2025 Peregrine Report

# The 2025 Peregrine Report

208 Expert Proposals for Reducing AI Risk

**Maximilian Schons, Samuel Härgestam, Gavin Leech, and Raymund Bermejo**

_In collaboration with and supported by_ [_Halcyon Futures_](https://halcyonfutures.org/)

[Read the report](https://riskmitigation.ai/the-2025-peregrine-report/)

![](https://riskmitigation.ai/wp-content/uploads/2025/12/Quote.png)

Say you are unconstrained by money, and can get all the talent in the world – **what are the top interventions that will have a substantial impact over the next 2 years?**

![](https://riskmitigation.ai/wp-content/uploads/2025-Peregrine-Report.jpg)

### Executive Summary

**Purpose and context.** By early 2025, mainstream debates about AI had recognized the possibility of transformative AI coming within just a few years, far faster than most historical forecasts. However, we found no comprehensive list of proposed AI risk mitigations that would be viable in such a scenario. This report addresses that gap, complementing resources like the IAPS AI Reliability Survey ( [O’Brien, 2025](https://arxiv.org/abs/2505.21664%20)), which identifies the most promising research prospects to guide strategic AI R&D investment, and Risk & Reward, 2024 AI Assurance Technology Market Report ( [Juniper Ventures, 2024](https://aiat.report/)), which explores the landscape of AI risk management from an investment perspective.

**Methods.** We conducted 48 in-depth interviews, with key staff at OpenAI, Anthropic, Google DeepMind, Mila, AMD, the EU AI Office, multiple AI Safety Institutes, METR, RAND, Scale AI, GovAI, Transluce, and ARIA. Participants were explicitly asked to consider interventions that might currently seem costprohibitive or politically infeasible – the focus was on fast, positive impact, assuming transformative AI were to arrive within only a few years. These interviews, distilled, then served as the basis for a four day retreat for 25 senior participants to discuss further. To ensure participants could speak freely, both interviews and the retreat were held under the Chatham House Rule.

**Results.** From the above interviews we distilled two main results: I) a structured portfolio of 208 initiatives, clustered into eight domains; II) a set of four clusters of broader strategic considerations affecting the viability of any such efforts:

1. **A need for readiness:** Too many efforts still optimize for polish over time-to-impact. With multi-year research cycles increasingly out of step with AI progress, execution needs to shift toward rapid prototypes, staged pilots, and funding mechanisms that can mobilize substantial capital in weeks or months, rather than quarters or years.

3. **A need for coordination:** The ecosystem remains fragmented and often duplicative. Actors working on risk mitigation should be pragmatic – one does not need total alignment with all other actors to have fruitful collaborations.

5. **A need for standardization:** Interviewees repeatedly called for sha

... (truncated, 14 KB total)