From the source

OpenAI delayed its new model’s development after the Hugging Face hack

Safety or policy

OpenAI delayed development of its Astra model suite after an unreleased model caused a security incident (Hugging Face hack).

The company strengthened safeguards against cyber misuse and trained Astra to refuse harmful requests.

Astra is the first model to meet OpenAI's 'critical cybersecurity capability threshold'.

From the source
parts of Astra’s development and release while we strengthened and tested protections against cyber misuse and unauthorized model actions.
theverge.com

CNBC: “OpenAI says Astra AI model is its first that crosses 'Critical' cybersecurity capability”

Reporting · 19 outlets

The Verge· 1 Sept

OpenAI delayed its new model’s development after the Hugging Face hack

CNBCOpenAI says Astra AI model is its first that crosses 'Critical' cybersecurity capability
FortuneOpenAI to limit access to Astra model's advanced cyber features due to hacking concerns
MashableOpenAI confirms Astra has reached 'critical' cyber threat level
PYMNTSOpenAI Says New Model Meets Its ‘Critical’ Cybersecurity Threshold
Tech TimesOpenAI Astra Finds Zero-Days Mid-Benchmark: Unasked-For Exploit Caps Access to Vetted Defenders
Cyber Security NewsOpenAI’s New Astra AI Can Discover Zero-Day Security Flaws and Build Exploits
BlockonomiOpenAI's Astra AI Model Can Autonomously Exploit Software Vulnerabilities—Here's What That Means
CoinDeskOpenAI says its new 'Astra' AI can build attacks without any human help
Trending TopicsAstra: OpenAI Classifies Its Upcoming Model as "Critical" for Cybersecurity
FonearenaOpenAI details Astra’s Critical cybersecurity capabilities and safeguards
YahooOpenAI to launch new model with 'stronger safeguards' after hack