AI Transparency & Policy

How PR Bot Uses Artificial Intelligence

1. Journalist Query Matching (Automated Decision-Making)

What It Does: Our AI systems analyze incoming journalist queries from various media opportunity platforms and match them to user profiles based on stated expertise, credentials, and experience.

Technology Used: Natural language processing (NLP) and semantic matching algorithms powered by enterprise AI providers.

Automated Decision: The system automatically determines whether a query is relevant to your profile based on expertise matching.

Your Control:

• You define your areas of expertise in your profile settings
• You can adjust relevance sensitivity to receive more or fewer matches
• You can review all matched queries before responses are sent (if enabled in settings)

2. Response Generation (AI-Assisted Content Creation)

What It Does: When a relevant journalist query is identified, our AI generates a draft pitch or response based on your profile data.

Technology Used: Enterprise AI services from industry-leading providers.

Training Data: The AI is trained using:

• Information you provide about your expertise and credentials
• Examples of your previous work or writing (if provided via LinkedIn, website, etc.)
• Your communication preferences and tone settings
• General knowledge from the AI provider's base models (not specific to you)

Your Control:

• You can disable “Automatic Pitch Sending” in settings to require manual approval for all pitches
• When automatic sending is disabled, pitches stay in “generated” status until you manually review and approve them
• You can edit AI-generated responses before sending them to journalists
• You can deactivate profiles entirely to stop all pitch generation

3. Quality Evaluation (Automated Quality Control)

What It Does: AI systems evaluate the quality of generated pitches for accuracy, relevance, and professionalism before sending.

Technology Used: Automated quality evaluation systems combined with human review.

Current Process: Every pitch is reviewed by our team before sending. This human-in-the-loop approach ensures accuracy, professionalism, and appropriate tone for each journalist interaction.

AI Model Providers & Data Retention

Third-Party AI Services

PR Bot uses enterprise AI service providers to generate content on your behalf. For a complete list of our sub-processors and detailed data processing information, please see our Data Processing Information page.

Your Rights Regarding AI

Right to Human Review (GDPR Article 22)

Under GDPR, you have the right not to be subject to decisions based solely on automated processing that produce legal effects or significantly affect you. While PR Bot uses AI extensively, you have:

• Right to Human Oversight: Currently, all pitches receive human review before sending. If automated options become available in the future, you can always require manual review via settings
• Right to Explanation: Request an explanation of how the AI matched you to a specific query
• Right to Challenge: Contest AI matching decisions you believe are incorrect by contacting support
• Right to Opt-Out: Deactivate your profile entirely to stop all automated processing

Right to Understand AI Decisions

You can request information about:

• Why a specific journalist query was matched to your profile
• What profile data elements influenced the AI's decision
• The confidence score of any AI-generated content
• Which AI model was used to generate specific content

Contact support@prbot.ai to request this information.

AI Limitations & Accuracy

Known Limitations

We are transparent about the limitations of AI systems used in PR Bot:

• Hallucinations: AI models may occasionally generate plausible-sounding but factually incorrect information. We use validation systems to minimize this risk.
• Context Understanding: AI may misinterpret subtle nuances in journalist queries or your expertise areas.
• Bias: AI models may reflect biases present in their training data. We actively monitor for and mitigate bias in our systems.
• Credibility Assessment: AI cannot verify the truthfulness of credentials or expertise claims you provide.

Our Quality Control Measures

To ensure AI-generated content meets quality standards, we:

• Manually review and edit every pitch before it is sent to journalists
• Verify factual accuracy against your profile data
• Ensure appropriate tone, professionalism, and relevance for each query
• Monitor feedback from journalists to improve accuracy
• Maintain continuous human oversight of all AI-generated content

Note: While we currently review all pitches manually, we may offer automated sending options in the future for users who prefer faster outreach. Human review will always remain available as a setting.

Bias Mitigation & Fairness

We are committed to ensuring our AI systems operate fairly:

Ongoing Monitoring

• Demographic Monitoring: We track whether our matching algorithms produce equitable outcomes across different user demographics (when data is available)
• Topic Diversity: We monitor to ensure AI systems don't systematically favor certain expertise areas over others
• Regular Audits: We conduct quarterly reviews of AI system performance and fairness metrics

Reporting Concerns

If you believe our AI systems have exhibited bias or unfair treatment, please report it to ai-ethics@prbot.ai. We investigate all reports and will take corrective action when warranted.

Data Used to Train AI Systems

Profile-Specific Training

To personalize AI-generated content for you, we use the following data:

• Professional Information: Credentials, job titles, company affiliations, expertise areas
• Public Content: LinkedIn profile, personal website content, published work (if linked)
• Communication Style: Tone preferences, writing style examples (if provided)
• Historical Performance: Previous successful pitches and journalist feedback

Third-Party AI Model Training

Critical Clarification: According to OpenAI, Anthropic, and Google's API Terms of Service:

Transparency in Communications

Journalist Disclosure Policy

When our AI systems communicate with journalists on your behalf, we include appropriate disclosures:

• Responses include a footer indicating they were facilitated through PR Bot
• Journalists are informed that content was reviewed by you (the expert)
• Your authentic credentials and contact information are provided
• We never misrepresent AI-generated content as purely human-written

AI System Updates & Changes

As AI technology evolves, we may update our systems. When we make significant changes to how AI is used in PR Bot, we will:

• Notify users via email at least 30 days before changes take effect
• Update this AI Transparency Policy
• Provide opt-out mechanisms if changes materially affect automated decision-making
• Allow users to review and update their AI preferences and settings

Security of AI Systems

We implement multiple security measures for AI systems:

• Input Validation: All profile data is sanitized before being sent to AI models
• Output Filtering: AI-generated content is scanned for inappropriate or harmful content
• Prompt Injection Prevention: Systems are protected against adversarial prompt manipulation
• Rate Limiting: AI API usage is monitored and rate-limited to prevent abuse
• Access Controls: Only authorized systems can access AI generation capabilities

Compliance with AI Regulations

EU AI Act Compliance

We are monitoring the implementation of the EU Artificial Intelligence Act and are committed to compliance with all applicable provisions, including:

• Transparency obligations for AI systems
• Risk management requirements for high-risk AI applications
• Technical documentation and record-keeping for automated decisions
• Human oversight and intervention capabilities

California AI Transparency (SB 942)

For California residents, we disclose that PR Bot uses generative artificial intelligence to create content based on your profile. You have the right to opt out of automated content generation and use the Service in manual mode only.

Exercising Your AI-Related Rights

To exercise any of the rights described in this policy or to ask questions about our AI systems:

• Email: ai-ethics@prbot.ai
• Subject Line: “AI Rights Request” or “AI Transparency Inquiry”
• Response Time: We will respond within 30 days

Changes to This Policy

We may update this AI Transparency Policy as our AI systems evolve or as new regulations are introduced. We will notify users of material changes via:

• Email notification to all active users
• Prominent notice in the application dashboard for at least 30 days
• Update to the “Last Updated” date at the top of this page