Translating Akkadian clay tablets with ChatGPT?

Their apparent superficial improvement may make for an even worse "translation". Instead of being obviously incorrect, they might beget some misplaced trust. With the GPT3 version, at least the single slice of hallucination was obviously useless. With the GPT4 result, some parts seem to be more correct. Mainly, it's not hallucinating locations and seems to actually have _some_ proficiency with the genitive case. Though I would say the GPT4 translation is an improvement over the one from GPT3, it has improved from something with no basis in reality to one that is more wrong than right.

From col. V line 55 and later, GPT4 gets the bottom of the barrel usage of the genitive mostly correct. Before that, it can't grasp the concept of an adjective. So I guess that's an improvement from "lacking a basis in reality" to "still completely wrong but inspiring confidence in the untrained eye". Not great. Although it attempts to transcribe some of the place names, meaning it is no longer hallucinating them, there seems to have been no effort made at translation. This is again (and again and again it seems) something that is both extremely basic and essential; (known) place names are not hard. On the topic of transcriptions, the names of kings are poorly transcribed. It clearly doesn't know what to do with alephs in both place and personal names. Another less than stellar effort. Everything before line 54 in the GPT4 "translation" is a blood bath. Picking it apart is sisyphean.

Taking a look at a different attempted GPT4 "translation" [1], this time of col. I, I don't have anything to add that I haven't already said. It is useless incoherent garbage that the popular imagination might mistake for a real translation. The best way to correct the result would be to re-translate it yourself. It smashes a bunch of words together to form an output that so thoroughly differs in meaning from the original that it has graduated from "wrong" to "fiction a layperson might believe".

The Claude attempt [2] is more of the same. An awful "translation". Incorrect information. General incoherency on the ancient near east. Extra points off for using unicode cuneiform rather just transliteration. Asking about something in "Babylonian cuneiform" is a meaningless way to start; what time period are we talking about here? They can be pretty different... The compliance and lack of clarification from the LLM is pathetic. And the topic at hand is NA cuneiform, not OB, MB, or NB (or LB). When you program in C#, do you say you're a C programmer?

Esarhaddon's prisms are not crazy exotic, grammatically complex, unknown, and/or untranslated texts. They are the assyriological equivalent of an uncrustable. The grammar is simple and the writing system (NA cuneiform) is refined. There nothing difficult here.

The incompetency and general idiocy displayed by the now three different (though perhaps incestuously related) LLMs in this thread is shocking. Who has the time and will to stem this indomitable wall faeces being spewed towards the ancient near east by these chatbots? Going through each response, line by line to demonstrate that yes, this translation is indeed wrong, is taxing. There is no joy here. And that's just the translations. I don't know where Claude [2] got its ideas about ancient near eastern writing/phonology/grammar/culture. Maybe it makes sense if you glaze over while skimming through a poorly written wikipedia article (which encompasses most on the ANE). Nearly every single line of response has something that ranges from subtly to flagrantly wrong. Claude (and GPT friends) are making bricks for your foundational understanding of the ANE. It just so happens that those "bricks" are frozen blocks of ground beef and you're building a foundation at Nineveh. The bricks will rot and fester in the pitiless sun; as they melt the surrounding baked clay, maggots will weave through your walls and your house will become an abomination left better un-built.

These LLMs have no place in Assyriology or for those working with aspects of the ANE. In academia, there is no room for "improved" "translations" that are the product of merged queries "strengthened" by the reprocessing of garbage [1]. An incorrect translation is still incorrect, no matter how much its readability has been improved through the chewing of regurgitated cud. "Partially correct" or "mostly correct" translations are worse than irrelevant, they waste the time and expertise of all they touch.

Many of the academically relevant translations done today are not of simple, easily read, and well preserved texts like Esarhaddon's prism seen in this thread. Simple texts with available transliterations/transcriptions that also somehow lack translations can be trivially read and translated on the fly off the page with experience, no LLM needed.

With complicated translations, there is really no room for error. An LLM translated passage that differs in meaning from the original text due to the LLM's fundamental inability to comprehend basic grammar is not something to celebrate. The correct interpretation of the text is the point. Until LLMs find/are given a non-hallucinated "understanding" of the cultures and writing systems in the ANE, any produced translation will be dangerous to the amateur and worthless to the academic.

If you fall into the category of amateur, rather than running the CDLI's transliterations through a chatbot, just find a real translation of the text from an academic source. Doing this will expose you to relevant context in the form of commentary, texts, and citations. References to literature that exists in our shared reality rather than those generated by the throes of a hallucinating LLM can be helpful for research.

The GPT3 translations of NA are a joke. Those done by GPT4 are similarly worthless while containing fragments of connection to the text. Although I would consider the GPT4 "translations" to be "better", I would also think them to be more dangerous to the reader than GPT3's due to their thin gilding of competence. Anthropic/Claude have a long way to go. I suggest its makers start by wiping all training data pertaining to the ANE and begin training again with fresh with data from competent authors.

[0] https://news.ycombinator.com/item?id=35955688

[1] https://news.ycombinator.com/item?id=35957726

[2] https://news.ycombinator.com/item?id=35960213

Museums, like libraries, are organizers of information. Think of them somewhat as databases. Of course some are better than others at this.

I once had cause to track down a beautiful statue I’d seen in the MoMa some years earlier because I wanted to find a reproduction. It was gone. I found a local docent and asked, describing it (bronze color, a few feet high, sweeping curves that led to points and formed a vaguely human shape)

They weren’t sure but went to inquire. I waited maybe half an hour before they came back and said it was titled “Unique Forms of Continuity in Space”, but it was not currently on display. I asked if it was somewhere in storage that would allow a quick viewing (sometimes possible) but unfortunately it wasn’t.

But being given the title, it enabled me to buy a small reproduction, which sits on my desk :)