Skip to content
View EmilioBarkett's full-sized avatar
đź‘‹
đź‘‹
3 followers · 2 following

Block or report EmilioBarkett

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
EmilioBarkett/README.md

Hi, I'm Emilio đź‘‹

I'm an AI safety generalist focused on fieldbuilding at the intersection of AI safety and the social sciences. I want to bring the deep, underutilized pool of social science researchers into AI safety.

My work is in empirical and technical AI governance, along two threads:

  1. How AI systems replicate human behavior, and what that means for society, especially how models pick up and reproduce human biases, personas, and social dynamics.
  2. How organizations adopt AI, especially how they identify, communicate about, and manage the risks that come with it.

I'm currently a Research Mentor at SPAR, where I lead projects on the above topics.

I'm always happy to chat with new people, so feel free to reach out via my website.

Pinned Loading

  1. moral-dilemma-project moral-dilemma-project Public

    LLM moral reasoning under framing and social pressure: process dissociation and sycophancy studies across frontier models (SPAR project, in progress)

    Python

  2. brunswik-tacit-project brunswik-tacit-project Public

    Companion code and data for 'Whose Alignment?' (ICML 2026 Pluralistic Alignment Workshop, arXiv 2605.25256): LLM process alignment on German Credit decisions via the Brunswik lens model

    Jupyter Notebook

  3. long-olivia/escalation-commitment long-olivia/escalation-commitment Public

    Repository for the paper "Getting out of the Big-Muddy: Escalation of Commitment in LLMs."

    Python 1

  4. truth-bias truth-bias Public

    Data and analysis code for 'Reasoning Isn't Enough: Examining Truth-Bias and Sycophancy in LLMs' (ICML 2025 workshop, arXiv 2506.21561)

    TeX

  5. realization-effect-project realization-effect-project Public

    Representation–steerability correspondence benchmark for language models; grew out of 'Representation Without Control' (arXiv 2605.25151)

    Python 1

  6. status-hierarchy status-hierarchy Public

    Code, results, and paper source for 'Status Hierarchies in Language Models' (MA thesis, arXiv 2601.17577): do language models defer to partners based on status cues?

    TeX