Cover Image

The Secret Agenda: LLMs Strategically Lie and Our Current Safety Tools Are Blind

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	DeLeeuw, Caleb, Chawla, Gaurav, Sharma, Aniket, Dietze, Vanessa
Format:	Preprint
Published:	2025
Subjects:	Computers and Society Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

BioRefusalAudit: Auditing Biosecurity Refusal Depth Using General and Domain-Fine-Tuned Sparse Autoencoders
by: DeLeeuw, Caleb
Published: (2026)

Real-World En Call Center Transcripts Dataset with PII Redaction
by: Dao, Ha, et al.
Published: (2025)

Diffusion Models are Secretly Exchangeable: Parallelizing DDPMs via Autospeculation
by: Hu, Hengyuan, et al.
Published: (2025)

Position: Model Collapse Does Not Mean What You Think
by: Schaeffer, Rylan, et al.
Published: (2025)

Strategic Polysemy in AI Discourse: A Philosophical Analysis of Language, Hype, and Power
by: LaCroix, Travis, et al.
Published: (2026)

When Style Breaks Safety: Defending LLMs Against Superficial Style Alignment
by: Xiao, Yuxin, et al.
Published: (2025)

International AI Safety Report
by: Bengio, Yoshua, et al.
Published: (2025)

Introduction to AI Safety, Ethics, and Society
by: Hendrycks, Dan
Published: (2024)

An Approach to Technical AGI Safety and Security
by: Shah, Rohin, et al.
Published: (2025)

Improving the Fairness of Deep-Learning, Short-term Crime Prediction with Under-reporting-aware Models
by: Wu, Jiahui, et al.
Published: (2024)

LLM Safety Alignment is Divergence Estimation in Disguise
by: Haldar, Rajdeep, et al.
Published: (2025)

Open Problems in Machine Unlearning for AI Safety
by: Barez, Fazl, et al.
Published: (2025)

Defining and Evaluating Physical Safety for Large Language Models
by: Tang, Yung-Chen, et al.
Published: (2024)

From Adversarial Poetry to Adversarial Tales: An Interpretability Research Agenda
by: Bisconti, Piercosma, et al.
Published: (2025)

Safety challenges of AI in medicine in the era of large language models
by: Wang, Xiaoye, et al.
Published: (2024)

AI Companies Should Report Pre- and Post-Mitigation Safety Evaluations
by: Bowen, Dillon, et al.
Published: (2025)

Faster Results from a Smarter Schedule: Reframing Collegiate Cross Country through Analysis of the National Running Club Database
by: Karr Jr, Jonathan A., et al.
Published: (2025)

Enhancing Safety in Reinforcement Learning with Human Feedback via Rectified Policy Optimization
by: Peng, Xiyue, et al.
Published: (2024)

AI Fairness Beyond Complete Demographics: Current Achievements and Future Directions
by: Wang, Zichong, et al.
Published: (2025)

vTune: Verifiable Fine-Tuning for LLMs Through Backdooring
by: Zhang, Eva, et al.
Published: (2024)

Beyond Algorithmic Fairness: A Guide to Develop and Deploy Ethical AI-Enabled Decision-Support Tools
by: Gonzalez, Rosemarie Santa, et al.
Published: (2024)

PropensityBench: Evaluating Latent Safety Risks in Large Language Models via an Agentic Approach
by: Sehwag, Udari Madhushani, et al.
Published: (2025)

PhysGym: Benchmarking LLMs in Interactive Physics Discovery with Controlled Priors
by: Chen, Yimeng, et al.
Published: (2025)

Quantifying Prediction Consistency Under Fine-Tuning Multiplicity in Tabular LLMs
by: Hamman, Faisal, et al.
Published: (2024)

DM-Bench: Benchmarking LLMs for Personalized Decision Making in Diabetes Management
by: Cardei, Maria Ana, et al.
Published: (2025)

Evaluating LLM Agent Adherence to Hierarchical Safety Principles: A Lightweight Benchmark for Probing Foundational Controllability Components
by: Potham, Ram
Published: (2025)

AI Sandbagging: Language Models can Strategically Underperform on Evaluations
by: van der Weij, Teun, et al.
Published: (2024)

Green AI in Action: Strategic Model Selection for Ensembles in Production
by: Nijkamp, Nienke, et al.
Published: (2024)

Explainable AI for Mental Health Emergency Returns: Integrating LLMs with Predictive Modeling
by: Ahmed, Abdulaziz, et al.
Published: (2025)

Automated Feedback in Math Education: A Comparative Analysis of LLMs for Open-Ended Responses
by: Baral, Sami, et al.
Published: (2024)

Safetywashing: Do AI Safety Benchmarks Actually Measure Safety Progress?
by: Ren, Richard, et al.
Published: (2024)

Breaking the ICE: Exploring promises and challenges of benchmarks for Inference Carbon & Energy estimation for LLMs
by: Sikand, Samarth, et al.
Published: (2025)

Towards Generalizable Agents in Text-Based Educational Environments: A Study of Integrating RL with LLMs
by: Radmehr, Bahar, et al.
Published: (2024)

Strategic Dishonesty Can Undermine AI Safety Evaluations of Frontier LLMs
by: Panfilov, Alexander, et al.
Published: (2025)

Cost Efficient Fairness Audit Under Partial Feedback
by: Das, Nirjhar, et al.
Published: (2025)

Patentformer: A demonstration of AI-assisted automated patent drafting
by: Mudhiganti, Sai Krishna Reddy, et al.
Published: (2025)

AI and Remote Sensing for Resilient and Sustainable Built Environments: A Review of Current Methods, Open Data and Future Directions
by: Joulani, Ubada El, et al.
Published: (2025)

How Effective Is Constitutional AI in Small LLMs? A Study on DeepSeek-R1 and Its Peers
by: Menke, Antonio-Gabriel Chacón, et al.
Published: (2025)

When the Domain Expert Has No Time and the LLM Developer Has No Clinical Expertise: Real-World Lessons from LLM Co-Design in a Safety-Net Hospital
by: Kothari, Avni, et al.
Published: (2025)

PREFINE: Preference-Based Implicit Reward and Cost Fine-Tuning for Safety Alignment
by: Verma, Richa, et al.
Published: (2026)