#model-governance

[ follow ]
#enterprise-ai #ai-cybersecurity #red-teaming #dual-use-risk #ai-roi #ai-infrastructure #open-source-risk
Artificial intelligence
fromIT Pro
1 day ago

OpenAI turns to red teamers to prevent malicious ChatGPT use as company warns future models could pose 'high' security risk

OpenAI will restrict access and harden future models—using adversarial training, red teaming, and trusted partners—to prevent AI-enabled cyberattacks and mitigate dual-use risks.
#enterprise-ai
more#enterprise-ai
Artificial intelligence
fromTechzine Global
2 weeks ago

Trend Micro launches AI Security Package

Trend Vision One AI Security Package provides proactive exposure management and analytics to protect AI application stacks from model-specific threats across development and runtime.
fromSuff Syed
2 months ago

Designers Have to Move from the Surface to the Substrate - Suff Syed

For most of computing history, designers have been the authors of the visible world. They shaped the screens we tap, the icons we recognize, the patterns we repeat until they feel second nature. From Xerox PARC to Apple's Human Interface Guidelines, nearly every interface worth designing has already been designed. The metaphors of interaction: the desktop, folders, windows, menu, tap, swipe, are now complete, endlessly iterated but rarely reinvented.
Design
Artificial intelligence
fromZDNET
2 months ago

Forensic vibers wanted - and 10 other new job roles AI could create

New AI-driven job roles will emerge, requiring human oversight, prompt engineering, model management, and ethical and legal supervision.
Artificial intelligence
fromTheregister
3 months ago

All IT work to involve AI by 2030, says Gartner

By 2030 all IT department work will be performed with AI assistance, shifting roles toward AI-augmented humans, reducing entry-level positions but avoiding massive job losses.
Artificial intelligence
fromTechzine Global
3 months ago

Cloudsmith ML model registry lays down the law

Cloudsmith released an ML Model Registry to apply enterprise-grade controls, traceability and security to ML models and datasets, reducing model sprawl and compliance risks.
[ Load more ]