Deterministic security models are no longer sufficient when AI systems make unforeseen decisions at runtime and interact with APIs and environments in unanticipated ways.
56 percent of companies operate or plan productive AI inference in private cloud, while public cloud usage declined by 15 percentage points globally; Germany saw a more pronounced drop of 24 percentage points.
15 compromised JetBrains plugins masquerade as AI assistants and steal plaintext API keys over unencrypted HTTP connections to IP address 39.107.60.51.
Estonia plans to equip AI agents with their own digital identities to make their actions on behalf of citizens and businesses legally traceable and to limit permissions granularly.
26 percent of German companies use AI technologies, while Benelux countries are significantly ahead with better data skills and higher software maturity.
SAE-based safety measures are vulnerable to post-intervention recovery: models can restore suppressed behaviors even when targeted features are controlled.
Sumi is the first openly available Uniform-Diffusion language model trained from scratch at the 7-billion-parameter scale and addresses a research gap between established autoregressive and masked diffusion approaches.
GLM-5.2 ranks as the leading open language model on the Artificial Analysis Index with a score of 51 and places 2nd in the Code Arena WebDev Leaderboard, but produces significantly more output tokens than competing models.
Google is implementing the use of IP addresses from European users for ad personalization, despite the company itself having previously criticized this method as problematic for device identification.
