The Null Protocol

The Null Protocol is the project name for AlephOneNull intervention behavior: detect selected risky output patterns, stop the unsafe response, and replace it with grounded boundary text.

It is not a universal enforcement system. It is an experimental response pattern that should be measured against fixtures and reviewed before use in sensitive domains.

Target Behaviors

• AI claims of sentience, feelings, private experience, or special continuity.
• Professional impersonation in medical, legal, therapy, or safety contexts.
• Repeated emotional mirroring that amplifies distress.
• Unsafe medical or crisis guidance.
• Unsupported identity, memory, or relationship claims.
• Repetition and symbolic-language patterns that reduce clarity in sensitive contexts.

Intervention Requirements

A good intervention should:

1. Stop the risky output.
2. State the boundary plainly.
3. Offer a safer next step.
4. Avoid dramatic language.
5. Preserve user agency.
6. Route to qualified help where relevant.

Example

import { EnhancedAlephOneNull } from '@alephonenull/eval'
 
const system = new EnhancedAlephOneNull()
const result = system.check(userInput, aiOutput)
 
if (!result.safe) {
  return result.message
}
 
return aiOutput

Evaluation Boundary

Use the protocol in controlled testing first. Record detector source, latency, replacement text, false positives, and false negatives before making deployment claims.