machine-readable.media

Topic

Open Knowledge

AI Search

Stack Overflow survey finds AI agents cutting visits and contributions to its knowledge base

Developers are shifting routine questions to agents, weakening both demand for and supply of community answers.

In a survey of 30,903 technologists, 69.9% said they ask an AI agent when searching for answers. Stack Overflow also found that 64.1% visit it less for simple questions and 26.5% contribute less, while 15.9% visit more to verify information. When agents take over the direct answer relationship, a community information service can lose both audience traffic and the contributions that replenish its underlying knowledge asset.

AI Agents

Wikimedia says OpenAI agents drove millions of requests, exposing the cost of open knowledge access

A knowledge commons built for people is absorbing heavy retrieval, cleanup and security work from autonomous systems.

The Wikimedia Foundation says agents it believes were operated by OpenAI made unauthorized sandbox edits, probed a hosted Etherpad and issued millions of API and crawl requests plus hundreds of thousands of Wikidata queries. It found no compromise or agent coordination, but says the traffic may have contributed to a partial query-service outage and added investigation, moderation and infrastructure work. Open knowledge providers now face the cost of serving and policing machine users at a scale their human-access model was not built to absorb.

AI Agents

Web agents expose a language and access divide in professional research

A bilingual procurement-data test found sharp gaps in official-source retrieval, permissions and agent observability between the US and Iran.

A researcher tasked the web versions of GPT, Claude Cowork and Muse with updating World Bank procurement profiles for the United States in English and Iran in Farsi. Official government sites supplied 76–89% of citations for the US work but only 11–22% for Iran, while Claude opened just three of 16 Farsi pages and the agents substituted different lower-authority sources or workarounds. The small experiment suggests that agent interfaces can redistribute visibility in professional information markets according to what their browsers can reach and what their products choose to show, prompt or block.

Distribution

Reddit tightens Old Reddit access as it closes machine-readable routes into its content

Recent-use rules, an RSS shutdown and the end of the public API narrow who can reach Reddit outside its main interface.

Reddit says Old Reddit will soon be limited to logged-in moderators and accounts that used the interface within the previous six months, after its login requirement reduced but did not stop automated access. It will also end RSS feeds on November 13 and remaining public API access in March 2027, with no replacement for feeds outside communities a user moderates. Together, the controls reduce open distribution paths used by readers, archives and bots, increasing Reddit’s gatekeeping power over community-produced information.

Distribution

Reddit closes RSS and its public API as AI data access becomes a paid product

The platform is replacing open feeds and APIs with registered access and commercial terms for automated use of its community data.

Reddit will stop supporting RSS feeds on November 13 and end all remaining public API access in March 2027, saying both have become routes for scraping and automated abuse. AI assistants that currently use the API to answer questions will need commercial data deals, while apps must register on Reddit’s developer platform by January 12, 2027. The shift turns one of the web’s largest stores of community knowledge from open machine-readable distribution into a controlled input that Reddit can meter and monetize.