PREVIEW. This is how it will look in an inbox; it was not sent to you. Each real email carries its own unsubscribe link and Keryn's postal address.

Keryn Headlines

Daily, Thursday 5 February 2026 · Retrospective · Covers 4 February 2026 00:00 to 5 February 2026 00:00 UTC

Read on the web

Top developments

Detecting backdoored language models at scale

Microsoft · Incident · Models

CISA warns of five-year-old GitLab flaw exploited in attacks

BleepingComputer · Incident · Remediation

Hackers compromise NGINX servers to redirect user traffic

BleepingComputer · Incident

SCA

The Birthmark Standard: Privacy-Preserving Photo Authentication via Hardware Roots of Trust and Consortium Blockchain

arXiv · Research · SCA

Eclipse Foundation Mandates Pre-Publish Security Checks for Open VSX Extensions

The Hacker News · Analysis · SCA

Trust

Pluralism in AI Governance: Toward Sociotechnical Alignment and Normative Coherence

arXiv · Research · Trust

Models

Inference-Time Backdoors via Chat Templates: From LLM Supply Chains to Agentic System Compromise

arXiv · Research · Models

Trust The Typical

arXiv · Research · Models

Routing-Aware Safety Alignment for Mixture-of-Experts Models

arXiv · Research · Models

AI Security

Bypassing AI Control Protocols via Agent-as-a-Proxy Attacks

arXiv · Research · AI Security

Advisories in this window

  • 54 npm advisories · 44 critical · 8 high · 2 moderate
  • 22 Go advisories · 3 critical · 6 high · 12 moderate · 1 low
  • 2 crates.io advisories · 1 high · 1 low
  • 1 Packagist advisory · 1 low
  • 1 PyPI advisory · 1 critical
  • 1 Maven advisory · 1 moderate

All advisories

Titles, links and short excerpts only; every item links to its source. Selection is automatic; the editor's note is AI-drafted and checked as described. Preview: the unsubscribe link and Keryn's postal address are added to each email when it is sent.