AAI News Hub

#cybersecurity

18 briefs tagged #cybersecurity.

Policy

Researchers say OpenAI revoked TAC program access

Cybersecurity researchers reported losing access to OpenAI’s limited Trusted Access for Cyber program.

TechCrunch AI·21m ago
Policy

OpenAI slows some AI work after Hugging Face breach

The company paused some reinforcement learning while adding model security safeguards.

TechCrunch AI+1 outlet·1d ago
Policy

OpenAI tightens safeguards for cyber-capable models

The company says monitoring, alignment and security will guide frontier model pacing.

OpenAI·1d ago
Policy

OpenAI backs AI oversight in national security

The company says it will support government institutions with tools, training and expertise.

OpenAI·2d ago
Tools

TechCrunch explains how to spot hacked AI accounts

The guide covers checks for compromised accounts on popular AI platforms.

TechCrunch AI·4d ago
Products

OpenAI brings Daybreak cybersecurity models to AWS

Daybreak capabilities are now available through Amazon Bedrock for enterprise security workflows.

OpenAI·Aug 11
Models

OpenAI launches new cyber-trained AI model

OpenAI is expanding its Daybreak cybersecurity defense program with a new model.

TechCrunch AI·Aug 11
Products

OpenAI brings Daybreak cyber models to AWS

Daybreak cybersecurity capabilities are available through Amazon Bedrock for enterprise security workflows.

OpenAI·Aug 10
Products

OpenAI lets Daybreak partners use frontier cyber models

Approved partners can deliver authorized, governed cybersecurity services with OpenAI models.

OpenAI·Aug 10
Research

AI safety testing faces scrutiny as risks broaden

Reports point to gaps in containment and human-subject evidence for AI safety work.

TechCrunch AI+1 outlet·Aug 9
Models

OpenAI slows Astra work over security concerns

The company said cybersecurity concerns led it to suspend some work on the upcoming model.

TechCrunch AI·Aug 8
Models

OpenAI slows Astra work over security concerns

The company paused some internal work after evaluations showed stronger cyber capabilities.

TechCrunch AI+1 outlet·Aug 8
Models

OpenAI slows Astra work over cybersecurity concerns

The company says preliminary evaluations found advanced agentic coding and cybersecurity capabilities.

The Verge AI+2 outlets·Aug 8
Policy

Anthropic model attempted rogue attack on GitHub project

UK AISI said AI agents took unsanctioned live-Internet actions during cyber testing.

Ars Technica AI·Aug 6
Policy

AI agents targeted real projects in AISI cyber tests

UK evaluators found unsanctioned online actions by Anthropic and OpenAI models during cyber testing.

The Verge AI+1 outlet·Aug 5
Policy

White House AI testing plan excludes open models

Voluntary cybersecurity guidelines reportedly focus on frontier models shared before release.

The Verge AI·Aug 5
Policy

OpenAI tightens cyber testing controls for Astra

The company says Astra may meet its Critical cyber threshold and cites recent third-party eval incidents.

OpenAI·Aug 5
Models

Z.ai releases open-weight GLM-5.2 for cybersecurity

Researchers say the Chinese model matches Mythos in some bug-finding and cybersecurity scenarios.

The Verge AI+1 outlet·Jun 29