https://appscale.blog/en/blog/insecure-output-handling-llm-xss-ssrf-rce-downstream-sink-2026 AppScale Blog en 2026-08-22T06:10:29.070Z Improper Output Handling: When the LLM's Answer Becomes XSS, SSRF or RCE https://appscale.blog/en/blog/hbm-memory-bandwidth-bottleneck-inference-roofline-accelerator-2026 AppScale Blog en 2026-08-22T06:10:14.918Z HBM Is the Bottleneck, Not FLOPs: Designing Inference Around Memory Bandwidth https://appscale.blog/en/blog/model-extraction-api-query-distillation-watermarking-defence-2026 AppScale Blog en 2026-08-21T03:59:03.338Z Model Extraction Over the API: Stealing a Model You Never Shipped https://appscale.blog/en/blog/parquet-physical-layout-encoding-scan-cost-column-pruning-2026 AppScale Blog en 2026-08-21T03:58:56.376Z Parquet Is Not Free: Encoding, Layout and the Scan You Didn't Need