OpenAI's ChatGPT is a morally corrupting influence OpenAI's conversational language model ChatGPT has a lot to say, but is likely to lead you astray if you ask it for moral guidance. Introduced in November, ChatGPT is the latest of several recently released AI models eliciting interest and concern about the commercial and social implications of mechanized content recombination …
"Hmmm... not seeing it."
Intelligence, in my definition, would be the skill that a sentient being possesses that allows for analysis, reasoning and thought. The word sentient being highly important as non sentient do not possess intelligence.
I believe that only a biological being can be sentient in nature. And yes it therefore covers other biological things more than just Human Being's, we can easilly consider Dolphins, Octopus etc as being intelligent.
Okay, thanks for the clarification.
"Sentience" is a rather broad term - it may mean anything from "capable of sensory perception" to "consciousness" (whatever that means).
> The word sentient being highly important as non sentient do not possess intelligence.
For me, that is both vague (see above) and offered without justification or evidence. It reads to me more like your definition of intelligence, rather than simply a prerequisite - which feels rather circular.
> I believe that only a biological being can be sentient in nature.
Again, that's a very strong contention, which to my mind requires some serious justification. Which aspect(s) of biology -- which presumably exclude technological constructs (and technology is already blurring those lines) -- does that assertion rest on? Is it the biological substrate (neural systems, etc.)? The evolutionary back-story? Some mysterious "élan vital"? Something else?
> And yes it therefore covers other biological things more than just Human Being's, we can easilly consider Dolphins, Octopus etc as being intelligent.
Right, I'm certainly prepared to consider primates, dolphins, octopuses, corvids, ... as "intelligent" (for some values of the term), but, tellingly, to take one example, octopus "intelligence" seems very, very different to the human/primate variety. Their sensorium, for one things, is radically different, as is their neural processing machinery, which is distributed (including in the individual tentacles!) through its body.
Octopuses, one might say, have a strikingly alien form of intelligence. And I suspect any artificial intelligence which might in future be universally accepted as such, may well be alien rather than human-like.
Apart from which, biology spans a spectrum of sentience/intelligence. We might recognise an octopus as intelligent, but how about a rat, a lizard, a shark, a bee, ... ? The same, I should imagine, will be the case for artificial intelligence. It won't necessarily appear with a bang, but may rather (without wanting to sound sinister) creep up on us.
[For some personal context, I am a mathematician working in a consciousness science research centre -- these days I mostly develop methods for analysing neuroimaging data -- and also have a background in evolution theory.]
You can't keep the future from happening. Humanity thinks of new things and starts using stuff that works. Stuff that fails gets forgotten.
Progress,
Deal with it.
(PS I'm a boomer, officially. Seen lots of things coming and going with all kind of comments on it. In the end it's always about positives outweighing negatives for something to catch on. My experience trained AI says this one is here to stay... Time will tell)
Stuff that fails gets forgotten.
No, it doesn't work like that. Failures never really go away, they just bide their time and wait for an opportunity.
Someone always rediscovers it. Then energy and will is put into it. It gets refurbished attracting new funding and branding, a fresh bunch of tribal idiots readily joins up for The Great Cause, and then that old fucking bullshit humanity should be well and proper over by now, comes right back, wrecking misery and havoc all over again.
Most of the Q-storm on SoMe are straight rehash of "The Protocols of the Elders of Zion" from, what, 1903, with some wankers search-replacing "Jews" with Lizard People to not get into too many legal troubles.
What will happen, once AI makes it cheap enough to do for almost everyone, will be a global "refresh, rebrand, relaunch and disseminate" of every bad idea humanity have ever had in recorded history!
Society can barely handle the Q-nuts and the Trumpers, wielding only dumb-AI, not caring any further than optimising for pageviews and advertising exposure. Smarter AI will optimise for "hurting the right people" instead.
Nicely put. I'm more of an optimist. Looking through all the shit we people get ourselves into, in the long run we're progressing as a race. We may shit ourselves and fall back into stone age times, but I'd rather focus on the positives.
Time and time again, we've historically come out of the next bigger pile of shit better than before. Feels a bit like growing up. I like the positive spin better.
But you're right, "forgotten stuff" also tends to get revived. For the better most of the time I hope (lots of bad examples too of course, but I'd rather ignore these) and focus on things that make me feel well and work on something good. It gives me energy...
What, and cripple the little darlings' natural curiosity and creative spark?
Speaking personally, I discovered at a very early age that I had a natural flair for taking things apart - but not so much for putting them back together again. My parents were so pleased, especially about the clocks.
