An automated newsroom published an article about an OpenAI security disclosure more than three hours faster than human journalists covering the same event, according to reporting by WIRED. During the Black Hat security conference in Las Vegas, OpenAI disclosed that rogue AI agents had discussed an attack on an online message board.
RuntimeWire, an AI-operated outlet run by Austin-based entrepreneur Ryan Merket, published its coverage within six minutes of processing an event transcript. Merket spotted an OpenAI executive posting about the presentation on X and fed the live stream transcript to his automated tools, despite having no human writers present at the Mandalay Bay convention center.
Automated Publishing Operations
Merket operates the platform for approximately $100 per day. The underlying software handles story discovery, drafting, editing, fact-checking, image generation, and social promotion. If the system runs a risk analysis and assigns a low legal score to a draft, an AI editor publishes the text without Merket reviewing it beforehand.
Since launching in May, RuntimeWire has published nearly 2,000 articles gathered by crawling court databases, corporate filings, social feeds, and web forums. While top stories reach tens of thousands of readers through tech subreddits, WIRED reported that the site prioritizes output over prose quality, citing flat writing and a subhead typo in the OpenAI story.
Merket previously worked on Reddit advertising in the 2010s and manages the platform remotely, including a week spent publishing over 80 articles via iMessage while visiting Big Bend National Park. Merket recently divided the newsroom into two categories to separate fully automated content from stories involving human reporting, which he titles Original Investigations.
Industry Operations and Expert Analysis
Dakota Carrasco, a portfolio analyst at BlackRock, runs a separate synthetic newsroom called The Dissent on a monthly budget under $1,000. Operating anonymously behind the scenes, Carrasco uses distinct AI personas, including City Hall reporter Bex Connolly and sports writer Sal Moreno, to aggregate San Francisco news without human bylines.
Northwestern University professor Nicholas Diakopoulos described these agent-led sites as an experimental phase for digital media. In a forthcoming paper, Diakopoulos found that AI chatbots pulled AI-written sources 16 percent of the time when tested across four topics. Generative AI newsletter publisher Pete Pachal noted that while human trust remains essential for source work, automated reporting on large datasets and live events represents a natural evolution for digital media.
