1. Data Collection and Usage
Our AI technology and partner currently do not utilize facial recognition or any means to correlate images or data to specific individuals. As such, we ensure that no personally identifiable information (PII) is stored or tracked through our AI systems.

2. Data Minimization and Anonymization
To protect sensitive data, we adhere to the principles of data minimization and anonymization:

• Data Minimization: We limit the amount of information provided to our partners and AI models to only what is necessary for their function and learning Cycles.

• Anonymization: There is no personally identifiable information (PII) provided to our partners or AI models, ensuring that no identifiable information is processed.

3. Data Privacy Requirements
We continuously measure and audit our institution’s data privacy requirements, where applicable, against regulatory obligations such as SOC2, GDPR, HIPAA, and other industry-specific standards. These compliance regulators allow us to maintain and update our data responsibly.

4. Federated Learning Approach
PB&J TV and its Ai partners deploy large language models (LLMs), in conjunction with, generative AI models on our own servers and in a private cloud environments. Our use of federated learning techniques enables model training on decentralized data sources without centralizing sensitive data. In this approach:

• The model is sent to the data sources for local training.

• Only model updates are shared, not the raw data, ensuring that sensitive information remains secure.

5. Differential Privacy
Our methodology applies differential privacy mechanisms to the data or the model itself. This involves adding noise or randomness to data points or model parameters to protect individual data while still allowing for meaningful insights.

6. Data Retention Practices

• Temporary Storage: Data collected through our AI systems will only be temporarily stored for customer reporting for a 30 day cycle and is not currently retained for extended periods.

• Model Training: Data used for model training will only be utilized within the same 30 day cycle for specific use cases, after which it will be purged.

7. Data Security
We implement robust security measures to protect data, including:

• Secure Transmission: Data sent to and from AI models is encrypted using strong encryption protocols. We use Virtual Private Networks (VPNs) or secure channels to protect data during transit.

• Red Team Testing: We periodically conduct “red team testing,” where ethical breaches are simulated on our systems to identify any vulnerabilities.

• Regular Compliance Audits: We periodically review and audit our systems to ensure compliance with data privacy regulations and internal policies.

8. User Rights and Transparency
We are committed to transparency regarding our data practices. If their questions not answered above regarding our Ai data policies, please reach out to contact@pbnjtv.com.

9. Policy Updates
We may update these policies periodically to reflect changes in our practices or regulatory requirements. Significant changes will be communicated to our users through appropriate channels.

PERFORMANCE.   AVAILABILITY.   RELIABILITY.

LET'S CONNECT!

M - F, 7am-8pm CST
S - S, 8am-4pm (email only)
1-877-722-6358
contact@pbnjtv.com