share Voice Recordings” toggle in Settings > Privacy. You will be presented with a clear disclosure of what data is collected and how it is used before opting in.
• How to opt out: Disable the “Share Voice Recordings” toggle at any time. You retain any free months already earned, but the ongoing discount stops when you opt out.
• Non-discrimination: The incentive is not unjust, unreasonable, coercive, or usurious. All app features work identically regardless of your participation in the incentive program. You do not lose functionality, receive degraded service, or face any penalty for declining to participate.
• Proportionality: The value of the financial incentive (one free month of Tier 3 Intelligence and 25% ongoing discount) is reasonably related to the value of the data provided, considering the cost of acquiring equivalent training data through commercial licensing.

Data Security

• All data is encrypted in transit (TLS 1.2+) and at rest (AES-256).
• Voice recordings are stored in isolated infrastructure with access controls limited to automated training pipelines.
• We conduct regular security audits of our training data infrastructure.
• In the event of a data breach affecting voice recordings, we will notify affected users in accordance with applicable breach notification laws, and will use our best efforts to provide notice within 72 hours, via in-app notification and email.

Children

DropHammer is designed for adult professional use in the construction industry. The Service is intended for users who are at least eighteen (18) years of age. We do not knowingly collect data from anyone under 18 years of age. If we learn that we have collected data from a minor, we will delete it immediately.

State-Specific Rights

California Residents (CCPA/CPRA)

You have the right to know what personal information we collect, request deletion of your data, and opt out of the sale of your data. We do not sell personal information. To exercise your rights, use the in-app privacy controls or contact [email protected].

Illinois Residents (BIPA)

Tier 2 Voice Recording collection is NOT available to Illinois residents or to businesses with a principal place of business in Illinois. The Illinois Biometric Information Privacy Act (740 ILCS 14/1 et seq.) imposes requirements on the collection and use of biometric information that are incompatible with our Tier 2 training data program. To protect Illinois residents and to eliminate any risk of non-compliance:

• Tier 2 is disabled for Illinois. If your account is registered to an Illinois address, or if you indicate during onboarding that you are an Illinois resident or that your business is principally located in Illinois, the Tier 2 “Share Voice Recordings” toggle will not be available to you. You will not be able to opt in to voice recording collection.
• Tier 1 is unaffected. Tier 1 noise and environment data consists of numerical measurements only. These are not audio recordings, cannot be reversed into speech, and do not constitute biometric identifiers or biometric information under BIPA. Tier 1 remains available to all users including Illinois residents.
• AI Receptionist calls. The AI Receptionist processes voice in real time for the purpose of conducting the call and generating a transcript. DropHammer does not extract, store, or use voiceprints, vocal pattern templates, or other biometric identifiers from AI Receptionist calls. Call recordings are stored as business records for quality assurance and are not used for biometric identification or voice model training. This applies to all users, including calls involving Illinois residents.
• Incentive program. The training data incentive program (discounts and free months in exchange for Tier 2 participation) is not available to Illinois residents or Illinois-based businesses.
• Interstate calls. For calls handled by the AI Receptionist where the caller is located in Illinois, DropHammer does not extract biometric identifiers from the call audio. The call recording and transcript are retained as standard business records subject to the retention schedule in this policy.
• If you move to Illinois. If you are a Tier 2 participant and subsequently move to or establish your principal place of business in Illinois, you must notify DropHammer at [email protected]. We will disable Tier 2 collection for your account and delete any raw recordings associated with your account within 30 days of notification.
• You may contact [email protected] for questions about how BIPA applies to your use of DropHammer.

Other States

We comply with all applicable state privacy laws. If your state provides additional privacy protections beyond what is described here, those protections apply to you. Contact [email protected] for state-specific questions.

Changes to This Policy

If we make material changes to this policy (changes to what we collect, how we use it, or who we share it with), we will:

1. Notify you in the app at least 30 days before the change takes effect
2. Clearly explain what is changing and why
3. Give you the opportunity to adjust your settings before the change takes effect

Minor changes (typo fixes, clarifications that do not change meaning) may be made without advance notice. The current version is always available at drophammer.ai/privacy.

Contact

Patent-pending technology. DropHammer is a trademark of DropHammer Technologies LLC.