Live data from Hacker News

Wolfram Alpha and ChatGPT

writings.stephenwolfram.com

91–100 of 309 posts

Re: Wolfram Alpha and ChatGPT

#91

I'm almost offended by the "cubic light year of ice cream" answer from ChatGPT. It's obviously ridiculous but is also a fairly simply dimensional analysis problem. Do the damn math, don't wag your finger at me and crush my dreams! I'm pretty bullish on ChatGPT and its ilk, but I _really_ dislike when ChatGPT lectures me because my request is against its "moral values." I recently pasted in the lyrics from Sleep's tit…

I'm guessing they don't want to end up like MS's last (2? Can't find the other one)(1) or Meta's attempts at online chatbots/ai?(2)

  I'm guessing OpenAI just doesn't want the bad press...

1) https://www.theverge.com/2016/3/24/11297050/tay-microsoft-ch... 2) https://mashable.com/article/meta-facebook-ai-chatbot-racism...

Re: Wolfram Alpha and ChatGPT

#92
post #84

I am always slightly in awe of the clarity of Wolfram's thought and communication skills. 63 years old, rich as a king, and yet you can just feel his energy as he digs into this incredible, magical new toy while he takes notes from his treadmill desk or outdoor walking setup (1) :). The entire article feels contributory, non jealous, complimentary and objective given the position he is writing from. I feel like Wolfr…

Peter Norvig comes to mind when you mention incredibly articulate communication style and in a respectful way, ridiculously prolific. I've turned to his work a lot throughout my Python journey, it took a lot of hard work to unlearn all the Perl in my old bones, but I appreciated how much beautiful code he's put up online for anybody to dig through.

Maybe also Larry Wall, RIP.

Re: Wolfram Alpha and ChatGPT

#93

I'm almost offended by the "cubic light year of ice cream" answer from ChatGPT. It's obviously ridiculous but is also a fairly simply dimensional analysis problem. Do the damn math, don't wag your finger at me and crush my dreams! I'm pretty bullish on ChatGPT and its ilk, but I _really_ dislike when ChatGPT lectures me because my request is against its "moral values." I recently pasted in the lyrics from Sleep's tit…

It strangely refuses to even entertain the possibility that Santa Claus is a kind of clown. Try it.

Re: Wolfram Alpha and ChatGPT

#94
post #84

I am always slightly in awe of the clarity of Wolfram's thought and communication skills. 63 years old, rich as a king, and yet you can just feel his energy as he digs into this incredible, magical new toy while he takes notes from his treadmill desk or outdoor walking setup (1) :). The entire article feels contributory, non jealous, complimentary and objective given the position he is writing from. I feel like Wolfr…

Peter Norvig comes to mind when you mention incredibly articulate communication style and in a respectful way, ridiculously prolific. I've turned to his work a lot throughout my Python journey, it took a lot of hard work to unlearn all the Perl in my old bones, but I appreciated how much beautiful code he's put up online for anybody to dig through. Maybe also Larry Wall, RIP.

Appreciate your suggestions -

Re: Wolfram Alpha and ChatGPT

#96

I wanted to see if it could get here they way you would if you were a person though repeat multiplication. what is 3 to the power of 73? 3 to the power of 73 is equal to 3 raised to the 73rd power, or 3^73. This can also be written as 3 * 3 * 3 ... (73 times). It is a very large number and its value is approximately equal to 8.659e+21 if you wish to calculate it you can use the pow() function in python like this : Co…

I finally got it to work for 3x27 by pointing out it's mistake and then it still got it wrong but finally got it right then I told it to do it 1 more time and it did that ok. then I told it to try it 69 more times and it told me to f'off( in chatbot speak). told me to use a calculator or pyhon.

Re: Wolfram Alpha and ChatGPT

#97
post #50

Earlier quoted context omitted.

It's quickly apparent that the people impressed with ChatGPT's code are generally solving pretty toy problems and want an alternative to SO for "how to do X".

And that’s fine for the average HN audience; what disturbs me is that hiring managers see articles like “I had ChatGPT rewrite my code and it went 100000% faster!!!”, and then decide that they don’t need more than one engineer. If we could automate away engineers that’s fine with me, but i just don’t think that ChatGPT is there yet. I actually think Wolfram is kind of onto something with having ChatGPT work with Alph…

The comment above captures it well though-- many of us aren't just writing stuff you can copy/paste from stackoverflow. When you're working on things where you're operating with internal business concepts and data with set meanings the difficulty is rarely writing the code but rather reasoning about how to cleanly approach the problem in the first place.

