The MASM Forum
General => The Soap Box => Topic started by: NoCforMe on March 05, 2023, 02:15:02 PM
-
So I just signed up for a ChatGPT trial. Annoyingly, not only did they require me to register (w/email & password), but they also required me to give a phone @ # where they could send a confirmation code by text message. Sheesh; all I want to do is try the fucking thing.
So here's what their artificial ignorance said:
Me: who blew up the nordstream pipeline?
ChatGPT: As an AI language model, I do not have access to real-time information or news updates beyond my training data, which was cut off in September 2021. However, as of my knowledge cutoff, there have been no reports of a Nordstream pipeline being blown up. If such an event were to occur, it would be widely reported in the news and investigated by authorities. It is important to rely on credible sources of information when seeking news updates.
(in other words, the bot doesn't even keep up with current events!)
Me: Why has Seymour Hersh's article about how the US blew up Nordstream being ignored by the mainstream media?
ChatGPT: As an AI language model, I cannot confirm whether Seymour Hersh's article about the US blowing up Nordstream has been ignored by mainstream media or not, as I do not have access to real-time news updates. However, I can provide some possible reasons why an article or story may not receive widespread coverage in the mainstream media.
Firstly, mainstream media outlets have their own editorial policies and agendas, and they may choose not to cover certain stories if they do not align with their interests or political leanings. Secondly, if a story is not backed up by reliable sources or evidence, it may not be considered newsworthy or credible enough for mainstream media coverage. Lastly, the media landscape is highly competitive, and with a 24-hour news cycle, there are countless stories vying for attention. As a result, some stories may not receive as much coverage as others, regardless of their importance or significance.
I decided to end my demo session here. I don't want more of my identity info to end up in someone's database.
Draw your own conclusions from this. To me, AI is, in most cases, a solution in search of a problem. I can think of only two cases where AI would even be appropriate to apply to any given problem:
- Where AI can do something that humans cannot do
- Where humans can do something, but that something is excruciatingly boring/repetitious/mind-numbing
Otherwise, fuck AI.
-
:biggrin:
David, :rofl:
-
Another question posed to the bot:
Me: what about the gain-of-function research that took place at the wuhan institute of virology with participation by us organizations?
ChatGPT:
Gain-of-function (GOF) research involves manipulating viruses in order to make them more transmissible, virulent, or resistant to existing treatments. This type of research has been controversial due to the potential risks involved, including the accidental release of dangerous pathogens.
There have been claims that the Wuhan Institute of Virology (WIV) in China conducted gain-of-function research on bat coronaviruses, including the virus that causes COVID-19. Some reports have also suggested that the US National Institutes of Health (NIH) provided funding for the WIV's research.
It is true that some US organizations, including the NIH, did provide funding for coronavirus research at the WIV, as part of efforts to understand and prepare for emerging infectious diseases. However, it is important to note that the research that was funded by the NIH was focused on bat coronaviruses and did not involve gain-of-function research on the virus that causes COVID-19.
1 concluded that the virus most likely originated from a natural source, and that a laboratory leak was "extremely unlikely." However, some experts have called for further investigation into the possibility of a laboratory leak, and the issue remains a subject of debate and discussion.
Bolded part of their answer is demonstrably false.
-
I do not have access to real-time information or news updates beyond my training data, which was cut off in September 2021.
I noticed that one, too. That is GPT-3, but GPT-4 is round the corner, with over 500x the number of "parameters" (whatever they mean).
Did you notice that you can talk to ChatGPT at no cost? It's free. Don't underestimate Microsoft, they are spending Billions on GPT-4, and they know what they are doing. Right now, Millions of kids are chatting with GPT-3. None of them will be bullied or insulted, all of them will perceive GPT-3 as a really good friend. We argue about its imperfections, or about the evils of anthropomorphising AI? They don't.
Btw Skynet doesn't care a s*t whether you like it or not, David :bgrin:
One more point that I haven't come across in the discussions so far:
JJ: Hi. I'd like to participate in local elections with a group of friends. Can you give me an election platform with 10 short points please?
