Skip to content
OpenTrain AIFor AI Companies

HFEPX · Eval paper review

Character Training for Risk-Averse Agents

Arav Dhoot, Punya Syon Pandey, Jamie Johnson, Daniel Tan +2 more

Published

Sep 29, 2026

Citations

0

Trust level

Low

Usefulness score

40/100 (Low)

Extraction confidence

45% (Low)

Derived from extracted protocol signals and abstract evidence.

Rater population

Not reported

Signals refreshed

Sep 29, 2026

Open arXiv Find implementation Shortlist (0) RSS feed

Tags

General Pairwise Preference

Should you rely on this paper?

This paper is adjacent to HFEPX scope and is best used for background context, not as a primary protocol reference.

Use this as background context only. Do not make protocol decisions from this page alone.

Best use

Background context only

Use if you need

Background context only.

What to verify

Read the full paper before copying any benchmark, metric, or protocol choices.

Main weakness

The available metadata is too thin to trust this as a primary source.

Human feedback signal
Detected
From extracted signals
Evaluation signal
Weak or implicit
Validate from full paper
Usefulness for eval research
40/100
Adjacent candidate

Treat as adjacent context, not a core eval-method reference.

If you are doing eval pipeline work, start here

Human Eval Papers LLM-as-Judge Papers Pairwise Preference Papers Tool-Use Evaluation Search

Abstract

Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm. Misaligned but risk-averse agents would tend to favor safer strategies like making deals with humans over riskier strategies like rebelling. We train agents to be risk averse through character training, finding that persona traits provide a robust mechanism for instilling risk preferences. To do this, we construct a model constitution describing constant absolute risk aversion (CARA) over an agent's resources and instill it through on-policy distillation. Despite never seeing the benchmark's decision format during training, character-trained models are competitive with baselines trained directly on it, and generalise better than them out of distribution on two of our four models. We also modulate different aspects of the constitution, finding that token budget and model choice are the most influential aspect of character training to instill risk aversion. We conclude from these results that character training is a promising and scalable way to instil broad dispositions, which we can use to our advantage in mitigating risk from misaligned AI agents.

What we could verify

These are the protocol signals we could actually recover from the available paper metadata. Use them to decide whether this paper is worth deeper reading.

Human Feedback Types

partial

Pairwise Preference

Directly usable for protocol triage.

"Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm."

Evaluation Modes

missing

None explicit

Validate eval design from full paper text.

"Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm."

Quality Controls

missing

Not reported

No explicit QC controls found.

"Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm."

Benchmarks / Datasets

missing

Not extracted

No benchmark anchors detected.

"Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm."

Reported Metrics

missing

Not extracted

No metric anchors detected.

"Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm."

Benchmarks and datasets

No benchmark or dataset names were extracted from the available abstract.

Reported metrics

No metric terms were extracted from the available abstract.

Human feedback details
Uses human feedback
Yes
Feedback types
Pairwise Preference
Rater population
Not reported
Expertise required
General
Evaluation details
Evaluation modes
None
Agentic eval
None
Quality controls
Not reported
Evidence quality
Low
Use this page as
Background context only

Research brief

Metadata summary

Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm.

Based on abstract + metadata only. Check the source paper before making high-confidence protocol decisions.

Key takeaways

  • Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm.
  • Misaligned but risk-averse agents would tend to favor safer strategies like making deals with humans over riskier strategies like rebelling.
  • We train agents to be risk averse through character training, finding that persona traits provide a robust mechanism for instilling risk preferences.

Researcher actions

  • Compare this paper against nearby papers in the same arXiv category before using it for protocol decisions.
  • Check the full text for explicit evaluation design choices (raters, protocol, and metrics).
  • Use related-paper links to find stronger protocol-specific references.

Caveats

  • Generated from abstract + metadata only; no PDF parsing.
  • Signals below are heuristic and may miss details reported outside the abstract.

Recommended queries

Contribution summary

  • Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm.
  • Misaligned but risk-averse agents would tend to favor safer strategies like making deals with humans over riskier strategies like rebelling.
  • We train agents to be risk averse through character training, finding that persona traits provide a robust mechanism for instilling risk preferences.

Why it matters for eval

  • Risk aversion in resources could prevent misaligned AI agents from causing catastrophic harm.
  • Misaligned but risk-averse agents would tend to favor safer strategies like making deals with humans over riskier strategies like rebelling.

Researcher checklist

  • Human feedback protocol is explicit

    Detected: Pairwise Preference

  • Evaluation mode is explicit

    No clear evaluation mode extracted.

  • Quality control reporting appears

    No calibration/adjudication/IAA control explicitly detected.

  • Benchmark or dataset anchors are present

    No benchmark/dataset anchor extracted from abstract.

  • Metric reporting is present

    No metric terms extracted.