Research
SoK: PIAtlas: Decomposing the Unbounded Prompt Injection Attack Space
Publication date unknown · Discovered arxiv.org
Page published
Publication date unknown · First observed: 21 Oct 2026
Coverage timeline
Single-source research — one report is available.
Why it matters
PIAtlas gives defenders a structured way to identify gaps in prompt-injection evaluations and broaden red-team coverage beyond commonly benchmarked attacks.
PIAtlas introduces a prompt-injection taxonomy organized around six instruction-subversion mechanisms. The researchers release PICorpus and map thousands of attack records using a human-verified protocol, finding that attacks commonly combine techniques and that widely used agent benchmarks underrepresent mechanisms favored by adaptive attackers.