I love to show that kind of shit to AI boosters. (In case you’re wondering, the numbers were chosen randomly and the answer is incorrect).
They go waaa waaa its not a calculator, and then I can point out that it got the leading 6 digits and the last digit correct, which is a lot better than it did on the “softer” parts of the test.
Why would you think the machine that’s designed to make weighted guesses at what the next token should be would be arithmetically sound?
That’s not how any of this works (but you already knew that)
Idk personally i kind of expect the ai makers to have at least had the sense to allow their bots to process math with a calculator and not guesswork. That seems like, an absurdly low bar both for testing the thing as a user as well as a feature to think of.
Didn’t one model refer scientific questions to wolfram alpha? How do they smartly decide to do this and not give them basic math processing?
You are in for all kinds of surprises if you think the people shoving AI into everything are doing anything logical. They want everything to fit their singular model.
Maybe some of the smaller ones that have to run locally might, because they care about resource usage.
You forget a few major differences between us and AI makers.
We know that these chatbots are low-quality stochastic parrots capable only of producing signal shaped noise. The AI makers believe their chatbots are omniscient godlike beings capable of solving all of humanity’s problems with enough resources.
The AI makers believe that imitating intelligence via guessing the next word is equivalent to being genuinely intelligent in a particular field. We know that a stochastic parrot is not intelligent, and is incapable of intelligence.
AI makers believe creativity is achieved through stealing terabytes upon terabytes of other people’s work and lazily mashing it together. We know creativity is based not in lazily mashing things together, but in taking existing work and using our uniquely human abilities to transform them into completely new works.
We recognise the field of Artificial Intelligence as a pseudoscience. The AI makers are full believers in that pseudoscience.
Kagis Multi-Step assistant does that:
I can imagine the reason is that Wolfram alpha does calculations but also much more, so they can throw a whole bunch of shit at it.
Mistral.ai easily gets it right too
under no circumstances is “my favorite stochastic parrot got it right / I reran the prompt and it worked this time hmmmm wonder why” an interesting post
I would not expect that.
Calculators haven’t been replaced, and the product managers of these services understand that their target market isn’t attempting to use them for things for which they were not intended.
brb, have to ride my lawnmower to work
Try asking my question to Google gemini a bunch of times, sometimes it gets it right, sometimes it doesn’t. Seems to be about 50/50 but I quickly ran out of free access.
And google is planning to replace their search (which includes a working calculator) with this stuff. So it is absolutely the case that there’s a plan to replace one of the world’s most popular calculators, if not the most popular, with it.
Also, a lawnmower is unlikely to say: “Sure, I am happy to take you to work” and “I am satisfied with my performance” afterwards. That’s why I sometimes find these bots’ pretentious demeanor worse than their functional shortcomings.
“Pretentious” is a trait expressed by something that’s thinking. These are the most likely words that best fit the context. Your emotional engagement with this technology is weird
let me just slip the shades on real quick
checks out
oh so we’ve reached the gaslighting phase of the product life have we
Pretentious is a fine description of the writing style. Which actual humans fine tune.
Given that the LLMs typically have a system prompt that specifies a particular tone for the output, I think pretentious is an absolutely valid and accurate word to use.
Also, these bots have been deliberately fine-tuned in a way that is supposed to sound human. Sometimes, as a consequence, I find it difficult to describe their answering style without employing vocabulary used to describe human behavior. Also, I strongly suspect that this deliberate “human-like” style is a key reason for the current AI hype. It is why many people appear to excuse the bots’ huge shortcomings. It is funny to be accused of being “emotional” when pointing out these patterns as problematic.
The funny thing is, even though I wouldn’t expect it to be, it is still a lot more arithmetically sound than what ever is it that is going on with it claiming to use a code interpreter and a calculator to double check the result.
It is OK (7 out of 12 correct digits) at being a calculator and it is awesome at being a lying sack of shit.
Random tokens can’t lie to you, because they’re strings of text. Interpreting this as a lie is an interesting response
lol the corollary of this is that LLMs are incapable of producing meaningful output, you insufferable turd
Im using it literally every single day to make huge gains. Every single day I disprove this comment
this you?
AI boosters are people who literally cannot tell good from bad, yet again
Visions of the promptfan walking into those kind of premium mediocre restaurants came to mind here
I knew you were a lying promptfondler the instant you came into the thread, but I didn’t expect you to start acting like a gymbro trying to justify their black market steroid habit. new type of AI booster unlocked!
now fuck off
cool story, bro
That’s why I say “sack of shit” and not say “bastard”.