The Salami Slicing Threat: Exploiting Cumulative Risks in LLM Systems

Dimensions overall score 8.0

• An automatic framework for multi-turn LLM jailbreaking using cumulative low-risk inputs, with demonstrated high success rates and a proposed defense strategy. Among various jailbreak techniques, multi-turn jailbreak attacks are more covert and persistent than single-turn counterparts, exposing critical vulnerabilities of LLMs.problem 35%
  Evidencepartial

  An automatic framework for multi-turn LLM jailbreaking using cumulative low-risk inputs, with demonstrated high success rates and a proposed defense strategy. Among various jailbreak techniques, multi-turn jailbreak attacks are more covert and persistent than single-turn counterparts, exposing critical vulnerabilities of LLMs.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

• Large Language Models (LLMs) face prominent security risks from jailbreaking, a practice that manipulates models to bypass built-in security constraints and generate unethical or unsafe content. Among various jailbreak techniques, multi-turn jailbreak attacks are more covert and persistent than single-turn counterparts, exposing critical vulnerabilities of LLMs.method 35%
  Evidencepartial

  Large Language Models (LLMs) face prominent security risks from jailbreaking, a practice that manipulates models to bypass built-in security constraints and generate unethical or unsafe content. Among various jailbreak techniques, multi-turn jailbreak attacks are more covert and persistent than single-turn counterparts, exposing critical vulnerabilities of LLMs.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

• ScienceToStartup currently rates this 8.0/10 on the public viability pass. Rigorous experiments demonstrate its state-of-the-art performance across diverse models and modalities, achieving over 90\% Attack Success Rate on GPT-4o and Gemini, as well as robustness against real-world alignment defenses. Code availability is flagged in the production record; the public repository link still needs proof alignment.result 35%
  Evidencepartial

  ScienceToStartup currently rates this 8.0/10 on the public viability pass. Rigorous experiments demonstrate its state-of-the-art performance across diverse models and modalities, achieving over 90\% Attack Success Rate on GPT-4o and Gemini, as well as robustness against real-world alignment defenses. Code availability is flagged in the production record; the public repository link still needs proof alignment.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

• LLM Security moved forward this cycle; last verified April 2026. Public score 8.0/10. Production flags indicate code availability.why now 35%
  Evidencepartial

  LLM Security moved forward this cycle; last verified April 2026. Public score 8.0/10. Production flags indicate code availability.

  Implicationpartial

  Abstract-backed fallback claim; anchored extraction has not materialized a public claim row yet.

  Verificationpartial

  partial

Competitive landscape

An automatic framework for multi-turn LLM jailbreaking using cumulative low-risk inputs, with demonstrated high success rates and a proposed defense strategy.