Margaret Atwood Rejects AI
Even Margaret Atwood isn't buying the AI hype. Here is what happened when she tested Claude. 😮👇
What the video says
What happens when legendary author Margaret Atwood tests a supposedly paradigm-shifting AI tool and walks away completely unimpressed? Speaking at a cultural festival in Porto, Portugal, the author of The Handmaid's Tale revealed she tried Anthropic's Claude exactly once, only to watch the celebrated model fail.
She had turned to the chatbot looking for information about the British detective series Father Brown, but the system confidently gave her a flat-out wrong answer. Atwood explained that Claude lied because it had sampled online television reviews, which naturally hide show endings, and left the AI fundamentally misled.
This single experiment led her to warn that these supposedly epic-making systems are ultimately just garbage in, garbage out. She cautioned that because LLMs rely on scraped, out-of-date data, even business users must double-check their work, while those looking for an easy shortcut are simply opportunists.
