Latest Articles for Technology Ethics

Improving trust and compliance in language models through accurate source attribution.

This study reveals LLMs' preference for their own output over human-written content.

Study reveals difficulties for humans and AI in recognizing each other.

Examining how language influences gender views through biases in AI models.

CEIPA helps uncover vulnerabilities in large language models to enhance their safety.

Examining the risks and safety measures for LLM-powered robots.

Examining the impact of LLMs on social stereotyping and ways to improve outcomes.

Study reveals surprising risks of sharing sensitive data with chatbots.

Examining differential privacy and NTK regression to protect user data in AI.

Exploring differential privacy to protect sensitive information in AI applications.

Honest Computing emphasizes trust and ethics in data handling.

A framework enabling data owners to check unauthorized use of their data in ML.

This method enhances privacy while improving model accuracy in AI.

Exploring techniques for effectively removing data from AI models.

Examining the societal and environmental effects of AI foundation models.

Examining differential privacy in natural language processing for better data protection.

Examining how AI tools may reinforce biases in resume screening.

CrossWalk method improves fairness in node embeddings while considering sensitive information.

Deceptive AI explanations can sway beliefs, complicating the fight against false information.

This paper examines the effectiveness of current machine unlearning verification methods.

Examining how T2Is portray caste identities and reinforce stereotypes.

Study reveals potential leaks of personal identity information by VLMs.

Examining the role of contestable AI for fairness and accountability.

A new framework improves alignment of language models with user preferences.

Research explores how to estimate statistics while respecting user privacy needs.

A new dataset highlights biases in speech models based on gender and age.

AI chatbots may create false memories, raising concerns for legal settings.

New methods enhance face recognition while addressing privacy concerns.

A method to improve deepfake detection by reducing data biases.

Analyzing how language models handle social biases in opinions.

A new method reduces biases in image and text models effectively.

A new method improves detection of harmful prompts in language models.

A look at the key differences between Explainable AI and Interpretable AI.

Exploring certified unlearning and its role in data privacy protection.

Examining societal bias in Chinese language models and search engines.

A new approach to improve fairness and accuracy in facial recognition.

Examining how social identities affect biases in language models.

Study reveals vulnerabilities in AI models due to backdoor attacks.

A new dataset aims to improve the understanding of gender bias in AI.

Examining how liability and insurance can mitigate AI risks.