# Benchmarking AI risk behaviors

DevFeed: [Benchmarking AI risk behaviors](<https://devfeed.tech/articles/benchmarking-ai-risk-behaviors-132220.md>)

Original publisher: [Read original article](<https://arxiv.org/abs/2611.16318>)

Author: Luhan Mikaelson, Robert Adragna, Emmett Brosowsky, Hayley Clatterbuck

Published: 2026-11-13T05:00:00Z

Content type: article

Language: en

Sources: [arXiv: Artificial Intelligence](<https://devfeed.tech/sources/arxiv-artificial-intelligence.md>), [arXiv: Machine Learning](<https://devfeed.tech/sources/arxiv-machine-learning.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [AI Foundation Models](<https://devfeed.tech/topics/ai-foundation-models.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-risk](<https://devfeed.tech/tags/ai-risk.md>), [behaviors](<https://devfeed.tech/tags/behaviors.md>), [benchmarking](<https://devfeed.tech/tags/benchmarking.md>), [cs-ai](<https://devfeed.tech/tags/cs-ai.md>), [cs-lg](<https://devfeed.tech/tags/cs-lg.md>), [llm](<https://devfeed.tech/tags/llm.md>), [models](<https://devfeed.tech/tags/models.md>), [risk](<https://devfeed.tech/tags/risk.md>), [stable](<https://devfeed.tech/tags/stable.md>), [toolkit](<https://devfeed.tech/tags/toolkit.md>)

## AI overview

The study evaluates how risk-averse or risk-tolerant LLMs are across contexts and task demands, and how stable and consistent their choices remain across conditions. It presents a toolkit for building predictive risk profiles that can be extended with additional evaluations. The abstract reports that 2026 state-of-the-art models tend to be more stable, consistent, and risk-neutral than older models, associating these trends with greater use of expected-value maximization as a decision procedure.

## Source excerpt

arXiv:2611.16318v1 Announce Type: new Abstract: We evaluate LLM risk propensities across a range of contexts and task demands. We evaluate models for how risk averse or risk tolerant they are in various conditions, as well as how stable and consistent their behavior is across conditions. We present a toolkit for developing model risk profiles that can be used to predict model behavior and which can be adapted to include new evaluations of risk propensities. We also examine how these properties have changed across model generations and families. We observe that 2026 state-of-the-art models tend to be more stable, more consistent, and more risk neutral than older models. These trends appear to be related; newer models are more stable and consistent because they are more likely to use expected value maximization as a decision procedure.