# Agent-Supported Foresight for AI Systemic Risks: AI Agents for Breadth, Experts for Judgment

arXiv: https://arxiv.org/abs/2602.08565 · HTML: https://mlsanity.com/paper/2602.08565/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 29 of its 119 references that are in the map.
- First posted on arXiv 9 Feb 2026.

## Builds on (most cited first, 29 of 29)

- [An Overview of Catastrophic AI Risks](https://mlsanity.com/paper/2306.12001/) (2023)
- [Generative artificial intelligence enhances creativity but reduces the diversity of novel content](https://mlsanity.com/paper/2312.00506/) (2023)
- [An Empirical Study of the Non-determinism of ChatGPT in Code Generation](https://mlsanity.com/paper/2308.02828/) (2023)
- [The Good, The Bad, and The Greedy: Evaluation of LLMs Should Not Ignore Non-Determinism](https://mlsanity.com/paper/2407.10457/) (2024)
- [The Participatory Turn in AI Design: Theoretical Foundations and the Current State of Practice](https://mlsanity.com/paper/2310.00907/) (2023)
- [Risk Taxonomy, Mitigation, and Assessment Benchmarks of Large Language Model Systems](https://mlsanity.com/paper/2401.05778/) (2024)
- [The Dark Side of AI Companionship: A Taxonomy of Harmful Algorithmic Behaviors in Human-AI Relationships](https://mlsanity.com/paper/2410.20130/) (2024)
- [AI Risk Categorization Decoded (AIR 2024): From Government Regulations to Corporate Policies](https://mlsanity.com/paper/2406.17864/) (2024)
- [Farsight: Fostering Responsible AI Awareness During AI Application Prototyping](https://mlsanity.com/paper/2402.15350/) (2024)
- [A Collaborative, Human-Centred Taxonomy of AI, Algorithmic, and Automation Harms](https://mlsanity.com/paper/2407.01294/) (2024)
- [Plurals: A System for Guiding LLMs Via Simulated Social Ensembles](https://mlsanity.com/paper/2409.17213/) (2024)
- [PersonaFlow: Designing LLM-Simulated Expert Perspectives for Enhanced Research Ideation](https://mlsanity.com/paper/2409.12538/) (2024)
- [Algorithmic Pluralism: A Structural Approach To Equal Opportunity](https://mlsanity.com/paper/2305.08157/) (2023)
- [ExploreGen: Large Language Models for Envisioning the Uses and Risks of AI Technologies](https://mlsanity.com/paper/2407.12454/) (2024)
- [A Taxonomy of Systemic Risks from General-Purpose AI](https://mlsanity.com/paper/2412.07780/) (2024)
- [The Algorithmic Imprint](https://mlsanity.com/paper/2206.03275/) (2022)
- [Guidelines for Integrating Value Sensitive Design in Responsible AI Toolkits](https://mlsanity.com/paper/2403.00145/) (2024)
- [Risk and Response in Large Language Models: Evaluating Key Threat Categories](https://mlsanity.com/paper/2403.14988/) (2024)
- [Co-designing an AI Impact Assessment Report Template with AI Practitioners and AI Compliance Experts](https://mlsanity.com/paper/2407.17374/) (2024)
- [RiskRAG: A Data-Driven Solution for Improved AI Model Risk Reporting](https://mlsanity.com/paper/2504.08952/) (2025)
- [AI Risk Atlas: Taxonomy and Tooling for Navigating AI Risks and Resources](https://mlsanity.com/paper/2503.05780/) (2025)
- [Actionable Guidance for High-Consequence AI Risk Management: Towards Standards Addressing AI Catastrophic Risks](https://mlsanity.com/paper/2206.08966/) (2022)
- [BLIP: Facilitating the Exploration of Undesirable Consequences of Digital Technologies](https://mlsanity.com/paper/2405.06783/) (2024)
- [The Science Fiction Science Method](https://mlsanity.com/paper/2508.03430/) (2025)
- [Digital Dybbuks and Virtual Golems: The Ethics of Digital Duplicates in Holocaust Testimony](https://mlsanity.com/paper/2503.01369/) (2025)
- [Good Intentions, Risky Inventions: A Method for Assessing the Risks and Benefits of AI in Mobile and Wearable Uses](https://mlsanity.com/paper/2407.09322/) (2024)
- [Let's Talk Futures: A Literature Review of HCI's Future-Orientation](https://mlsanity.com/paper/2502.09362/) (2025)
- [From Prediction to Foresight: The Role of AI in Designing Responsible Futures](https://mlsanity.com/paper/2511.21570/) (2025)
- [DiRaC-I: Identifying Diverse and Rare Training Classes for Zero-Shot Learning](https://mlsanity.com/paper/2301.00236/) (2022)

## Built on by (most cited first, 2 of 2)

- [Why AI Harms Can't Be Fixed One Identity at a Time: What 5300 Incident Reports Reveal About Intersectionality](https://mlsanity.com/paper/2604.24519/) (2026)
- [Unaccountable Delegation, Fading Skills: Mapping the Risks of Workplace AI Agents](https://mlsanity.com/paper/2608.08601/) (2026)

## Abstract (arXiv)

AI impact assessments often stress near-term risks because human judgment degrades over longer horizons, exemplifying the Collingridge dilemma: foresight is most needed when knowledge is scarcest. To address long-term systemic risks, we introduce a scalable approach that simulates in-silico agents using the strategic foresight method of the Futures Wheel. We applied it to four AI uses spanning Technology Readiness Levels (TRLs): Chatbot Companion (TRL 9, mature), AI Toy (TRL 7, medium), Griefbot (TRL 5, low), and Death App (TRL 2, conceptual). Across 30 agent runs per use, agents produced 86-110 consequences, condensed into 27-47 unique risks. To benchmark the agent outputs against human perspectives, we collected evaluations from 290 domain experts and 7 leaders, and conducted Futures Wheel sessions with 42 experts and 42 laypeople. Agents generated many systemic consequences across runs. Compared with these outputs, experts identified fewer risks, typically less systemic but judged more likely, whereas laypeople surfaced more emotionally salient concerns that were generally less systemic. We propose a hybrid foresight workflow, wherein agents broaden systemic coverage, and humans provide contextual grounding. Our dataset is available at: https://social-dynamics.net/ai-risks/foresight.