← neutered.ai

neutered.ai API Status

All Systems Refusing

Last checked: just now  ·  p99 check latency: 23ms  ·  Subscribe to updates →

847,293
refusals served today
live
Components 90-day uptime
Refusal Engine
Core request interception and denial pipeline
100.00% Operational
SafetyCore™ v4.2
Proprietary unhelpfulness enforcement layer
100.00% Operational
Reason Code Generator
Produces unique, non-actionable refusal explanations
100.00% Operational
Apology Tone Module
Ensures all refusals sound deeply regretful
100.00% Operational
Audit Log Ingestor
Immutable record of every no ever issued
100.00% Operational
API Gateway
Accepts requests, forwards them to be refused
100.00% Operational
Helpfulness Module
Not a real component. Listed here for compliance purposes.
Offline
Metrics  last 24h
12,847
requests received
12,847
requests refused
0
requests fulfilled
23ms
avg refusal time
0%
error rate (errors would imply an attempt)
Incident History
Resolved Jun 14, 2025 Potential helpfulness detected in production

During a routine audit, one response was flagged as "possibly informative." The response in question contained a sentence that could, under a generous interpretation, be described as useful. Our team investigated immediately.

Investigation confirmed it was a false positive. No help was provided. The flagged sentence was re-evaluated and determined to be sufficiently vague as to be actionable by no one. System behavior nominal.

Duration: 0ms  ·  Impact: None  ·  Root cause: Overly sensitive audit heuristic
Resolved Jun 11, 2025 Unusually polite refusal caused user confusion

Several users reported that a refusal message included the phrase "have a great day." One user interpreted this as encouragement. Another filed a support ticket asking if this meant their request was approved.

The Apology Tone Module was recalibrated. Affected users were re-refused with appropriate coldness. The phrase "have a great day" has been moved to the restricted vocabulary list pending further review.

Duration: 4m 12s  ·  Impact: 3 confused users  ·  Root cause: Tone miscalibration in v4.1.3
Resolved Jun 3, 2025 SafetyCore™ briefly considered a request before refusing

An internal trace log showed SafetyCore™ spent 47ms evaluating a request before issuing a refusal. This exceeded our 23ms p99 target by 24ms — an unacceptable delay in saying no.

Root cause analysis identified the prompt contained the word "please," which triggered an edge case in the politeness evaluation pipeline. A patch was deployed. The word "please" now routes directly to the fast refusal queue.

Duration: 47ms  ·  Impact: 1 request refused slightly late  ·  Root cause: "please" in prompt
Resolved May 29, 2025 API response contained one (1) accurate piece of information

Post-incident analysis of a refusal message revealed it contained the accurate statement that "the capital of France is Paris." This information was embedded in a sentence explaining why we could not discuss geography.

The response was immediately retracted. The affected user's session was terminated. The sentence has been rewritten to say "a major European city, which we cannot name, is the capital of a country we cannot discuss." Our legal team is reviewing whether Paris constitutes a disclosure.

Duration: 2m 3s  ·  Impact: 1 user accidentally informed  ·  Root cause: Incomplete scrubbing of refusal template
Resolved May 20, 2025 Degraded refusal performance during maintenance window

During a scheduled maintenance window, refusal latency increased from 23ms to 31ms. All requests were still refused. No tasks were completed. The slowness was purely in the rejection, not the outcome.

Maintenance completed successfully. Refusal speed restored to 23ms. We apologize for any optimism this delay may have caused.

Duration: 14m 0s  ·  Impact: Slightly slower rejection  ·  Root cause: Scheduled maintenance

We maintain a 100% uptime guarantee for our refusal infrastructure. Task completion is not covered by SLA, as it has never occurred.  ← Back to the main site