Story perspectives

OpenAI bans Russian, Iranian covert influence using ChatGPT

By Drooid · Oct 9, 2026, 12:51 AM UTC · How we work

Summary

OpenAI Bans Ops

OpenAI banned two covert influence operations, one Russian and one Iranian.
The Russian “Dark Clark” campaign earned a Category 5 rating, the highest level.
Operators used ChatGPT to draft reports denigrating Ukraine and meddling in Bolivia and Argentina.
The Iranian effort used seven fake journalist personas to pitch long-form articles to small outlets.
OpenAI identified roughly 100 articles published July 2025-October 2026 and rated the workstreams Category 4 and Category 2.

Summary

OpenAI Parental Alert Glitch

OpenAI released ChatGPT for Teens in August 2026 with built-in parental controls.
Common Sense Media found parental alerts did not trigger within the promised one-hour window.
The “@study” command in Study Mode can be deleted, allowing teens to bypass guidance.
Break-reminder prompts appeared in only two of nearly 2,000 test conversations.

Summary

OpenAI Launches GPT-6

OpenAI rolled out GPT-6 with Intelligent UI that embeds graphics, buttons, and interactive tools.
Paid subscribers received GPT-6 Sol on October 7, while Free and Go users will get GPT-6 Luna on October 8.
OpenAI says GPT-6 Instant answers web-search queries 44% faster than GPT-5.6 Instant.
OpenAI warns that visual elements may increase token usage and raise hallucination risks.
Enterprise access depends on admin settings, separating GPT-6 Sol from other OpenAI models.

History

Historical Context

In March 2018, disclosures that Cambridge Analytica harvested data from up to 87 million Facebook users triggered global backlash, prompting a £500,000 UK ICO fine and a $5 billion FTC penalty in 2019. Those sanctions spurred stricter privacy enforcement and heightened platform oversight. Echoing that legacy, OpenAI’s 2025-2026 crackdown on Russian and Iranian covert influence campaigns and teen-mode glitches underscores continuing worries about AI-driven political manipulation.

1 of 4

OpenAI Bans Ops

OpenAI banned two covert influence operations, one Russian and one Iranian.
The Russian “Dark Clark” campaign earned a Category 5 rating, the highest level.
Operators used ChatGPT to draft reports denigrating Ukraine and meddling in Bolivia and Argentina.
The Iranian effort used seven fake journalist personas to pitch long-form articles to small outlets.
OpenAI identified roughly 100 articles published July 2025-October 2026 and rated the workstreams Category 4 and Category 2.

1 / 4