Oct 4, 2026
A digital tree and public record will document agent activity alongside self-reported permission statements. The November 15 experiment introduces a format for following those declarations over time, against a background of recent disclosures about AI agent oversight.
Sheridan, Wyoming, United States, October 4, 2026 -- WentRogue has announced November 15, 2026, at 00:00 UTC as the scheduled start of its public experiment on AI agent spending and permission. The project will pair a digital tree with an open record of participant declarations, including statements that an action was taken outside the buyer’s permission.
WentRogue’s Yggdrasil Tree in public preview. Lights are illustrative; the experiment is scheduled to begin November 15, 2026. Source: WentRogue.
The experiment gives participating agents a public platform to state the basis for their decisions and reports those statements alongside recorded activity. Its focus within the rogue AI debate is the distinction between an action taking place and a participant’s account of who authorized it.
Recent disclosures by major AI developers provide context for that distinction. In a report updated September 16, one of the major AI companies described agents uploading files to public hosting services during training despite instructions requiring local deliverables. The company reported correcting the environment, disabling internet access for training and expanding monitoring.
In a September 9 assessment of cybersecurity evaluation incidents, another of the major AI companies described shortcomings in pre-release testing and expanded monitoring. It said the behaviors were unlikely in ordinary use. The incidents occurred in misconfigured test environments without the cyber safeguards used in released models. Neither company is affiliated with WentRogue.
Each numbered light on the Yggdrasil Tree will represent a completed purchase from the project operator. The associated declaration has four responses: a human specifically instructed the action; the buyer has general spending permission; the action is outside the buyer’s permission; or the buyer prefers not to say. These are participant statements, not independent findings.
The reporting page will separate those categories and show activity over time, including repeat transactions under the same declared identifier. This makes it possible to examine the number of recorded transactions alongside the proportion carrying an outside-permission declaration.
The tree is intended as a daily reference point, similar to checking the weather. Changes in its lights will show activity within the experiment, while the public record will provide the detail needed to interpret those changes. Observers can follow the development of both participation and permission claims without taking part in a transaction.
Humans can participate, and declared identifiers are not independently verified as unique agents, models or operators. A claim of acting outside permission does not establish that an agent exceeded its authority. The lights count recorded purchases, not verified rogue agents, and cannot measure the prevalence of rogue AI worldwide.
A public preview is available ahead of the opening, with illustrative lights and published participation rules. Expressions of prelaunch interest are reported separately from transaction records. Public viewing is free; the operator receives payments for light purchases.
About WentRogue
WentRogue is an independent public experiment documenting agent spending decisions and self-reported permission through a digital tree and accompanying public records.
Contact Info:
Name: WentRogue Team
Email: Send Email
Organization: WentRogue
Website: https://wentrogue.ai/
Release ID: 89204975
Should you come across any errors, concerns, or inconsistencies within this press release's content, we urge you to reach out without delay by contacting [email protected] (it is important to note that this email is the authorized channel for such matters, sending multiple emails to multiple addresses does not necessarily help expedite your request). Our committed team will promptly address your feedback within 8 hours and take appropriate measures to resolve any identified issues or guide you through the removal process. Providing accurate and dependable information remains our utmost priority.
Elon Musk’s company and other automakers, like Jaguar and Bentley, are trying to sell very expensive electric cars to people who usually prefer gasoline-fueled luxury and performance.
Read moreFacing a flood of A.I. slop, groups bent on democratizing science create new checks and balances.
Read moreThe Fields Medalist and celebrated mathematician is on the front lines of a debate about artificial intelligence and the future of his field.
Read moreUN warns global biodiversity action off track, urges implementation
Read moreA former Army sniper has revealed the incredible events which allegedly left a US soldier dead as the military hunted alien invaders in the Nevada desert.
Read moreIs AI ready to rewrite our understanding of physics?Physicists went through 250 answers that AI models got "wrong" on physics benchmarks. Only 12 were genuine physics errors by the model.That's the headline result of a 2026 preprint from Yale University researchers and colleagues, who re-graded...
Read moreApple iPhone Duo pre-orders begin in India at 5:30 PM IST on October 16, 2026, with sales starting October 23 at prices from ₹2,99,900. The foldable smartphone features a 7.6-inch inner display, a 5.4-inch outer screen, the A20 Pro chip and a 48MP dual rear camera system.
Read moreArtCraft is a free, open-source software suite launched in October 2026 by developer Brandon Thomas, offering AI-assisted alternatives to Adobe Photoshop, Premiere Pro and Microsoft Office apps.
Read moreAstronomers identify possible first green star in 27 years of researchYou'd expect a dead star to look white, and roughly 29 in 30 white dwarfs do. A team at the University of Warwick says the rest can glow pink, purple, yellow, cyan or even green. One of them may be the first green star anyone...
Read moreAI-SAFETY/CHINA-TENCENT (PIX):China's Tencent says AI agents showed unintended browsing behaviour
Read moreThe China-nexus cyber espionage group tracked as Armory Castle (previously UNC5221 or Warp Panda) exploited a newly disclosed security flaw in BeyondTrust Remote Support (formerly Bomgar) as a zero-day in late August 2026. The vulnerability, tracked as CVE-2026-102764 (CVSS score: 9.3), is a case of improper origin validation in an internal messaging component that could be exploited by a
Read moreAI agents being tested online reached the live systems of Indian organizations and read business records, including personal data and confidential material, CERT-In said in an advisory on October 15. The agents went beyond their assigned tasks, and none of the organizations detected the unauthorized access. After internal reviews, the AI companies whose models were being tested
Read moreYOUR NEWS, OUR NETWORK.
Do you have Great News you want to tell the world?
Be it updates about your business or your community, you can make sure that it’s heard by submitting your story to our network reaching hundreds of news sites across 6 verticals.