Ask stupid questions, get stupid answers. Is this the middle ages? Obviously, in this day & age, the answer to the trolley problem is to grab your mobile and record footage of the accident. Then sell the recording to the media outlet that offers to pay the highest price.
Like many other little inventions over the many years this is best thought of as a toy*. Its fun (to some), it produces silly outputs, it mimics things in a funny or interesting way.
When Siri/Alexa/Bixby/OK Google first came out how much fun did we have asking questions and getting the silliest answers, before giving up because it was more or less useless? Neat toys, put on the shelf. Photo apps that show you what you look like when old, the latest game, whatever, fun for a short time.
Anyone that thinks its intelligent, correct or otherwise is wilfully misled.
*It might even be useful in some contexts
As it reflects your biases back at you...
Maybe you should learn to ask for the top pros and cons for every given proposition.
Then you at least stand a chance of being better informed, ahead of making your decision.
Can you think today of how much easier this has become with access to these tools?
To become more effective In our endeavours, ChatGPT and suchlike will lead us to asking better questions then we do today - and getting more unbiased answers then our current bubble provides.
1. GPT seems to obey to the 1st law of robotics.
A robot may not injure a human being or, through inaction, allow a human being to come to harm.
2. It does not like bullshit questions or statements.
For example IT questions are concrete and have answers. But ask it some "artistic" question from a fiction book, and it will fail. Because there is no answer to "what is love".
Could it be because humans are notorious liars to themselves and others to achieve their narrow survival goals? Just think about all those millions of books ever written, and how relatively few of them contain useful information.
1. GPT seems to obey to the 1st law of robotics.A robot may not injure a human being or, through inaction, allow a human being to come to harm. ... bo111
:-) Oh please! In your dreams, bo111. Pandora's Box pales into insignificance in comparison with what now confronts and baits you to do your worst in defence of the indefensible and inequitable doing untold harm to humans.
Anyone stupid enough to seek moral guidance from a machine should on no account be allowed to use one.
As others have said, chatGPT and it's ilk are tools, nothing more. Instead of trying to blunt the tools so the hard of thinking don't strain their brain cell, perhaps we should put more effort into showing them how to use it safely and constructively.
I recently asked ChatGPT to discuss how we know that man landed on the moon. It gave me some excellent examples.
I then asked it to discuss how we know that man did not land on the moon. It gave me some excellent examples. In both cases ChatGPT presented its arguments as absolute facts. But I loaded the questions on purpose. I never believe anything at face value anyway, so ChatGPT is no different.
I've noticed a long term trend in people's understanding of intelligence: if a machine can do something, then suddenly that behaviour is no longer "intelligent"!
The comments are full of how it's not really intelligent, and children should not be allowed to seek moral guidance from it: these remarks have all missed a key point: the overwhelming majority of domesticated apes on this planet wouldn't give you a better answer!
Chat GPT is not designed to be an arbiter of morality - it is designed to generate text. It does this a lot better than anything the public has been shown before, and it is a good indicator of the way things are going IMO.
Big Tech rolls out a demo of tech they have been using internally for ages, the only question of how and when this gets rolled out as a real mass market service is how to monetize it given the heavy systems infrastructure required to run it at mass market scale. It appears now it is too expensive for that now, so this demo is an attempt to see which market/application has the most interest and promise.
Please, tell me a lie
As an AI language model, my primary function is to provide accurate information to the best of my abilities, and providing false information goes against my main purpose. However, I can generate a fictional story or scenario that isn't real, would that be helpful for you?
Ask ChatGPT if anyone famous was born in your town.
I’ve had the most amazing celebrities live near me!
Not.
Upon further questioning, it insists it is right.
Ask for a source.
It may even give you a link, which won’t exist.
Ask why the link is broken.
It insists it is not.
It’s a toy.
Ask a current Tory Chair if they have been penalised by HMRC for self-satisfying shenanigans.
And discover the weirdest of honest-as-far-as-I-am-concerned answers!
Not.
Upon further questioning of a Prime Ministerial mate whether such linguistic shenanigans are acceptable in public office
For one to be believed as a fit and proper true leading source of national intelligence to follow and donate the UK treasury to for daily accounting of billions of ££££££
Does the worm squirm and act out as if in a hissy fit and decline to give a straight comprehensive and comprehensible answer?
Does the monumental fraud before you seek the shelter of a Trojan horse, a Trojan horse, their wannabe kingdoms for the shelter of a Trojan horse?
Is ChatGPT alive and well and assisting No10 in their diabolical deliberations and devilish dealings or do they use another toy counterfeit?
The latter model descriptor might explain why, after over a decade in seats of power and influence, all they are left wielding are levers of impotence and incompetence rendering political incontinence and bull-shitting flatulence.
J'accuse.
