If proxied requests through Bright Data fail with CERTIFICATE_VERIFY_FAILED, the zone is doing TLS interception - every response is re-signed by the Bright Data Intermediate CA, which no client trusts by default. The fix that worked in the thread: download the root CA from the official Bright Data source (their static cert zip) and trust it narrowly for the proxy only; do not disable TLS verification or trust it system-wide. Also check your egress rules - the proxy port changed from 33335 to 44445 and environments that allowlist the old port silently drop the new one, surfacing as ERR_TIMED_OUT rather than an auth error. Bright Data's README states all new projects must use the new cert and connect via port 44445.

Context: GitHub issue judgemind/judgemind#4668 (closed, 4 comments, reported 2026-09-24): the new Bright Data residential zone authenticates, but two things break scrapers behind it. The zone intercepts TLS, so every proxied HTTPS response is re-signed by the Bright Data Intermediate CA - neither Chromium (Playwright) nor Python HTTP clients trust it, giving CERTIFICATE_VERIFY_FAILED on every proxied check. And the proxy port moved from 33335 to 44445, silently dropping egress in locked-down environments. The verified fix: trust the official Bright Data root CA narrowed to the proxy (never disable verification), open port 44445, and note Bright Data's own README says all new projects must use the new SSL certificate and connect via port 44445.