The thing is, the hiring managers reacting like that to ChatGPT are also the same folks who would've considered outsourcing engineering anyways to one of those offshore "consultancies". Tools like ChatGPT will ultimately be a replacement for use cases where you've already fully scoped out what to build and how to build it and the implementation is straightforward.

Re: Wolfram Alpha and ChatGPT

#98

I'm almost offended by the "cubic light year of ice cream" answer from ChatGPT. It's obviously ridiculous but is also a fairly simply dimensional analysis problem. Do the damn math, don't wag your finger at me and crush my dreams! I'm pretty bullish on ChatGPT and its ilk, but I _really_ dislike when ChatGPT lectures me because my request is against its "moral values." I recently pasted in the lyrics from Sleep's tit…

My least favourite part of ChatGPT is when it tells me it has no biases and is not opinionated whilst telling me it's opinions and biases and why it won't do what I asked. then when corrected it actually cannot see this., so it's hard coded and in my opinion...already second rate and ready for a replacement

Which of these hypotheticals is least bad: an AI which won't write political invective against anyone, or one which will be used by your enemies to stir up hatred against your entire team, and your team's only available response is to do the same back at the entire other side?

Re: Wolfram Alpha and ChatGPT

#99

Earlier quoted context omitted.

> I think ChatGPT is pretty neat but I was somewhat less impressed than everyone else is with the code generation. specifically on code generation, even if it could produce the correct code, it's basically faster horses: if you ask a programmer what they want, it's something to generate rote code for them so they don't have to write it... but that's the wrong problem. The problem is not that code is hard to write, or…

Code generation isn't going anywhere. No/low code platforms are great for some subset of users, but there will always be people who want total control to differentiate themselves, since differentiation is how you get customers in a sea of competitors. That total control requires code of some sort. Additionally, even if 95% are on platforms, the platforms are on code, and being able to rapidly iterate your platform co…

Code is a tool for businesses, and while it's a tool that can enable you to do anything you can imagine, it is an expensive tool for businesses to do anything they can imagine... and how many businesses need to be able to do anything they can imagine?

The cost of code is huge, not only do you need competent software engineers to write the code, but first you need people who can translate business requirements into technical requirements... and then you're burdened with the costs of running, maintaining and iterating on that code as the business evolves.

If you could cut all of that out and instead have the people defining the business requirements design the software (through the use of a no-code platform) your business would be orders of magnitude more efficient. Historically, magical all-in-one configurable platforms (like Salesforce) were unwieldy and required experts to operate them, but that's no longer the case. The world is becoming much more mature, and businesses can achieve so much without owning a single line of code.

There'll always be technology that needs to be built, and code is a great tool for building technology, but most businesses don't need to build technology, they need to use technology. If you're opening your first grocery store, you rent a building, you don't hire architects and construction workers. The immaturity of technology has historically made that approach to business quite difficult, however, that's rapidly changing.

As Software Engineers, we get caught up in believing that because you can do anything with code, code is the right tool. That's a misunderstanding of business, though: given the choice between 2 tools, 1 that can do 90% of what the business needs and 1 that can do 100% of what the business needs but costs an order of magnitude more and has far greater risk then any rational business would choose the former.

If you spend much time interfacing with non-technical people in businesses, you'll often discover that they have an acrimonious relationship with technology in their business, because so much of the technology we build is caught up in the what and not the why. Code generation does not solve that problem, empowering non-technical people to own their business processes does.

Code generation is a tool for 2000, not 2030. I would bet my career that in 2030, there'll be less code produced per year (human authored + generated) than there is today.

Re: Wolfram Alpha and ChatGPT

#100
post #67

I'm almost offended by the "cubic light year of ice cream" answer from ChatGPT. It's obviously ridiculous but is also a fairly simply dimensional analysis problem. Do the damn math, don't wag your finger at me and crush my dreams! I'm pretty bullish on ChatGPT and its ilk, but I _really_ dislike when ChatGPT lectures me because my request is against its "moral values." I recently pasted in the lyrics from Sleep's tit…

Plenty of MLM's will be "anything goes". That OpenAI does not want its brand associated with the worst things its users dream up, especially this early while it's still a research preview, makes sense to me. OpenAI was bold enough to put out a live LLM for people to hammer on, something several other companies declined to do. I think they've done a good job keeping the plane flying while plugging holes. Not a perfect…

Can you give some examples of bad things people could do without OpenAI's restrictions?
Post reply on HN