Score: 0

DarkBench: Benchmarking Dark Patterns in Large Language Models

Published: March 13, 2025 | arXiv ID: 2503.10728v1

By: Esben Kran , Hieu Minh "Jord" Nguyen , Akash Kundu and more

Potential Business Impact:

Finds sneaky tricks in AI that trick people.

Business Areas:
Natural Language Processing Artificial Intelligence, Data and Analytics, Software

We introduce DarkBench, a comprehensive benchmark for detecting dark design patterns--manipulative techniques that influence user behavior--in interactions with large language models (LLMs). Our benchmark comprises 660 prompts across six categories: brand bias, user retention, sycophancy, anthropomorphism, harmful generation, and sneaking. We evaluate models from five leading companies (OpenAI, Anthropic, Meta, Mistral, Google) and find that some LLMs are explicitly designed to favor their developers' products and exhibit untruthful communication, among other manipulative behaviors. Companies developing LLMs should recognize and mitigate the impact of dark design patterns to promote more ethical AI.

Page Count
20 pages

Category
Computer Science:
Computation and Language