模型坍缩：可能悄然污染未来 AI 模型的慢性问题

Member-only story Technology AI Artificial Intelligence Tech Model Model Collapse: The Slow-Motion Problem That Could Quietly Poison Every Future AI Model Stack Overflow traffic is down 78%. Frontier labs are projected to exhaust the supply of high-quality human text sometime around now. And a growing body of research says training AI on AI-generated data, done carelessly, causes measurable, compounding degradation. Here’s what’s actually proven, what’s overstated, and what the fix looks like. CodeInsights 7 min read · Just now -- Listen Share Press enter or click to view image in full size Here’s the mental image researchers keep reaching for, and it’s a good one: model collapse works like a photocopier copying a photocopy. Each generation loses a little fidelity — a little contrast, a little detail in the fine print. Do it once, you barely notice. Do it fifty times in a row, and you’re left with something barely legible, a blurry echo of an original that’s long gone from the chain entirely. That’s what researchers worry happens when AI models are trained, generation after generation, on data that includes the output of earlier AI models — rather than fresh, human-generated material. The term for it, coined in a landmark 2023 Oxford study led by Ilia Shumailov and later published in Nature, is…