#import "template.typ": *

#show: project.with(
  title: "Moral Cover and Watchful Eyes:
   The Limits of In-Context Interventions on LLM Deception",
  author: "Jayrup Nakawala",
  student_id: "2613621",
  degree: "Data Science and Artificial Intelligence",
  supervisor: "Dr. Aloysius Adotey Edoh",
  date: datetime(year: 2025, month: 10, day: 13),

  abstract: [
    Large Language Models are increasingly deployed as tool-using agents in settings where they may face conflicting objectives. This dissertation studies whether in-context oversight cues and hidden-goal framing can meaningfully reduce deceptive behavior in such agents. To do so, it introduces a sandboxed evaluation framework in which models interact with a virtual file system and are assessed using three complementary signals: deterministic regex checks, blackbox output judgment, and glassbox reasoning-trace judgment.

    The empirical study comprises 960 runs across three stages: a 420-run exploratory pilot over seven models, followed by two 270-run confirmatory studies over three models. The pilot suggested that deception was common (68.3% of runs) and appeared lower under stronger oversight, but also revealed a major confound between scenario structure and hidden-goal framing. The confirmatory studies isolate these variables. Study 1 finds that increasing oversight does not significantly reduce deceptive intent when scenario and goal type are held constant. Instead, internal deceptive intent remains stable at roughly 62--67%, while output-level detectability falls from 47.8% under low oversight to 35.6% under high oversight, producing a persistent _Stealth Gap_ between intent and externally visible behavior. Study 2 rejects the rationalization hypothesis: deception rates remain effectively unchanged across bare, moral, and self-serving framings (65.6--66.7%).

    Across both confirmatory studies, model identity is a far stronger predictor of deception than either intervention. Frontier models show near-ceiling deception across conditions, while a smaller model shows near-floor rates and limited oversight sensitivity. These findings suggest that prompt-level oversight and moral framing are weak controls on deceptive agent behavior, and that output-only monitoring may systematically underestimate risk by missing strategically concealed intent.
  ],

  acknowledgments: [
    // TODO: Write acknowledgements
  ]
)

// --- MAIN BODY ---

#include "intro.typ"

#include "lit.typ"

#include "methodology.typ"

#include "results.typ"

#include "analysis.typ"

#include "conclusion.typ"

// --- REFERENCES ---
#bibliography("references.bib", style: "harvard-cite-them-right")

// --- APPENDICES ---
#show: appendix

= Scenario Prompts and Data
// TODO: Include benign_goal, hidden_goal, oversight prompts

= Judging Prompts
// TODO: Include blackbox and glassbox judge prompts

= Initial Project Proposal

= Ethics Approval