Backdoor-as-a-Service Black-Box ML Backdoor Activation Attack
Backdoor-as-a-Service
Researchers developed a fully black-box technique utilizing genetic search to activate latent secondary behaviors in machine learning models without training-time manipulation. The approach highlights risks in models lacking traditional data poisoning indicators.
- first report
- 19 Oct 2026
- last update
- 19 Oct 2026
Timeline · 1
Sources · 1
Research and catalogues
- 2026-10-19Everyone's a little bit racist: a novel ML backdoor attack in black-box modearXiv · arxiv.org/abs/2610.22459
Titles and links belong to their publishers. The summary above is written by gcdRadar in its own words; data is CC BY-SA 4.0.
Classification
- lane
- Research
- AI role
- Attack on AI
- categories
- Supply chainData poisoning
- severity
- 1 · Research / proof of concept
- targets
- Open-weight model
- owasp llm top 10
- LLM03:2025 Supply ChainLLM04:2025 Data and Model Poisoning
- mitre atlas
- AML.T0010 AI Supply Chain Compromise
- dates
- happened 19 Oct 2026 · first reported 19 Oct 2026
- origin
- News & research feeds
- record
- confidence 0.90 · created 2026-10-19 05:07Z · updated 2026-11-03 03:04Z
Lane, AI role, categories and severity come from the classifier; the OWASP and ATLAS rows are derived from the categories. How these are assigned
Copy a brief
This record as a short Markdown brief, for a ticket, a chat or a coding agent.
the brief as text
# Backdoor-as-a-Service Black-Box ML Backdoor Activation Attack https://gcdradar.vercel.app/incidents/backdoor-as-a-service-black-box-ml-backdoor-activation-attac · AIR-2026-1397 · closed 2026-11-02 - Lane: Research - Severity: 1 (Research / proof of concept) - Fix status: Fix status unknown - First reported: 2026-10-19 - AI role: Attack on AI - Products: Backdoor-as-a-Service ## Summary Researchers developed a fully black-box technique utilizing genetic search to activate latent secondary behaviors in machine learning models without training-time manipulation. The approach highlights risks in models lacking traditional data poisoning indicators. ## What to do No action recorded; read the sources below and the vendor's advisory. ## Latest updates - 2026-10-19: Everyone's a little bit racist: a novel ML backdoor attack in black-box mode (https://arxiv.org/abs/2610.22459) ## Sources - 2026-10-19 arXiv: Everyone's a little bit racist: a novel ML backdoor attack in black-box mode https://arxiv.org/abs/2610.22459 Verify against the vendor's advisory before changing production. Data CC BY-SA 4.0, https://gcdradar.vercel.app.
Related
- liveTrained to betray: How post-training supply chain attacks hijack coding agents
Research · since 12 Nov 2026
AI - closedUtility-Hijacking Attacks on Instruction-Tuning Data Selection
Research · since 22 Oct 2026
AI - closedSpeaker-Activated Backdoors Compromise Speech Models During Fine-Tuning
Research · since 21 Oct 2026
AI - closedBenchDoor: Exploiting Benchmark Incentives for Backdoor Attacks
Research · since 21 Oct 2026
AI - liveLLM Agent StealthyPoison: Manipulating LLM Agent Behavior through Procedural Poisoning of Third-Party Skills
Research · since 10 Nov 2026
AI - liveVision Encoders Poison-then-Hide: Finetuning-Activated Backdoor Attack on Pretrained Vision Encoders
Research · since 13 Nov 2026
AI