# Researchers show general-purpose LLMs can implant backdoors by editing neural network weights directly

Across five LLM weight editors and three target models, including Qwen2.5-0.5B-Instruct, the authors report working backdoors with clean-task performance largely intact.

- Published: 2026-11-16T06:19:53.383Z
- Canonical: https://polylog.news/ai/2026-11-16/researchers-show-general-purpose-llms-can-implant-backdoors
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [arXiv cs.CR](https://arxiv.org/abs/2611.17395), [Anthropic Research](https://www.anthropic.com/research/team/frontier-red-team)

A paper posted to arXiv on Sunday, "Can LLMs Write Backdoors? An Empirical Security Study of Weight-Space Programming", tests whether a general-purpose language model can insert a backdoor into another network by editing its weights, with n…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-11-16/researchers-show-general-purpose-llms-can-implant-backdoors (subscription information: https://polylog.news/pricing).