The cemetery of Minab, photographed as it prepares to bury more than 100 young girls, is a defining image of the US-Israeli war on Iran. But is it real?

位于米纳布的这处墓地，在准备安葬一百多名年轻女孩之际被拍摄下来，成为美以对伊战争的标志性影像之一。但它是真的吗？

Ask Google's Gemini, and the answer you receive is no — it claims the photo is of a 2023 mass burial in Turkey after an earthquake. Others turned to X's AI assistant Grok, which said it was a stock photo of Covid burials in Indonesia from 2021. In both cases, the AI answers sound sure but are simply wrong.

如果去问谷歌的 Gemini，你得到的答案是否定的——它声称这张照片实际上是 2023 年土耳其地震后的一次大规模埋葬。另一些人则转向 X 的人工智能助手 Grok，它则表示这是一张来自 2021 年印尼新冠疫情葬礼的图库照片。在这两种情况下，AI 的回答听起来都十分笃定，但却完全错误。

The cemetery image is authentic. Researchers have cross-referenced it with satellite images that confirm its location, and it can be cross-referenced again with video footage.

这张墓地照片是真实的。研究人员已将其与卫星图像进行交叉比对，确认了其位置，它的真实性还可以通过视频画面进一步核实。

Shayan Sardarizadeh of BBC Verify says AI now makes up nearly half of all the viral falsehoods his team debunks.

BBC Verify 的沙扬·萨达里扎德表示，如今他所在团队辟谣的所有热门虚假信息中，近一半是由 AI 生成的。

That has partly been driven by the ease with which anyone can now generate a realistic video or photo. But the other enormous shift is in people using AI to summarise the news or answer questions. Often, however, AI summaries are simply wrong.

这在一定程度上是因为如今任何人都可以轻易生成逼真的视频或照片。但另一个巨大的变化在于，人们开始使用 AI 来总结新闻或回答问题。然而，AI 的总结常常是错误的。

Part of the problem is how LLM AI models work. At a very basic level, they are probabilistic language models, constructing sentences piece by piece based on which next word has the highest likelihood of being appropriate, which doesn't mean the AI has actually analysed the material in front of it.

问题的一部分在于大语言模型（LLM）的运作方式。从最基本的层面看，它们是基于概率的语言模型，会从概率上判断接下来哪个词最恰当，从而一步步构建句子，这并不意味着 AI 真正分析过它所处理的材料。

For those investigating human rights abuses, the trend poses new challenges. Chris Osieck, an independent open-source investigator, said researchers' time was being wasted debunking AI material.

对于那些调查侵犯人权行为的人来说，这一趋势带来了新的挑战。独立的开源调查员克里斯·奥西克表示，研究人员的时间正浪费在对 AI 生成材料的辟谣上。

Debunking AI videos, for example, often involves carefully inspecting them frame by frame for visual discrepancies. "That time should be devoted to what matters most: reporting on the impact this brutal war has on the people caught in the crossfire."

例如，对 AI 视频进行辟谣通常需要逐帧仔细检查其中的视觉不一致之处。“这些时间本应投入到更重要的事情上：报道这场残酷的战争，对被卷入战火之中的人们所造成的影响。”