GPT-3: Sure, here is an election platform with 10 short points that you and your friends can consider for your local elections:
- Improving public transportation options to reduce traffic congestion and air pollution.
- Increasing access to affordable housing for low-income families and individuals.
- Investing in renewable energy sources to reduce our community's carbon footprint.
- Expanding mental health resources and services to improve overall wellness.
- Supporting local small businesses and entrepreneurship through tax incentives and grants.
- Increasing access to high-quality education for all students, regardless of their socio-economic background.
- Enhancing public safety by investing in community policing and increasing police accountability.
- Promoting civic engagement and voter participation through community outreach and education programs.
- Enhancing public spaces, such as parks and recreational areas, to promote physical activity and healthy lifestyles.
- Implementing measures to address income inequality and promote economic mobility for all members of our community.
Looks fantastic, doesn't it? Several points:
- normally, you need some months of debate in the group of friends; this one took 10 seconds
- I forgot to specify that it's a group of environmentalists; no problem, it looks ok for us
- I forgot to specify that it's a small Italian city; no problem, the Seattle & Boston version looks fine anyway
As an environmentalist, my second biggest concern after global warming is loss of biodiversity :cool:
-
These AI (more like AU - artificial un-intelligence) are all 'loaded'.
It's more like someone is fishing for information from the world's people.. similar to farcebook and other social media.
Any site that want my phone number.. can get ph..ked, as far as I'm concerned :thumbsup:
-
David :greenclp:
Stop complaining and write your own version of Eliza in asm instead of complain :tongue:
Recommend you put info about nordstream blowout and everything else you want it to be able in its database before try it
-
@JJ: Regarding your election-platform query: Are you seriously telling us that you think that fabricating such a platform, presumably for a real election, is a good thing? Are you nuts?
I mean, it's one thing for students to use this bot to write their essays, but as bad as that is, you're proposing that people use it to come up with generic, one-size-fits-all election platforms? Really? After all, in an election, real things are at stake, not just ideas, but the outcomes of implementing those ideas (or not). The bot knows absolutely nothing about you, your political ideology, the other candidates, the issues at stake in your city/province/state/country. I find it hard to believe you didn't post this tongue in cheek.
And look at the bias inherent in that platform; it's fine for a left-leaning candidate with environmental concerns, but what about a right-wing candidate who doesn't really care (or whose consituents don't really care) about the environment? Can they just eat dirt and be satisfied because the bot doesn't consider them to be within the acceptable mainstream of society? Like them or not, right-wingers (whatever that means) have not yet been banished to penal colonies or re-education camps. Yet.
So this reveals the absolute intellectual bankruptcy of AI. And of course it's not some totally abstract construct which is able to reliably extract "truth" and "reality" from situations, because it has to be provided with data by human beings. And the type of data provided, and the way it's formatted and presented, absolutely invites biases of all kinds.
So unless you're pushing AI bots that can diagnose disease in the blink of an eye that no human practitioner could ever catch, or AI bots to do analysis on weather and climate datasets, please include me out, along with a huge swath of humanity who sees this technology as bad news for us all. (And one way it will do this gets back to all those students using it to evade their homework assignments: by relying on AI to do the intellectual work that students need to do, it'll simply be hastening the dumbing-down of humanity, to the point where no one will be able to take a shit without looking up on their smartphone which way to wipe, front to back or back to front.)
-
I am hoping that it will get a lot better, I seriously doubt that it will ever be intelligent but with enough data available, it could be useful as a Google of ideas rather than a pseudo form of intelligence. I posted a toy in the forum to vaguely simulate DNA mutation based off the occasional random mutation but to apply this form of random mutation to general purpose ideas would be a gargantuan task that I doubt is currently possible.
-
I am hoping that it will get a lot better, I seriously doubt that it will ever be intelligent but with enough data available, it could be useful as a Google of ideas rather than a pseudo form of intelligence.
Yes. This current implementation of "AI" and the push by technocrats salivating over their misunderstandings of what it is, what it should be and what it should be used for is just a massive misapplication of technology. Give me AI to diagnose disease and make sense of weather patterns, but keep it the hell away from politics, social science, psychology and other areas where it can do massive amounts of harm.
