AI
Startups strip censorship from Chinese open-weight AI models for Western buyers
Rest of World details startups editing Chinese AI for Western firms
1 Nov 2 6:00 AM · 9d ago · 2 articles · 1 post · 2 sources · development 1 of 1
Rest of World reports that Hirundo, Lazarus and other startups are using machine unlearning and weight edits to harden Chinese open-weight models against cyberattacks, reduce bio-weapons knowledge, and strip state-aligned bias, with Hirundo CEO Ben Luria describing the process as 'like neurosurgery'.
“What we do is like neurosurgery, an intervention that's surgical and very specific.”
Ben LuriaBen Luria Co-founder and CEO of HirundoHirundo AI safety startupDeepSeek Chinese AI labCrowdStrike Cybersecurity firm
The whole story articlesposts the bright band is this development · numbered dots are the others · click one to jump
Nov 3Nov 4Nov 5Nov 6Nov 7Nov 8Nov 9yesterdaynow · 11:49 PM ET
What was reported 2 claims about this development
-
first by Rest of World, 9d ago
-
first by Rest of, 9d ago
All 1 developments of Startups strip censorship from Chinese open-weight AI… →
NewswiresMastodonReddit