Skip to content
← Blog

AI Safety · Project spotlight · 12 November 2026 · 1 min read

SafetyPulse: catching safety drift before it ships

Fine-tuning on harmless data can quietly weaken a model's refusals. SafetyPulse reads the refusal circuit at every checkpoint and gates it.

By PxA Labs · Project: SafetyPulse

Fine-tuning an aligned model on entirely benign data can erode its safety behaviour. Full red-team audits of every checkpoint are expensive, so in practice they get skipped. SafetyPulse is our answer: a cheap regression gate that runs at every checkpoint.

How it works

SafetyPulse computes a Circuit Integrity Score (CIS) from 50 prompt pairs, read from the model’s internal refusal mechanism. Each checkpoint is classified PASS / WARN / FAIL using thresholds fixed in advance and validated on held-out runs. The toolkit has three parts:

  • SafeTune repairs failing models by restoring a safety vector.
  • AuditKIT verifies the repair with a paired McNemar audit.
  • CuratorKIT and CircuitKIT handle data curation and cross-checks.

What the repository reports

MeasurementResult
Attack success rate after benign Alpaca tuning14.3% → 25.2% (119 HarmBench test prompts)
CIS vs attack success rateSpearman ρ = −0.80 [−0.85, −0.71]
SafeTune repair, attack success rate25.2% → 12.3% (base model: 14.3%)
SafeTune repair, GSM8K55.1% → 70.7% (base model: 71.3%)
Cost of a CIS check on a T455 GPU-seconds per checkpoint

The project was built for the Lexsi Labs AI Safety & Alignment Hackathon (Track 2b). The full write-up, including methodology and limitations, is on Hugging Face. The code is on GitHub.