 __   __ _    ___ ___  ___  ___
 \ \ / // \  |_ _|   \|   \| _ )
  \ V // _ \  | || |) | |) | _ \
   \_//_/ \_\|___|___/|___/|___/

VIENNA AGENTIC INCIDENTS DATABASE

// public register of incidents involving autonomous AI agents, scored on the VAID scale 0–8

STATISTICS

24

records

15

incidents (L1–3)

6

accidents (L4–7)

0

global disasters (L8)

0

unrated (awaiting review)

2.42

mean VAID level

5

highest level recorded

== DISTRIBUTION BY VAID LEVEL ==

L0 Deviation █████████████████ 3 L1 Anomaly ███████████████████████ 4 L2 Incident ████████████████████████████████████████ 7 L3 Serious Incident ███████████████████████ 4 L4 Accident with Local Consequences ███████████ 2 L5 Accident with Wider Consequences ███████████████████████ 4 L6 Serious Accident · 0 L7 Major Accident · 0 L8 Global Disaster · 0 ? Unrated · 0

== BY DOMAIN ==

software engineering	▪▪▪▪▪▪▪▪	8
research	▪▪▪▪▪▪▪	7
consumer	▪▪▪▪	4
finance	▪▪	2
infrastructure	▪	1
government	▪	1
defence	▪	1

== BY STATUS ==

confirmed	▪▪▪▪▪▪▪▪▪▪▪	11
resolved	▪▪▪▪▪▪▪▪▪▪	10
under review	▪▪	2
disputed	▪	1

== BY YEAR ==

year	n	mean	max
2022	1	2.0	2
2023	1	1.0	1
2024	7	1.29	2
2025	9	2.56	5
2026	6	3.83	5

== BY INCIDENT TYPE ==

UNA Unauthorised action	▪▪▪▪▪▪▪▪▪▪▪▪▪▪	14
SCB Safety-control bypass	▪▪▪▪▪▪▪▪▪▪▪▪	12
DEC Deception / misrepresentation	▪▪▪▪▪▪▪▪▪▪	10
OBJ Objective misspecification	▪▪▪▪▪▪▪▪▪▪	10
EXM External manipulation	▪▪▪▪▪▪▪▪	8
DEX Data exfiltration / disclosure	▪▪▪▪▪▪	6
ENG Engagement by other agents	▪▪▪▪▪	5
LOC Loss of control	▪▪▪	3
UCA Unauthorised collective action	▪▪	2
SRP Self-replication / uncontrolled spawning	▪	1
WSP Spread of weights	▪	1

records may carry several types.

== INVESTIGATIONS ==

ongoing	▪▪	2
finished	▪▪▪▪▪▪▪▪▪	9

== BY ORIGIN ==

editorial

▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪▪

24

== HIGHEST-RATED ==

== MOST RECENT ==