OpenAI says it paused AI training for two weeks and announces new security protocols following Hugging Face hack
Headline as filed by Fortune. PageNews does not write or rewrite headlines.
PageNews Brief
2 publishers have filed 2 reports that PageNews has grouped as one story.
The earliest report detected came from Fortune at 18:33 UTC on 18 August 2026, and the most recent from Channel News Asia 30 minutes later.
Hugging Face is named in all 2 reports.
The coverage is filed under Technology. Each publisher's own headline, summary and link is below.
Order of coverage
- first Fortune — earliest coverage indexed by PageNews
- +30m Channel News Asia — latest publisher to join
Names shared across publishers
Details appearing in more than one publisher
Wording appearing in 2 or more publishers’ own summaries, with the number that used each. It records that they used the same words, not that anything is confirmed.
Compiled by PageNews from the reports listed below — counts, timestamps and wording taken from them. No publisher wrote this section.
Coverage from 2 publishers
-
Fortune
OpenAI says it paused AI training for two weeks and announces new security protocols following Hugging Face hack
The AI company says its unreleased 'Astra' model presents a 'critical' cybersecurity risk and it has put its largest training runs on hold while it tests new safety procedures.
Read on Fortune → -
Channel News Asia
OpenAI slows model training to bolster security after Hugging Face hack
Read on Channel News Asia →
How the coverage developed
- first Fortune coverage detected OpenAI says it paused AI training for two weeks and announces new security protocols following Hugging Face hack
- +30m Channel News Asia coverage detected OpenAI slows model training to bolster security after Hugging Face hack