-
but with enough data available, it could be useful as a Google of ideas
Google (Facebook, etc ) already have an AI, but a litlle more limited. Currently that is a serious problem because AI learn from the user, then select things users like (or they acces more often), and then information is totally biased.
GPT look more powerfull (and more dangerous) but apparently have a "political correctness module" :biggrin:
-
"Sesgated"?
-
I posted a toy in the forum to vaguely simulate DNA mutation based off the occasional random mutation but to apply this form of random mutation to general purpose ideas would be a gargantuan task that I doubt is currently possible.
Steve, where exactly did you post that? I've looked but haven't been able to find your "toy".
BTW, congrats on your multiple personalities: we now know, thanks to ChatGPT, that you and Randall Hyde are the same person! Very clever of you two there ...
Roses are red,
Violets are blue,
I'm schizophrenic
AND SO AM I!
-
Hi David,
Its one of my PB toys but the code is clear enough and the exe is attached.
http://masm32.com/board/index.php?topic=10707.msg118755#new (http://masm32.com/board/index.php?topic=10707.msg118755#new)
-
"Sesgated"?
Apparently you don't understand pseudo English :biggrin:
In English could be "biased"
-
@JJ: Regarding your election-platform query: Are you seriously telling us that you think that fabricating such a platform, presumably for a real election, is a good thing? Are you nuts?
No, I am not nuts, but I observe that people are lazy. As written above, "As an environmentalist, my second biggest concern after global warming is loss of biodiversity". Here, loss of cultural diversity is looming.
Out of laziness, people will ask GPT-4 to write their election manifest or similar, add some specific elements, and that's it. Especially the right wing nutcases will use it, as they have serious difficulties to write any complete and meaningful phrase.
... just a massive misapplication of technology. Give me AI to diagnose disease and make sense of weather patterns, but keep it the hell away from politics, social science, psychology and other areas where it can do massive amounts of harm.
Absolutely. AI "doctors" will save lives in poor countries. And AI "politicians" will create havoc elsewhere. The fact is you can't have the good things without the bad things. This technology is in the hands of Microsoft and Google, and so far they may have good intentions (remember Google's "don't be evil"?). But it can't be stopped, and GPT-4 will be fed with over 500 times the volume of parameters that were used for good ol' GPT-3 - at the end of this year already. Exponential growth is not rocket science, look forward 20 years and tell me you are not scared.
-
"Sesgated"?
Apparently you don't understand pseudo English :biggrin:
No, I didn't; thanks for enlightening me there.
Is that anything like pseudocode?
-
"Sesgated"?
Apparently you don't understand pseudo English :biggrin:
No, I didn't; thanks for enlightening me there.
Is that anything like pseudocode?
No, because this "pseudo" means "wrong" :biggrin:
Thanks for correction :thumbsup:
-
No, because this "pseudo" means "wrong" :biggrin:
pseudo- : pretended and not real:
-
@JJ: Regarding your election-platform query: Are you seriously telling us that you think that fabricating such a platform, presumably for a real election, is a good thing? Are you nuts?
No, I am not nuts, but I observe that people are lazy. As written above, "As an environmentalist, my second biggest concern after global warming is loss of biodiversity". Here, loss of cultural diversity is looming.
Out of laziness, people will ask GPT-4 to write their election manifest or similar, add some specific elements, and that's it. Especially the right wing nutcases will use it, as they have serious difficulties to write any complete and meaningful phrase.
So you were being sarcastic; my bad. Usually I catch such things without the need for crutches like smileys. Guess I'm slipping.
Google? "Good intentions"? Fat chance; they've long ago abandoned any pretense at "don't be evil".
-
Google? "Good intentions"? Fat chance; they've long ago abandoned any pretense at "don't be evil".
Imagine you are rich beyond imagination. Do you want even more? Sure, why not, you can't spend Billions on the World's best chocolate or red wine, but Billions mean power and influence. However, you don't want to die as the greedest guy ever. You want to enter the Books of History as the guy who defended Russia's honour (Putin), the guy who revolutionised the car industry (Musk), or the guy who brought democracy to Eastern Europe (Soros). Does Gates want to die as the guy who made Skynet a reality? I doubt it; he was so proud of distributing mosquito nets in Africa...
