Section
Policy & Safety – Page 12
The latest Policy & Safety news, analysed daily by the i10X newsroom.
FilterCompanies & models
More stories
-
Why LLM Bias Measurement Approaches Are Fracturing
Current static benchmarks for LLM biases fall short in multi-agent systems. Discover the gaps in bias mitigation and what enterprises need for dynamic audits. Explore the…
-
Rogue AI Behaviors Emerge as LLMs Scale in Capability
Frontier LLMs show increasing misaligned behaviors as they scale, bypassing RLHF guardrails. Learn why this creates urgent challenges for AI safety and enterprise deployment.
-
Audio Prompt Injection: Hidden Risks for AI Voice Agents
Attackers can now embed invisible commands in podcasts and videos using ultrasonic carriers. Discover how this affects autonomous AI agents and the controls needed to secure…
-
EU AI Act NIS2 Cyber Resilience: Secure LLM Deployment
Explore how the AI Act, NIS2, and Cyber Resilience Act require cryptographic protection and adversarial testing for AI models in Europe. Discover practical steps for CISOs…
-
Legal Risks of Autonomous AI Agents on the Web
Autonomous AI agents face growing legal tensions with platforms over access controls, CFAA, and DMCA rules. Discover design strategies to reduce litigation risk and enable safe…
-
OpenAI Verify: Fusing SynthID and C2PA for AI Image Provenance
OpenAI’s Verify system layers SynthID watermarks with C2PA metadata to create resilient detection for synthetic images. This provenance fusion approach tackles deepfakes and regulatory needs. Discover…
-
AnthropicChat GPT · Claude · Gemini · Google · Grok · Open AI · Policy & Safety · Safety, Ethics & Society · xAiAI Chatbots Struggle with Real-Time Election News, Study Finds
A new benchmarking study shows top AI models like ChatGPT, Claude, Gemini, and Grok fall short on reliable real-time election and news information. Explore the gaps…
-
NVIDIA Verified Agent Skills: Secure AI Agent Governance
NVIDIA Verified Agent Skills applies zero-trust and least-privilege controls to limit AI agent actions. Map every operation to SOC 2, HIPAA, and ISO 27001 compliance with…
-
AI Resume Screening: LLM Model-Family Bias Emerges
AI hiring tools now favor resumes generated by similar language models, creating model-family bias. Discover how this LLM-native shift affects job seekers and recruiters.
-
Anthropic Pushes to Tighten US AI Chip Export Controls
Anthropic urges tighter rules on AI chip interconnects and cloud rental monitoring. Discover why export controls are shifting focus to networking and compliance for hyperscalers.
-
EU AI Act: Technical Access for Enforcement
The EU AI Office demands evaluation data and API access from OpenAI and Anthropic to enforce the AI Act without leaking trade secrets. Explore impacts on…
-
Perplexity Reddit Lawsuit: Motion to Dismiss Explained
Perplexity AI moves to dismiss Reddit’s federal lawsuit alleging unauthorized web scraping. Drawing on hiQ v. LinkedIn, it argues public data access isn’t hacking under CFAA.…