Websites taken offline worldwide after critical PHP flaws disclosed
publishers 30articles 30first reported 21 Oct, 17:00 UTCdeveloping since 21 Oct · 9 editions
Government agencies and companies worldwide took websites offline on Wednesday after critical security flaws were disclosed in PHP, a programming language estimated by security firm Wiz to power about 70% of websites.
The vulnerabilities could allow attackers to execute unauthorized commands on servers without a password, potentially giving them access to data and control of affected websites. Because the flaws lie in PHP’s core system rather than in the code of individual sites, unpatched installations may be exposed across online stores, news outlets and government services.
Fortune reported that a researcher at security firm Calif discovered a flaw in August using an advanced OpenAI model and alerted the nonprofit organization that manages PHP. Information about the vulnerability leaked before patches were released, and Amazon Web Services officials said attackers then began attempting to exploit it “broadly.” Sky News reported that a major technology company had recorded “millions” of exploitation attempts.
PHP’s developers released a patch on Wednesday. The U.S. National Security Agency advised organizations to install it immediately or take vulnerable systems offline. Government and corporate shutdowns followed as administrators raced to secure affected servers.
HOW THIS STORY WAS MADE
Written from 4 articles, 2 with the publisher's own text; 4 independent newsrooms once syndicated copies count as one; this version written 54 h after the record first saw the story; the editor kept it as written.
- Publishers
- 30
- Source articles
- 30
- Given to the writer
- 4, 2 with the article's own text
- Independent newsrooms
- 4
- This version written
- 54 h after the record first saw the story
- Second model (editor)
- kept as written
- Publication gate
- passed
- Human review
- none
WHAT THE EDITOR DID
The second model checked the writer's draft against the same sources and published it as written.
Written by a language model from the sources above, then checked by a second model that may only cut, attribute or correct. How it works →