AI Daily for 29 May: Claude Opus 4.8, Anthropic 65B Round, AI Permission Fatigue, LLM Smells

AI Daily for 29 May recaps 5 major AI Hacker News stories, moving through claude opus 4.8, anthropic 65b round, ai permission fatigue, llm smells. 1. Claude Opus 4.8 The next story is Anthropic's Claude Opus 4.8, which the company says improves coding, agentic work, judgment, and speed while keeping the same price, making it an important update for developers already building around Claude. Hacker News reacted with a mix of fatigue and curiosity, with many readers calling it a modest bump rather than a breakthrough and questioning whether today's benchmarks still capture real progress. Story link Hacker News discussion 2. Anthropic 65B Round The next story is Anthropic's new 65 billion dollar Series H round at a 965 billion dollar post-money valuation, which the company says will fund safety research, more compute, and product expansion as Claude adoption keeps climbing. Hacker News focused less on the victory lap and more on the mechanics and incentives of private capital, asking how long a company can keep stacking gigantic rounds before investors force an IPO. Story link Hacker News discussion 3. AI Permission Fatigue The next story is a Show HN project called Continue? Y slash N, a sixty-second game that tests how carefully people read AI tool commands and permission prompts, turning agent safety habits into a simple reflex challenge. Story link Hacker News discussion 4. LLM Smells The next story is an essay called Various LLM Smells, where the author catalogs recurring tells in AI-assisted writing and design, from punchy sentence rhythms to overused interface tropes, and argues that these patterns are now recognizable across the internet. Hacker News largely agreed that these stylistic fingerprints are real, though readers split on whether exposing them is useful criticism, temporary pattern literacy, or just another way the models will quickly adapt. Story link Hacker News discussion 5. AI Jobs Walkback The next story is a report that Sam Altman and Dario Amodei are dialing back their earlier warnings about an imminent white-collar jobs apocalypse, arguing instead that AI has not yet displaced workers at the scale they predicted and may expand productivity more than it destroys roles. Hacker News met that reversal with heavy skepticism, with many readers reading it as IPO-era message discipline rather than a sincere update to the evidence. Story link Hacker News discussion That’s it for today.