Sheer Karny

Cathy Mengying Fang, Sheer Karny, Chayapatr Archiwaranguprok et al.

Alignment research on large language models (LLMs) increasingly depends on understanding how these systems are used in everyday contexts. Yet naturalistic interaction data is difficult to access due to privacy constraints and platform control. We present AI-Wrapped, a prototype workflow for collecting naturalistic LLM chatbot usage data while providing participants with an immediate "wrapped"-style report on their usage statistics, top topics, and behavioral patterns. We report findings from an initial deployment with 82 U.S.-based adults across 48,495 conversations from their 2025 chat histories. Participants used LLMs for both instrumental and reflective purposes and had topics with emotional or existential themes. Some usage patterns reflect potential over-reliance or perfectionism. Heavy users showed comparatively more reflective exchanges than primarily transactional ones. Methodologically, even with zero data retention and PII removal, participants may remain hesitant to share chat data due to perceived privacy and judgment risks, underscoring the importance of transparent design when building measurement infrastructure for alignment research.

Sheer Karny, Anthony Baez, Pat Pataranutaporn

Chatbot behavior is often opaque to users, as responses can shift unpredictably across a conversation, drifting toward sycophancy, toxicity, or other unsafe responses. This can leave users vulnerable, either being misled by overly agreeable AI or manipulated by a harmful chatbot that no longer behaves as intended. To address this, we introduce multi-turn neural transparency, an interface that surfaces an LLM's internal neural activations in real time to help users anticipate and recognize how behaviors change across turns. We construct behavioral vectors for six personality traits using methods from mechanistic interpretability, identifying directions in activation space that correlate with trait expression ($R^2 \geq 0.9$) via contrastive system prompts, and visualize trait expression using a sunburst and drift panel that updates at each turn. In a randomized controlled study (N = 246), participants predicted trait expression from a system prompt alone, then rated observed behavior after interacting with the chatbot for both assistant and role-play personas. We find that participants without visualization struggled to accurately evaluate traits (RMSE $\approx$ 0.6-0.7), while the inclusion of neural transparency significantly improved both anticipation and evaluation compared to no visualization (d = -0.34 to -0.49). The multi-turn dynamic visualization additionally outperformed the static single-turn visualization on holistic evaluation of model behavior (d = -0.32). Transparency also reduced overconfidence: participants without visualization grew more confident despite no gain in accuracy. These findings suggest that surfacing internal model representations to everyday users is a meaningful step toward more transparent and informed human-AI interaction.

Sheer Karny, Anthony Baez, Pat Pataranutaporn

Millions of users now design personalized LLM-based chatbots that shape their daily interactions, yet they can only loosely anticipate how their design choices will manifest as behaviors in deployment. This opacity is consequential: seemingly innocuous prompts can trigger excessive sycophancy, toxicity, or inconsistency, degrading utility and raising safety concerns. To address this issue, we introduce an interface that enables neural transparency by exposing language model internals during chatbot design. Our approach extracts behavioral trait vectors (empathy, toxicity, sycophancy, etc.) by computing differences in neural activations between contrastive system prompts that elicit opposing behaviors. We predict chatbot behaviors by projecting the system prompt's final token activations onto these trait vectors, normalizing for cross-trait comparability, and visualizing results via an interactive sunburst diagram. To evaluate this approach, we conducted an online user study using Prolific to compare our neural transparency interface against a baseline chatbot interface without any form of transparency. Our analyses suggest that users systematically miscalibrated AI behavior: participants misjudged trait activations for eleven of fifteen analyzable traits, motivating the need for transparency tools in everyday human-AI interaction. While our interface did not change design iteration patterns, it significantly increased user trust and was enthusiastically received. Qualitative analysis indicated that users' had nuanced experiences with the visualization that may enrich future work designing neurally transparent interfaces. This work offers a path for how mechanistic interpretability can be operationalized for non-technical users, establishing a foundation for safer, more aligned human-AI interactions.

Lukas William Mayer, Sheer Karny, Jackie Ayoub et al.

Despite the growing interest in collaborative AI, designing systems that seamlessly integrate human input remains a major challenge. In this study, we developed a task to systematically examine human preferences for collaborative agents. We created and evaluated five collaborative AI agents with strategies that differ in the manner and degree they adapt to human actions. Participants interacted with a subset of these agents, evaluated their perceived traits, and selected their preferred agent. We used a Bayesian model to understand how agents' strategies influence the Human-AI team performance, AI's perceived traits, and the factors shaping human-preferences in pairwise agent comparisons. Our results show that agents who are more considerate of human actions are preferred over purely performance-maximizing agents. Moreover, we show that such human-centric design can improve the likability of AI collaborators without reducing performance. We find evidence for inequality-aversion effects being a driver of human choices, suggesting that people prefer collaborative agents which allow them to meaningfully contribute to the team. Taken together, these findings demonstrate how collaboration with AI can benefit from development efforts which include both subjective and objective metrics.

Mark Steyvers, Heliodoro Tejeda, Aakriti Kumar et al.

As artificial intelligence (AI) systems, particularly large language models (LLMs), become increasingly integrated into decision-making processes, the ability to trust their outputs is crucial. To earn human trust, LLMs must be well calibrated such that they can accurately assess and communicate the likelihood of their predictions being correct. Whereas recent work has focused on LLMs' internal confidence, less is understood about how effectively they convey uncertainty to users. Here we explore the calibration gap, which refers to the difference between human confidence in LLM-generated answers and the models' actual confidence, and the discrimination gap, which reflects how well humans and models can distinguish between correct and incorrect answers. Our experiments with multiple-choice and short-answer questions reveal that users tend to overestimate the accuracy of LLM responses when provided with default explanations. Moreover, longer explanations increased user confidence, even when the extra length did not improve answer accuracy. By adjusting LLM explanations to better reflect the models' internal confidence, both the calibration gap and the discrimination gap narrowed, significantly improving user perception of LLM accuracy. These findings underscore the importance of accurate uncertainty communication and highlight the effect of explanation length in influencing user trust in AI-assisted decision-making environments. Code and Data can be found at https://osf.io/y7pr6/ . Journal publication can be found on Nature Machine Intelligence at https://www.nature.com/articles/s42256-024-00976-7 .