5 ms·
I just had a huge and surreal argument with Google's Gemini about the current time in other time zones. It was Sunday, 10/13. I asked what time it was in Londo
by AStonesThrow 2y ago
I just had a huge and surreal argument with Google's Gemini about the current time in other time zones.
It was Sunday, 10/13. I asked what time it was in London:
It is currently 1:39 PM MST. As GMT timezone is 7 hours behind MST, the current time in London is 6:39 AM GMT on Sunday, October 13, 2024.
Well, no, it's not.
So I started grilling it on simple math ("What is 13:39 plus seven hours") and the real offsets of the time zones, and it would respond correctly, and then I'd ask for the current time again, it'd apologize profusely, and give me some B.S. calculations right on the same line!
I started inquiries about other places, like NZ and across the International Date Line, and it was spectacularly wrong. Some places like PDT it was OK. And it was repeatedly getting tripped up and apologizing and immediately supplying the wrong info again.
I was not trying to trick it; I was not going into ambiguous situations or half-hour zones. My locale doesn't observe DST, but it was clearly adapting for MST vs. DST zones.
The other comical thing was that Gemini would sometimes refuse outright to give the current time, referring me to other methods, as if there were some security blocks on that query. But only sometimes!
I simply must conclude that time zones are one of the most difficult programming problems. And I don't know how LLMs do time/date calculations. But Gemini clearly has a long way to go from these simple math confabulations!
- mtsolitary 2y agoLLMs don’t do time/date calculations at all. They’re just good enough at predicting the next word to trick us into thinking they can.
- jasomill 2y agoWolframAlpha is my go-to for one-off calculations involving either durations and units. For your example, the WolframAlpha query "time in London at 1:39 pm MST on October 13" returns the correct answer, "9:39:00 pm BST | Sunday, October 13, 2024"[1]. While WolframAlpha's natural language processing is hit-or-miss, its responses clearly state assumptions made in the face of ambiguity. In this example: Assuming "MST" is a named time zone Assuming "time" is referring to a calendar computation Assuming month/day Assuming Mountain Time (United States; no observed DST rule) Assuming London (United Kingdom) [1] https://www.wolframalpha.com/input?i=time+in+London+at+1%3A39+pm+MST+on+October+13 https://www.wolframalpha.com/input?i=time+in+London+at+1%3A3...
- AStonesThrow 2y agoThank you kindly! It seems that WolframAlpha is venerable and stable. I first heard of it long, long ago. (Was it only released in 2009? Seems older.) It must have amazing staying power. It seemed like a quite powerful tool. Why is it such a well-kept secret?