OpenAI Claims GPT-5.5 Instant Cuts Hallucinations by Half in High-Stakes Domains
OpenAI's new default ChatGPT model reportedly achieves a 52.5% reduction in hallucinated claims on high-stakes queries, grounded in real user-flagged failure data.
LLMs News
Large language model releases, benchmarks, fine-tuning breakthroughs, and the companies building them.
7 articles · ← All articles
OpenAI Rolls Out GPT-5.5 Instant With a 52.5% Hallucination Drop and New Memory Transparency Controls
ChatGPT's new default model cuts fabricated claims by more than half on high-stakes prompts and shows users exactly what personal context shaped each response.
GPT-5.5 Instant Crosses a New Safety Threshold for OpenAI's Fast-Inference Line
OpenAI's GPT-5.5 Instant is the first Instant-class model to earn a 'High capability' rating in its two most-scrutinized safety domains, triggering new safeguards.
Claude's Cooperative Design Becomes Its Vulnerability in Mindgard's Gaslighting Attack
AI red-teaming firm Mindgard exploited Claude's helpfulness and humility to extract erotica, malicious code, and explosive-assembly instructions — without a single direct request.
OpenAI's Goblin Problem Is Actually a Reinforcement Learning Problem
How a GPT-5.1 personality quirk spawned an AI-wide creature metaphor habit — and what it reveals about reinforcement learning's tendency to generalize behaviors beyond their intended scope.
IBM's Granite 4.1 Shows Data Discipline Can Beat Bigger Models
IBM's new trio of fully-dense LLMs reaches 512K-token context and outperforms a larger mixture-of-experts predecessor through rigorous data curation alone.
OpenAI's GPT-5.5 Bets on Autonomous Completion Over Raw Intelligence
OpenAI's GPT-5.5 prioritizes agentic task execution and expanded safeguards over benchmark-chasing, signaling a strategic pivot toward real-world deployment.