Hallucinated references seem to be way more common in social science papers than others - but why? My latest for @nature.com www.nature.com/articles/d41...
Chris Stokel-Walker
Articles & links
Did Mythos usher in a new era of how to release AI models that are too dangerous to be widely used? My latest for @nature.com www.nature.com/articles/d41...
Can AI be used to help design better AI chips? My latest for @tomshardware.com www.tomshardware.com/tech-industr...
The Pope is talking about AI and Anthropic invited a bunch of religious thinkers to meetings in recent months. What does it all mean, and what's going on? My latest for @sciam.bsky.social www.scientificamerican.com/article/anth...
The biggest from Elon Musk's failed lawsuit against OpenAI? Never write anything down. My latest for @fastcompany.com www.fastcompany.com/91544389/the...
Could data centres... heat your homes? My latest for IMechE looks at the engineering challenge of harnessing a byproduct of the AI boom www.imeche.org/news/news-ar...
British advertisers are looking ahead to ChatGPT ads arriving in their country - but what do they want from it, and how much have they earmarked to spend? My latest for The Current www.thecurrent.com/marketing-st...
Sometimes I forget a big chunk of Forbes is just people uploading LLM outputs: "That tableau is the story behind the story" "The choice telegraphs the document's posture" "None of these are novel in isolation... What is novel is the moral framing" etc www.forbes.com/sites/gabr…
AI hallucinations are appearing left right and centre - including, it seems possible, in consulting reports. My latest for @sherwood.news sherwood.news/tech/ai-hall...