#AIEngineering #aihype #LLMLimitations

https://bigdataboutique.com/blog/thinking-fast-and-failing-slow-why-llm-as-a-judge-fails-aa1d67

#AIEngineering #llms #aisecurity #LLMLimitations

https://pub.towardsai.net/rag-doesnt-neutralize-prompt-injection-it-multiplies-it-329c91b2e5a3

#AIEngineering #LLMLimitations #airisks

https://arstechnica.com/ai/2025/12/syntax-hacking-researchers-discover-sentence-structure-can-bypass-ai-safety-rules/

#AIEngineering #LLMLimitations #airisks

https://news.mit.edu/2025/shortcoming-makes-llms-less-reliable-1126

#AIEngineering #aiagents #LLMLimitations

https://thenewstack.io/why-load-tests-lie-harsh-truth-about-ai-agent-performance/

#AIEngineering #LLMLimitations #airisks

https://www.theregister.com/2025/11/21/poetry_llm_guardrails/

#AIEngineering #airisks #LLMLimitations

https://www.theregister.com/2025/11/14/ai_guardrails_prompt_injections_echogram_tokens/

#AIEngineering #aisecurity #LLMLimitations

Only 250 poisoned documents can compromise models across all model and dataset sizes.

Poisoning Attacks on LLMs Require a Near-constant Number of Poison Samples

https://arxiv.org/abs/2510.07192

#AIEngineering #GenAI #LLMLimitations

https://restofworld.org/2025/south-korea-ai-textbook/

#AIEngineering #llm #airisks #LLMLimitations

Deloitte refunds Aussie gov after AI fabrications slip into $440K welfare report

https://www.theregister.com/2025/10/06/deloitte_ai_report_australia/

#AIEngineering #llm #LLMLimitations

https://arstechnica.com/ai/2025/09/science-journalists-find-chatgpt-is-bad-at-summarizing-scientific-papers/

#AIEngineering #LLMLimitations #aisecurity

https://www.theregister.com/2025/08/26/breaking_llms_for_fun/

#AIEngineering #llm #airisks #LLMLimitations

https://medium.com/data-science-in-your-pocket/grok-4-failed-these-benchmarks-elon-lied-again-412a78fcabf9

https://www.infoq.com/news/2025/07/apple-illusion-thinking/

#AIEngineering #llm #LLMLimitations