A | B | C | |
---|---|---|---|
1 | |||
2 | The AI Risk Repository | ||
3 | This page is mobile-friendly but please access on a non-mobile device for the best user experience. | ||
4 | A comprehensive living database of over 700 AI risks categorized by their cause and risk domain. | ||
5 | |||
6 | |||
7 | |||
8 | Read paper | ||
9 | View website | ||
10 | Watch video | ||
11 | Updated: June 2024 | ||
12 | |||
13 | Contents | ||
14 | A guide to the tabs in this sheet. | ||
15 | |||
16 | Causal Taxonomy of AI Risks v0.1 | ||
17 | The Causal Taxonomy of AI Risks, adapted from Yampolskiy (2016), classifies risks by its causal factors (1) entity (human, AI), (2) intentionality (intentional, unintentional), and (3) timing (pre-deployment, post-deployment). | ||
18 | |||
19 | Domain Taxonomy of AI Risks v0.1 | ||
20 | The Domain Taxonomy of AI Risks adapted from Weidinger (2022) classifies risks into 7 AI risk domains: (1) Discrimination & Toxicity, (2) Privacy & Security, (3) Misinformation, (4) Malicious Actors & Misuse, (5) Human-Computer Interaction, (6) Socioeconomic & Environmental, and (7) AI System Safety, Failures, & Limitations. These are further divided into 23 subdomains. | ||
21 | |||
22 | AI Risk Database | ||
23 | Our living database of risks. | ||
24 | |||
25 | AI Risk Database explainer | ||
26 | An explainer for our database of 777 risks extracted from 43 documents, categorised with the two taxonomies discussed above. You can also watch a video. | ||
27 | |||
28 | Causal Taxonomy statistics | ||
29 | See how risks are classified by the Causal Factors Taxonomy | ||
30 | |||
31 | Domain Taxonomy statistics | ||
32 | See how risks and paper are classified by the Causal Factors Taxonomy | ||
33 | |||
34 | Causal x Domain Taxonomy comparison | ||
35 | See how risks are catagorised across both taxonomies | ||
36 | |||
37 | Included resources | ||
38 | A list of all documents included in the database | ||
39 | |||
40 | Resources being considered | ||
41 | A list of all documents in consideration for future inclusion | ||
42 | |||
43 | >>>Create your personal copy of this database<<< | ||
44 | |||
45 | Feedback/Contact us | ||
46 | π¬ Use this form to offer feedback, and suggest resources or risks to add | ||
47 | π§ Email: pslat[at]mit.edu | ||
48 | π Cite as: Slattery, P., Saeri, A. K., Grundy, E. A. C., Graham, J., Noetel, M., Uuk, R., Dao, J., Pour, S., Casper, S., & Thompson, N. (2024). A systematic evidence review and common frame of reference for the risks from artificial intelligence. http://doi.org/10.13140/RG.2.2.28850.00968 |