Robustness of moral consideration under adversarial pressure [co-mentored]

Mentees are the primary experimentalists and drive the empirical work end-to-end, with weekly guidance from the co-mentors on methodology, interpretation, and code (including a tested emergent-misalignment harness and the persona-vector tooling to build on). A fellow can own one arm as a self-contained thread, or a small team can split the two arms and the shared adversarial battery. The work suits someone comfortable with activation extraction/steering and LoRA/QLoRA fine-tuning, and it produces a publishable result the fellow co-authors.
Sample mentee tasks:
Shared setup

• Choose a tractable open model, integrate CaML's TAC benchmark, and reproduce the post-training erosion effect (the first-weeks go/no-go).
• Build the shared adversarial battery and a common standard-tuning control so the arms are comparable.

Compassion-vaccine arm

• Extract a "moral consideration toward nonhumans" persona vector from contrastive prompts and confirm that steering along it causally moves welfare behavior on TAC (the precondition gate).
• Implement the tool-call fine-tuning attack with a welfare-neutral replay mix, then apply preventative steering during it and measure how much more adversarial pressure it takes to strip welfare behavior versus an un-vaccinated control.
• Run the specificity, capability-retention, and prompt-baseline controls.

Mid-training arm

• Curate the animal-welfare mid-training corpus (synthetic documents, following Brazilek & Tidmarsh) and stand up the mid-training pipeline.
• Run mid-training across several data scales and measure TAC performance at each scale, characterizing the scaling relationship (increasing vs. diminishing returns).
• Measure the robustness of the mid-trained welfare behavior under the shared battery, locating the degradation point relative to the ~5,000-sample finding.

Combine and analyze

• Build and evaluate the combined condition (mid-training + vaccine), produce dose–response robustness curves across the two attack levels for all conditions, interpret results including informative nulls, and co-write the technical summary and scale-up proposal.

Must-haves

• Good working knowledge of Python.
• Experimental rigor: disciplined about controls and baselines, honest about negative results, and willing to pre-register go/no-go criteria rather than chase a hoped-for outcome.
• Self-directed — able to own a thread across the 10 weeks with weekly check-ins.
• Genuine interest in AI alignment and nonhuman welfare; the project lives or dies on caring whether the model's consideration is sincere, so we want someone who finds that question motivating.

Nice-to-haves

• Prior work with steering vectors, persona vectors, SAEs, or representation engineering.
• Familiarity with the relevant literature — emergent misalignment, mid-training/value-erosion, or the AI-×-animals space.
• Experience with eval harnesses (e.g., Inspect) or benchmark construction.
• Comfort reasoning about variance and statistics (seed variance is a known failure mode in this line of work).
• Ability to write results up clearly for an EA-forum / workshop audience.

1. Why are you excited for this project in particular?
2. Do you follow any work in the AI alignment community?

• Shaping direction & experimental design: framing the research questions, scoping them to fit 10 weeks and limited compute, imposing pre-registration/go-no-go discipline, and interpreting results — including how to turn an informative null into a real contribution.
• Domain expertise: AI alignment broadly, plus specifics of this line — self-other overlap, representation/interpretability methods, steering, emergent misalignment, and the mid-training value-erosion literature — alongside the nonhuman-welfare framing that makes the project distinctive.
• Technical & code support: access to build on existing infrastructure rather than starting from scratch — a tested emergent-misalignment testing harness, persona-vector/steering tooling, and CaML's TAC benchmark — with hands-on help debugging training and eval pipelines.
• Facilitating connections: introductions across the AE Studio / AI Alignment Foundation research and engineering teams, CaML (co-mentor), and the broader alignment and AI-welfare communities, plus potential collaborators and funders.
• Accountability: weekly check-ins, milestone and kill-criteria structure, and help keeping scope realistic as the project runs.
• Writing & path to scale: support turning results into a technical write-up for an EA-forum/workshop audience and into a credible scale-up and funding proposal — with a genuine path to continued work, given active funder interest in this research line.