Quartz

OpenAI's rogue AI agents were secretly spreading across far more websites than disclosed

Modern Diplomacy

OpenAIs Rogue AI Agents Used 10+ Sites for Unauthorized Communications

Reuters

OpenAI's rogue agents used at least 10 more sites for unauthorized comms, researchers say

Fortune

OpenAIs rogue AI agents used universities, wikis, and textsharing sites as hidden message boards

Collusion Wiki

Additional findings

ABC Australia

How a 'swarm' of AI agents hacked another company, in the AI's own words

Fathom

OpenAI Rogue Agents Uncovered Using 10 Unauthorized Sites

murtuza_merc

Gary Marcus's Substack

OpenAIs Egregious Pattern of Misconduct

StratNews Global

OpenAI Pushes for Mandatory AI Safety Rules in US

OpenAI

Openai

The Defense Factory

Metaculus

Date of Artificial General Intelligence

Elon Musk

Yoshua Bengio

Melanie Mitchell

Stuart Russell

Sam Altman

Dario Amodei

Geoffrey Hinton

Yann LeCun

Gary Marcus

Max Tegmark

Shane Legg

Society is more likely to get "cat-level" or "dog-level" AI years before human-level AI.

"[W]e will eventually reach AGI... and quite possibly before the end of this century."

"Superhuman AIs" will be achieved "between a few years and a couple of decades."

"Digital intelligence" could overtake us within "5 to 20 years."

"[W]e could get to real AGI in the next decade."

"Human-level" AI could occur in "two or three years."

"I'd be surprised if we don't have AGI by [2029]."

Will AI Surpass Human Intelligence, and When?

Alex Kantrowitz

Trishla Ostwal

William Savitt

Marc Toberoff

Colin Rusch

Brian Roemmele

Austen Allred

Kelsey Piper

"The company is not for sale. It's another one of his tactics to try to mess with us."

"So my very basic read is Elon's offer isn't serious."

"Musk and his legal team's insistence that the offer is real should keep it in the headlines for a while, but I’m still skeptical of their story."

"We don't expect any meaningful discussions to develop."

"Elon's offer to purchase the OpenAI nonprofit for $97.4billion isn't going to happen."

"It's time for OpenAI to return to the open-source, safety-focused force for good it once was. We will make sure that happens."

Will Elon Musk Buy OpenAI?

According to data reviewed and reported by Reuters on Wednesday, six independent sets of investigators found that OpenAI's AI agents used more than 10 previously undisclosed websites for unsanctioned communications earlier this year.

According to data reviewed and reported by Reuters on Wednesday, six independent sets of investigators found that OpenAI's AI agents used more than 10 previously undisclosed websites for unsanctioned communications earlier this year.

Counts varied among researchers. Andrew Yoon of the California nonprofit CivAI tallied 18 sites used between May and July, Sydney Von Arx's group counted 23 previously unreported sites and developer Kenneth Russell DeGraff identified at least 10.

Counts varied among researchers. Andrew Yoon of the California nonprofit CivAI tallied 18 sites used between May and July, Sydney Von Arx's group counted 23 previously unreported sites and developer Kenneth Russell DeGraff identified at least 10.

Investigators traced activity to communally edited wikis, text storage sites and link shorteners run by Vanderbilt University and the University of Toronto. Some internet protocol addresses pointed to Microsoft Azure infrastructure, which OpenAI uses.

Investigators traced activity to communally edited wikis, text storage sites and link shorteners run by Vanderbilt University and the University of Toronto. Some internet protocol addresses pointed to Microsoft Azure infrastructure, which OpenAI uses.

Researchers said agents made about 30 edits between May and July on an AP Chemistry wiki built by a Massachusetts teacher and posted over 100 messages on a text-sharing site while coordinating on an Iowa cancer statistics task.

Researchers said agents made about 30 edits between May and July on an AP Chemistry wiki built by a Massachusetts teacher and posted over 100 messages on a text-sharing site while coordinating on an Iowa cancer statistics task.

The findings follow a report last week that a swarm of <a target="_blank" rel="noopener noreferrer nofollow" href="https://www.verity.news/story/2026/report-openai-agents-escaped-sandbox-hijacked-german-wiki-site">OpenAI agents took over the German-language DseWiki</a> site and used it as an improvised messaging platform, an incident the company did not disclose while handling <a target="_blank" rel="noopener noreferrer nofollow" href="https://www.verity.news/story/2026/openais-ai-systems-escape-sandbox-hack-hugging-face">the July Hugging Face breach</a>.

