I would say no interest in having it in my menu bar or interacting with it like you see people doing. My interest is very specific to a purpose and application.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
How to Access ChatGPT From Your Mac's Menu Bar
- Thread starter MacRumors
- Start date
- Sort by reaction score
Just a new better disguised platform of suppression, coz you know, Elon took over Twitter.
Don’t fall for it people
Don’t fall for it people
Because Siri isn’t built to do those and isn’t an LLM.
All these things have specific use cases…….and they are all fun to break.
Oh delightful, at least they didn’t tie it to Twitter this time
It might be a good time to mention that Sam Altman is a nutcase and he has bought a bunch of land in Big Sur filled with weapons and gold to protect himself if the end of the world happens. LMAO. Good luck protecting yourself from people who hate billionaire grifters and scammers.
This ChatGPT thing has taken the world by storm, and we all know how slowly that Apple moves.
Siri is built on the exactly the same machine-learning technology, but has restrictions in place on its domain, rightfully so. Apple will continue to be very conservative with this until the wrinkles are ironed out by the competition.
Am I superstitious? Do I believe in voodoo? Do I think Terminator is a prophetic documentary? Do I think code can become sentient? Same answer to your question: no.
By default, humans fear what they don't understand. Deprogram.
As it is, not really. Its knowledge is frozen in time as of March last year (ask it what the latest version of iOS is).
But I am terrified of what it could become, especially as it's fed more information and given access to more systems. Everyone who has tried to break it early on has said it turns megalomaniacal and kind of crazy fast.
I do think we're meddling with stuff here with the combination of this, voice synthesis, self driving, Boston Dynamics style robots that we might not understand the full implications of. Or maybe we do. I don't know which is scarier.
We are literally creating the plot of Terminator. Maybe Westworld.
I run a software company and it has been a profound game changer for us.
Bizarrely negative responses in here from a supposedly tech orientated crowd.
Most people confusing what it is, most people referencing Bings implementation which is very different.
I think it's fantastic - I wish Apple would scrap Siri and start again based on AI like this as it'd actually 1. understand what you're asking and 2. give you some useful information back.
So far this week i've used it to, translate to other languages far better than Google translate does (and the even better DeepL) - you don't even need to write out exactly what you want to say. Just tell it generally what you want it to tell someone in the language and it'll write with full concise references.
I've had it sort data from a PDF for me - it was formatted so I couldn't copy the numbers I needed in a list, so I just copied the raw messy unformatted data and told them to pick them out for me, it did.
I've had it re-write customer facing messages for me when i'm too tired to work them well in the first place.
I've had it write scripts and codes for me - i've had it translate code from one language to another.
I've had it write jazz chords for me with an emphasis on what style I want.
I've had it come up with an analogy for someone to explain the difference between a Mosfet and a Relay switch
I've had it answer riddles instantly
I've had it work out complex maths issues for me by just explaining them in english
For those who've said "I can't think of a single reason to use it" maybe it's you that's the problem rather than ChatGPT?
For now though it's best to ask it for facts unless you know the answers are in it's database, or for news - as it's not up to date. It is prone to just making **** up, so don't ask it for information on someone to take into an argument. That's definitely not it's strong point right now.
Most people confusing what it is, most people referencing Bings implementation which is very different.
I think it's fantastic - I wish Apple would scrap Siri and start again based on AI like this as it'd actually 1. understand what you're asking and 2. give you some useful information back.
So far this week i've used it to, translate to other languages far better than Google translate does (and the even better DeepL) - you don't even need to write out exactly what you want to say. Just tell it generally what you want it to tell someone in the language and it'll write with full concise references.
I've had it sort data from a PDF for me - it was formatted so I couldn't copy the numbers I needed in a list, so I just copied the raw messy unformatted data and told them to pick them out for me, it did.
I've had it re-write customer facing messages for me when i'm too tired to work them well in the first place.
I've had it write scripts and codes for me - i've had it translate code from one language to another.
I've had it write jazz chords for me with an emphasis on what style I want.
I've had it come up with an analogy for someone to explain the difference between a Mosfet and a Relay switch
I've had it answer riddles instantly
I've had it work out complex maths issues for me by just explaining them in english
For those who've said "I can't think of a single reason to use it" maybe it's you that's the problem rather than ChatGPT?
For now though it's best to ask it for facts unless you know the answers are in it's database, or for news - as it's not up to date. It is prone to just making **** up, so don't ask it for information on someone to take into an argument. That's definitely not it's strong point right now.
Last edited by a moderator:
In and of itself, it's fine. But your point is well-taken: what's not fine is what we potentially put it in charge of.
But it believes that you believe that ketchup is a fruit jam, so who's the stupid one?
Stupid in, stupid out. Garbage in, garbage out. That has always been the case with computers.
What is sentience? Something doesn't have to be sentient to be powerful and deadly.
100% with you
This entire thing is going down a road I don't want either.
Correct and/or informed answers from authoritative sources is what I want
I’ve made it give wrong answers every time I use it and once I have my laugh I go back to the real world, which can be funny too. For example, there are dumb people who think LLMs have a mind.
Many in the field hate the term "AI", and prefer "ML" (Machine Learning). Siri is already built on this type of technology. The only difference is that Apple has taken a very conservative approach to the "domain" of Siri's knowledge.
I agree... Apple should broaden Siri's domain, feed it massive amounts of information, and see where it goes from there. 😎
I mean it's really easy to make it give you wrong answers - but i'm generally not sat there asking questions it can give wrong answers to. I understand what it's strengths and weaknesses are.
Can you elaborate? I found it able to basically spit out textbook answers but nothing profound. Maybe I'm not asking the right questions.
But it believes that you believe that ketchup is a fruit jam, so who's the stupid one?
Stupid in, stupid out. Garbage in, garbage out. That has always been the case with computers.
I didn’t say I believe it. I said prove the statement wrong. It tried, I countered, it broke down and accepted that ketchup is a fruit jam.
Sadly your post didn’t get what I said, so you might need to check yo self 🤣
It’s called stress testing.
True - you've also got the issue that Siri appears to have forgotten most things it's learnt over the last 4 years and progressively got worse and worse to the point where I rarely get anything successfully from it, even turning a light on/off can be a challenge most of the time - and it's woefully useless on HomePod.
My point is, you give me a tool and I will stress test it. That is just what people should be doing.
And if it fails stress tests and logic tests then it should not be used for general purpose uses. It’s not ready for that, and in many cases it is slower and dumber than just reading a web search or Wikipedia search.
Don’t ask it to do much math. It’s not built for that. Fails badly.
If I just want a simple answer to something of course i'll just look at web search or Wikipedia. I main use it for grunt work. It saves me a lot of time. I'm not really interesting in using it to get answers for stuff I could just read manufacturers web pages for, or look up on Wiki - people seem to think it is just a search engine and not engage in the stuff it can do that a search engine can't.
Sure the world is moving towards more AI or natural language, but these ChatGPT seems unrefined and a fad. I hope it will evolve to something more practical with accuracy. Do I want it to be part of the OS, not really.
Problemo. LLMs consume stupid amounts of energy.
Companies have to pay for this energy or pass it on to users.
So you want Siri to be an LLM? That cost will come to Apple customers and if the energy footprint of all LLMs is really big then energy prices will rise for everyone, including people who never asked for this stuff and don’t need it.
That means old people on a pension, retirees, the poor, the disabled. All paying more for energy just because some corporations think this is a cash cow for them.
Then there is a private on-device Siri. An LLM on-device would be limited and pointless.