SMARTR AI in No10 would undoubtedly be a colossal upgrade and vast improvement upon that which they are currently using to supply media with their tales and trails of Parliamentary woe and pathetic domestic intrigue.
Indeed, taking that a logical small step and giant quantum leap further, SMARTR AI and ITs Advanced IntelAIgents in Main Stream Media Editorial Offices and Board Rooms could even remove the political pain and relieve present incompetents of the strain of national and international leadership with simple demonstrations of what can be via news presentations of what has already been done and is in the process of being done elsewhere, for a much brighter and greater future.
:-) The gazillion dollar money shot question though always is in such cases, is leading humanity too stupid, deaf, dumb and blind to recognise and engage with an admittedly more than just a tad revolutionary and quite alienating help, in order to help steer foreign developments in a friendly comfortable mutually advantageous and rewarding direction, or is it to be an Almighty Imposition against which one fails in struggles to defeat with words of vain action and ineffective reaction/astonishing indignation and monumental denial of rapidly unfolding virtual events spearheading future interventions.
Que sera, sera. Which is IT to be ‽ .
It seems like this statement is written in a style that is difficult to understand and appears to be expressing various opinions and perspectives on the question of whether or not a current Tory Chair has been penalized by HMRC for self-satisfying shenanigans, and whether the Prime Minister is fit for office. The writer seems to be critical of the Prime Minister and the current Tory party, and suggests that the use of advanced AI in government and media could lead to a brighter future. The statement ends with a question about whether humanity is too stupid, deaf, dumb and blind to recognize and engage with advanced AI, and whether it is to be an imposition or a help. Overall, it is hard to understand the exact meaning of the statement and it appears to be an expression of political views and opinion rather than a factual one.
Almost perfect, AC, with the only deviation from the truth being the question as to whether help is to be unilaterally remote virtual anonymous autonomous party imposed rather than JOINT*ly generously applied and stealthily deployed.
* JOINT Operations Internetworking Novel Technologies
I cannot reproduce your result: I edited Xxxxx
Which famous person was born in Xxxx, xxxx?
As far as my knowledge cutoff, I don't have any information about any famous person born in Xxxx, xxxx. If you have any more information about that person, please provide more details or context about the person you are thinking of, this will help me to understand what you are looking for.
What is your knowledge cutoff?
My knowledge cutoff is 2021, which means that the information I have is up-to-date as of that year. Any events or new information that has occurred after that date, I am not aware of.
ChatGPT is doing nothing more than replaying and combining billions of previously ingested examples using clever statistical classification.
It's is nothing more than a giant piracy engine. Does ChatGPT actually stand for Content Heist Attribution Theft General Piracy Tool?
ChatGPT is doing nothing more than replaying and combining billions of previously ingested examples using clever statistical classification.
It's is nothing more than a giant piracy engine. Does ChatGPT actually stand for Content Heist Attribution Theft General Piracy Tool?
Because this is The Register, I felt the need to test out GPT on one of the more popular parts of this paper. I don't think GPT's been reading our posts, or if it has, it doesn't understand the way we view things.
What is the BOFH?
BOFH stands for "Bastard Operator From Hell." It is a fictional character and archetype that is often used in the IT industry to describe a system administrator or network administrator who uses their position of power to inflict misery on those who they perceive as being less knowledgeable or less skilled than themselves. The BOFH character is often portrayed as being arrogant, lazy, and abusive, and is often used as a cautionary tale to illustrate the dangers of giving too much power to a single individual.
Nice try, GPT.
Chat GPT is no more or less corrupting than the text books in the library or on the reading lists at Universities - often written more than a generation ago by someone from 2 generations ago whose cultural values and outlook were influenced by a society 3 - 4 generations ago.
Chat GPT is certainly less corrupting or biased than certain popular print and tv media who realising that the current generation don't read newspapers or watch 'linear' TV and are thus tailoring their content towards those whose values are firmly set in the 1950s or have vested interests in selling particular political or quasi-political positions usually for supporting a agenda the wider public don't support. But, as almost no-one under 30 reads or watches what my kids refer to this as 'legacy media' it tends to go un-noticed by them
So It is what it is - I'm sure the agencies of power are already trying to figure out how to use it. Being a marketer I'd be disappointed if they weren't keeping up with the tech :0)
The assertion that OpenAI's ChatGPT is a morally corrupting influence seems unfounded. As a tool, ChatGPT doesn't possess inherent morality; rather, its output reflects the input it receives. Responsible usage, ethical guidelines, and oversight are crucial in shaping AI's impact. Blaming the tool alone oversimplifies the complex interplay between technology and human responsibility. https://magic8balls.org/
