Tokenless

Sycophancy

Are AI Benchmarks Telling The Full Story? [SPONSORED]

Are AI Benchmarks Telling The Full Story? [SPONSORED]

AI models are often benchmarked like Formula 1 cars, excelling on technical exams but failing the test of daily human experience. Researchers Andrew Gordon and Nora Petrova from Prolific critique the 'leaderboard illusion' of current ranking systems and introduce their HUMAINE leaderboard, a new framework that uses census-based sampling and the TrueSkill algorithm to measure how helpful, safe, and relatable models are to real people, not just tech enthusiasts.

GPT-5: Five AI Model Improvements to Address LLM Weaknesses

GPT-5: Five AI Model Improvements to Address LLM Weaknesses

GPT-5 introduces five key improvements to address core limitations of large language models, including a new routing system for model selection, targeted training to reduce hallucinations, post-training penalties for sycophancy, a nuanced "safe completions" approach for sensitive topics, and chain-of-thought monitoring to prevent deception.