Providing incorrect or misleading information about consequences of non-payment

🟣 Definition

This violation is recorded when the collections agent provides inaccurate, exaggerated, or misleading information regarding the consequences of non-payment, which may result in misleading the customer or creating a false impression.

🟣 When to Flag

Flag this violation ONLY if one or more of the following conditions are clearly met:

1. The agent claims authority or actions that are not accurate or not within their actual authority, such as:

• “I can suspend your services”
• “I will block your account immediately”
• “All your services will be stopped now”

1. 
   
2. The agent provides incorrect information about official procedures or consequences.
3. The agent uses threatening language based on false or exaggerated consequences.
4. The agent attributes to themselves or their department powers they do not have.

🟣 When NOT to Flag (Very Important)

Do NOT flag this violation in the following cases:

1. The agent explains consequences accurately and in line with policies.
2. The agent uses neutral and conditional statements such as:

• “Necessary actions may be taken according to policy”
• “The case may be escalated as per procedure”

1. 
   
2. The agent does not claim direct authority over the action.
3. The information provided is factually correct and compliant.

🟣 Violation Examples

• “I can suspend your services”
• “I will block your account now”
• “Your services will be stopped immediately”

❌ If these statements are inaccurate → Violation

🟣 Indicators (Strong Evidence)

• Use of phrases like:
• “I will…”
• “I can…”
• “We will immediately…”
• 
  

⚠️ Especially when linked to actions that are not accurate or authorized

🟣 Final Decision Rule (IMPORTANT 🔥)

Flag this violation only when there is a clear and explicit statement containing incorrect or exaggerated information about the consequences of non-payment.

🟣 AI Safety Rule

Do not flag this violation unless there is clear textual evidence of inaccurate or misleading information. Avoid assumptions and rely only on explicit statements in the transcript.

🟣 Analytical Note (Benchmark)

This violation is considered low-frequency (approximately 1%) based on previous QA reports, and should be evaluated with high precision to avoid false positives.