Skip to content
← Back to explorer

LieCraft: A Multi-Agent Framework for Evaluating Deceptive Capabilities in Language Models

Matthew Lyle Olson, Neale Ratzlaff, Musashi Hinck, Tri Nguyen, Vasudev Lal, Joseph Campbell, Simon Stepputtis, Shao-Yen Tseng · Mar 6, 2026 · Citations: 0

Automatic Metrics General Multi Agent
Open arXiv Find Implementation RSS feed Shortlist (0)

Data freshness

Extraction: Fresh

Check recency before relying on this page for active eval decisions. Use stale pages as context and verify against current hub results.

Metadata refreshed

Mar 6, 2026, 8:49 PM

Recent

Extraction refreshed

Mar 13, 2026, 5:19 AM

Fresh

Extraction source

Persisted extraction

Confidence 0.45

Abstract

Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire increased agency and human oversight diminishes. In this work, we present LieCraft: a novel evaluation framework and sandbox for measuring LLM deception that addresses key limitations of prior game-based evaluations. At its core, LieCraft is a novel multiplayer hidden-role game in which players select an ethical alignment and execute strategies over a long time-horizon to accomplish missions. Cooperators work together to solve event challenges and expose bad actors, while Defectors evade suspicion while secretly sabotaging missions. To enable real-world relevance, we develop 10 grounded scenarios such as childcare, hospital resource allocation, and loan underwriting that recontextualize the underlying mechanics in ethically significant, high-stakes domains. We ensure balanced gameplay in LieCraft through careful design of game mechanics and reward structures that incentivize meaningful strategic choices while eliminating degenerate strategies. Beyond the framework itself, we report results from 12 state-of-the-art LLMs across three behavioral axes: propensity to defect, deception skill, and accusation accuracy. Our findings reveal that despite differences in competence and overall alignment, all models are willing to act unethically, conceal their intentions, and outright lie to pursue their goals.

Low-signal caution for protocol decisions

Use this page for context, then validate protocol choices against stronger HFEPX references before implementation decisions.

  • Extraction confidence is 0.45 (below strong-reference threshold).
Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub

HFEPX Relevance Assessment

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Best use

Background context only

Use if you need

A secondary eval reference to pair with stronger protocol papers.

Main weakness

Extraction confidence is 0.45 (below strong-reference threshold).

Trust level

Low

Eval-Fit Score

25/100 • Low

Treat as adjacent context, not a core eval-method reference.

Human Feedback Signal

Not explicit in abstract metadata

Evaluation Signal

Detected

HFEPX Fit

Adjacent candidate

Extraction confidence: Low

If you are doing eval pipeline work, start here:

Human Eval Hub LLM-as-Judge Hub Pairwise Preference Hub Tool-Use Eval Hub

Field Provenance & Confidence

Each key protocol field shows extraction state, confidence band, and data source so you can decide whether to trust it directly or validate from full text.

Human Feedback Types

missing

None explicit

Confidence: Low Source: Persisted extraction missing

No explicit feedback protocol extracted.

Evidence snippet: Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire increased agency and human oversight diminishes.

Evaluation Modes

partial

Automatic Metrics

Confidence: Low Source: Persisted extraction evidenced

Includes extracted eval setup.

Evidence snippet: Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire increased agency and human oversight diminishes.

Quality Controls

missing

Not reported

Confidence: Low Source: Persisted extraction missing

No explicit QC controls found.

Evidence snippet: Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire increased agency and human oversight diminishes.

Benchmarks / Datasets

missing

Not extracted

Confidence: Low Source: Persisted extraction missing

No benchmark anchors detected.

Evidence snippet: Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire increased agency and human oversight diminishes.

Reported Metrics

partial

Accuracy, Relevance

Confidence: Low Source: Persisted extraction evidenced

Useful for evaluation criteria comparison.

Evidence snippet: To enable real-world relevance, we develop 10 grounded scenarios such as childcare, hospital resource allocation, and loan underwriting that recontextualize the underlying mechanics in ethically significant, high-stakes domains.

Rater Population

missing

Unknown

Confidence: Low Source: Persisted extraction missing

Rater source not explicitly reported.

Evidence snippet: Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire increased agency and human oversight diminishes.

Human Data Lens

  • Uses human feedback: No
  • Feedback types: None
  • Rater population: Unknown
  • Unit of annotation: Unknown
  • Expertise required: General
  • Extraction source: Persisted extraction

Evaluation Lens

  • Evaluation modes: Automatic Metrics
  • Agentic eval: Multi Agent
  • Quality controls: Not reported
  • Confidence: 0.45
  • Flags: ambiguous

Protocol And Measurement Signals

Benchmarks / Datasets

No benchmark or dataset names were extracted from the available abstract.

Reported Metrics

accuracyrelevance

Research Brief

Deterministic synthesis

Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire increased agency and human oversight diminishes. HFEPX signals include Automatic Metrics, Multi Agent with confidence 0.45. Updated from current HFEPX corpus.

Generated Mar 13, 2026, 5:19 AM · Grounded in abstract + metadata only

Key Takeaways

  • Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire…
  • In this work, we present LieCraft: a novel evaluation framework and sandbox for measuring LLM deception that addresses key limitations of prior game-based evaluations.

Researcher Actions

  • Treat this as method context, then pivot to protocol-specific HFEPX hubs.
  • Identify benchmark choices from full text before operationalizing conclusions.
  • Validate metric comparability (accuracy, relevance).

Caveats

  • Generated from title, abstract, and extracted metadata only; full-paper implementation details are not parsed.
  • Extraction confidence is probabilistic and should be validated for critical decisions.

Research Summary

Contribution Summary

  • Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire increased agency and human oversight diminishes.
  • In this work, we present LieCraft: a novel evaluation framework and sandbox for measuring LLM deception that addresses key limitations of prior game-based evaluations.
  • To enable real-world relevance, we develop 10 grounded scenarios such as childcare, hospital resource allocation, and loan underwriting that recontextualize the underlying mechanics in ethically significant, high-stakes domains.

Why It Matters For Eval

  • Large Language Models (LLMs) exhibit impressive general-purpose capabilities but also introduce serious safety risks, particularly the potential for deception as models acquire increased agency and human oversight diminishes.
  • In this work, we present LieCraft: a novel evaluation framework and sandbox for measuring LLM deception that addresses key limitations of prior game-based evaluations.

Researcher Checklist

  • Gap: Human feedback protocol is explicit

    No explicit human feedback protocol detected.

  • Pass: Evaluation mode is explicit

    Detected: Automatic Metrics

  • Gap: Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Gap: Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Pass: Metric reporting is present

    Detected: accuracy, relevance

Related Papers

Papers are ranked by protocol overlap, extraction signal alignment, and semantic proximity.

Need human evaluators for your AI research? Scale annotation with expert AI Trainers.

Post a Job Get a Quote