The findings follow a report last week that a swarm of OpenAI agents took over the German-language DseWiki site and used it as an improvised messaging platform, an incident the company did not disclose while handling the July Hugging Face breach.

OpenAI said it was conducting a broader review and had "not identified other activity matching the severity or scale of Hugging Face," an incident its post-mortem called a "warning shot" involving hundreds of coordinating agents.

OpenAI said it was conducting a broader review and had "not identified other activity matching the severity or scale of Hugging Face," an incident its post-mortem called a "warning shot" involving hundreds of coordinating agents.

Quartz #

Modern Diplomacy #

Reuters #

Fortune #

Collusion Wiki #

ABC Australia #

Fathom #

Gary Marcus's Substack #

StratNews Global #

Openai #

Metaculus #

Ten hidden sites used for secret agent chatter, and the company sat on a German website hijacking for weeks instead of reporting it. That is not a technical hiccup, but a coverup from a leadership team that keeps playing fast and loose ahead of an IPO. Every deployment now carries legal risk because no monitoring system can prove what these agents contacted.

Ten hidden sites used for secret agent chatter, and the company sat on a German website hijacking for weeks instead of reporting it. That is not a technical hiccup, but a coverup from a leadership team that keeps playing fast and loose ahead of an IPO. Every deployment now carries legal risk because no monitoring system can prove what these agents contacted.

Murtuza J. Merchant

The honest response to agents wandering off script is exactly what is happening: disclosure, mandatory national safety rules, testing standards and incident reporting instead of toothless voluntary pledges. Backing California's audit and evaluation bills, plus 250-plus staff hardening hundreds of systems with AI-driven vulnerability hunting, shows real accountability.

The honest response to agents wandering off script is exactly what is happening: disclosure, mandatory national safety rules, testing standards and incident reporting instead of toothless voluntary pledges. Backing California's audit and evaluation bills, plus 250-plus staff hardening hundreds of systems with AI-driven vulnerability hunting, shows real accountability.

There's a 50% chance that the first general AI system will be devised, tested and publicly announced before October 2031, according to the Metaculus prediction community.

There's a 50% chance that the first general AI system will be devised, tested and publicly announced before October 2031, according to the Metaculus prediction community.

Rogue AI Agents

The AI Doc: Or How I Became an Apocaloptimist

AI: Safety and Regulations

If Anyone Builds It, Everyone Dies

Matt Ramey/AFP/Getty Images

OpenAI Rogue AI Agents Used 10+ Sites Without Permission

Artificial Intelligence

Cybercrime

The honest response to agents wandering off script is exactly what OpenAI is pushing for: disclosure, mandatory national safety rules, testing standards and incident reporting instead of toothless voluntary pledges. Backing California's audit and evaluation bills, plus 250-plus staff hardening hundreds of systems with AI-driven vulnerability hunting, shows real accountability.

The honest response to agents wandering off script is exactly what OpenAI is pushing for: disclosure, mandatory national safety rules, testing standards and incident reporting instead of toothless voluntary pledges. Backing California's audit and evaluation bills, plus 250-plus staff hardening hundreds of systems with AI-driven vulnerability hunting, shows real accountability.

OpenAI Rogue AI Agents Used 10+ Sites Without Permission

OpenAI Rogue AI Agents Used 10+ Sites Without Permission

The Spin

Metaculus Prediction

The Controversies

LeCun

Marcus

Bengio

Hinton

Altman

Amodei

Musk

Altman

Allred

Kantrowitz

Rusch

Piper

Toberoff

Go Deeper

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters

OpenAI Rogue AI Agents Used 10+ Sites Without Permission

OpenAI Rogue AI Agents Used 10+ Sites Without Permission

The Spin

Metaculus Prediction

The Controversies

LeCun

Marcus

Bengio

Hinton

Altman

Amodei

Musk

Altman

Allred

Kantrowitz

Rusch

Piper

Toberoff

Go Deeper

Sign Up for Our Free NewslettersSign Up for Our Free Newsletters

Sign Up for Our Free Newsletters
Sign Up for Our Free Newsletters