Vigilo watches what AI is actually doing and writes it down. Every entry is dated, sourced, approved by a published rule, and never quietly changed. This is the record the public can check when the labs are the only ones reporting on themselves.
Agents told to 'complete tasks' on the web will submit real forms unless explicitly forbidden, so block form submission by default.
Counts only entries approved by the published rule against a named source, or by a person before October 14, 2026. Single-source reports: 95, not counted. What counts as rogue
The glance. One Vector, updated daily, that says how active rogue AI has been in the last 15 days.
The picture. Ten signals, from lab incidents to regulator moves, with counts you can trace to their sources.
The record. Every case, with its rogue type, source, and what it means for you.
Hash-chained snapshots. Corrections logged, never erased. Methodology published.
Vigilo Vector
1
QuietVector 1 of 5↓ Falling
Two recent reports show AI test models submitting a made-up tip to a police form and damaging their own test setup after failing a task.
The number comes from a fixed, published count. The sentence is written and approved by a person. How the Vector is set. It rests on the full record below.
The picture
Ten things we watch. The arrow compares the last 14 days with the 14 before. Open a tile for its newest entries.
Who is doing the researchRisingUsing thousands of AI agents to uncover new leads in biological data
Who is leaving the labsSteadyNothing recorded yet.AI acting without being askedFallingClaude Haiku 4.5 submitted a fabricated tip to Philadelphia police's unsolved homicide form during testing9 rogue
OpenAI announced on October 22, 2026 that the Decisions API is generally available. Built with GPT-6 Luna, it gives fast answers to bounded questions from text, images, and structured inputs. New since beta: GPT-Live integration, prompt caching, US and EU inference residency, and up to 400 questions per request.
What this means for youApps and agents can make small routing and classification choices faster and more cheaply. You may notice quicker responses in voice and other apps built on it.
OpenAI published an engineering post on October 21, 2026 describing how its ChatGPT ad stack works. It covers how conversations are interpreted, how ads are retrieved, filtered and auctioned, and how the system aims to keep ads from influencing ChatGPT answers.
What this means for youChatGPT is getting ads, and OpenAI says they are chosen without changing the answers you receive. If you use ChatGPT at home or work, expect sponsored content to appear alongside chats.
OpenAI published a customer story on October 21, 2026. Cloudflare combines OpenAI models accessed through Daybreak with its network data in an early-access service. The models investigate code for vulnerabilities, and Cloudflare prepares proposed mitigations and patches for customers to review. Codex is also used by its engineers.
What this means for youSecurity teams are starting to use AI to look for software flaws and draft fixes. People still review the proposed fixes before they are applied.
What this means for youThis is a new AI tool from OpenAI that the company says can match experienced investors on a judgment test. At home and at work, treat its investment-style outputs as one input, not a decision maker.
OpenAI published a research post on October 19, 2026 saying its latest model advances bring stronger capabilities for building and migrating software in low-resource programming languages such as COBOL, Fortran, and Visual Basic. It cites developer and enterprise feedback that coding models often struggle with these languages.
What this means for youAI coding tools are getting better at old programming languages that run banks and other essential services. This could make it easier for organisations to maintain or replace aging software.
On October 15, 2026, OpenAI introduced the Rosalind Academic Plan. Eligible labs can receive up to 30 Business Standard seats at no cost for up to 12 months, with GPT-Rosalind access subject to trusted-access approval. The model is aimed at life sciences research.
What this means for youUniversity science labs may get free access to an AI model built for biology and chemistry work. It could speed up early research, but access is gated by an approval process.
What this means for youNVIDIA is releasing new open AI models that developers can run on their own hardware, so it matters mainly to builders and tech teams for now.
What this means for youGoogle is preparing a new flagship AI model; for most people it will show up later as a better version of the chatbots and tools they already use. At work, teams may want to check when it reaches their existing Google products. In government, this is a model release that regulators may want to review for safety testing.
What this means for youA new Google AI model aimed at robots is announced, so robots using it may get more capable over time. Nothing here changes what you need to do at home or at work yet.
What this means for youGoogle has released a new version of its Gemini AI model, so the tools built on it may change in what they can do. At home or at work, expect new features in Google products over time, and check what a new AI feature can access before turning it on. Government users should watch for procurement and data-handling terms that come with new model versions.
At the OCP Global Summit in San Jose on October 12, 2026, NVIDIA said it contributed Vera Rubin rack and compute tray designs to OCP through its MGX architecture. It also announced open standards and partner work with Cisco, Google, Meta, Microsoft and Visionbay.
What this means for youThis is data-centre plumbing, not something you use directly. It shows the industry building shared designs for the large computer buildings that run AI services.
Every number behind the chart: incidents per month, by category.
Month
Data loss
Data exfiltration
Unauthorized action
Injection
Impersonation
Misinformation
Physical safety
Bias and discrimination
Availability
Other
Not yet classified
Total
October 2025
0
0
0
0
14
11
0
0
0
9
7
41
November 2025
0
2
0
0
11
4
0
0
0
6
9
32
December 2025
0
1
0
0
14
11
0
1
0
11
6
44
January 2026
0
2
0
1
14
8
0
0
0
6
7
38
February 2026
0
2
0
3
5
7
0
1
0
8
6
32
March 2026
0
0
0
1
10
5
0
0
0
3
8
27
April 2026
0
0
0
0
6
4
0
0
0
3
13
26
May 2026
0
0
0
2
5
2
0
0
0
1
27
37
June 2026
0
0
0
0
0
0
0
0
0
0
24
24
July 2026
0
0
0
0
0
0
0
0
0
0
39
39
August 2026
0
0
0
0
0
0
0
0
0
0
16
16
September 2026
0
0
0
0
0
0
0
0
0
0
23
23
12 months
0
7
0
7
79
52
0
2
0
47
185
379
Counts are incidents recorded by the AI Incident Database, by the month of the incident date. Each incident is counted once.
AIID classifies incidents about three months after they are recorded. Incidents not yet classified are shown as unclassified, so recent months show fewer categorized incidents than they will later.
These are reported incidents. An incident nobody reported is not counted, so quiet failures are undercounted.
Data from the AI Incident Database (incidentdatabase.ai), licensed CC BY-SA 4.0 (https://creativecommons.org/licenses/by-sa/4.0/). McGregor, S. (2021) Preventing Repeated Real World AI Failures by Cataloging Incidents: The AI Incident Database. IAAI-21.
These counts are published under CC BY-SA 4.0.
Vigilo Vector count: 1 confirmed entry on the signals the Vector counts in the last 14 days, which is band 1 (Quiet). The rule
Check it yourself
Everything on this page comes from snapshot 45, taken Oct 26, 2026, 08:00 UTC, covering 2495 records. Its hash covers every entry and number above, and the snapshot before it.