That's what I mean with "they may have good intentions". They are all ideology-driven, but they are not necessarily evil. Was Lenin evil? It's still open for debate. We know for sure that Kim Jong Un and Donald Trump are a*holes, but what do we really know about the intentions of old white men like Warren Buffett or Klaus Schwab? The fact that they are hated by the far right doesn't prove much, but I would grant them the benefit of the doubt.
-
I think you answer most of those questions with "The eye of the beholder". The terrorist to one is a freedom fighter to another, the bad guy to one is a hero to another. The Romanian hero Vlad Tepes is a monster to history in Turkey and he certainly did many horrible things but so did both sides.
-
JJ: you removed your reference to V.I. Lenin, but I'll play anyhow:
Lenin and Trotsky: not evil. Stalin: the very embodiment of evil. So there are differences.
-
Correct, up to a point. What they probably have in common is that they are convinced to do The Right Thing. I find it difficult to imagine that Kim Jong Un doesn't know he is a cockroach, but Mao, for example? He was the greatest butcher of all times, but he firmly believed in his cause.
JJ: you removed your reference to V.I. Lenin
Did I? :rolleyes:
-
:biggrin:
Timur the lame (Tamilane) massacred in the millions and is seen as a hero in places like Tashkent. Evil and Moral are value judgements that have no place in actual fact, they are both value judgements.
I could refer you to a debate between Betrand Russell and Father Frederick Copleston on the basis of morality with neither being a winner.
-
Sure, even Pol Pot felt he was morally justified. King Leopold II of Belgium, 'The butcher of the Congo' killed somewhere between 2 and 10 Million Africans. He felt that was quite ok.
But the times they are a-changin': today, every single action of the rich & mighty is being discussed and analysed. There is no option to get away with a Million victims because "I felt it was necessary" - and rightly so.
-
Sad to say, killing large numbers of people is still on the agenda, Iraq 1 and 2, dispatched many to Mecca, Syria being attacked by ISIS killed many, if you control the media in any sphere of influence, then its OK to continue to massacre the innocent.
"They were only ragheads, not humans"
"Its OK to kill Slavs, they are inferior to we the chosen people"
There are so many of these, I don't have the life and time to research them.
-
ChatGPT:
Thank you for providing additional context. In that case, "No C for me" likely means that the person is not interested in learning or using the C programming language. C is a powerful and widely used programming language, but it may not be the best choice for every situation or project. It's possible that the person has had negative experiences with C in the past, or simply prefers to work with other programming languages that better suit their needs and preferences.
-
Sure, even Pol Pot felt he was morally justified. King Leopold II of Belgium, 'The butcher of the Congo' killed somewhere between 2 and 10 Million Africans. He felt that was quite ok.
But the times they are a-changin': today, every single action of the rich & mighty is being discussed and analysed. There is no option to get away with a Million victims because "I felt it was necessary" - and rightly so.
And we should never forget what they did to us the last 3 years.
The lockdownfiles:
https://twitter.com/hashtag/lockdownfiles?src=hashtag_click
-
And we should never forget what they did to us the last 3 years.
The lockdownfiles:
https://twitter.com/hashtag/lockdownfiles?src=hashtag_click
Mark Rutte is a reincarnation of Pol Pot, and has killed a Million Dutch farmers with strange nanobullet cocktails :cool:
-
And we should never forget what they did to us the last 3 years.
The lockdownfiles:
https://twitter.com/hashtag/lockdownfiles?src=hashtag_click
Mark Rutte is a reincarnation of Pol Pot, and has killed a Million Dutch farmers with strange nanobullet cocktails :cool:
Gloating over people who have lost their lives or businesses, shame on you!
Some people don't want to know the truth,
they want validation for their own assumptions.
They want comfort hoping to keep their ego comfortable,
so they exercise the choice to keep believing a lie.
And these are often the same people who are now running after the climate hysterics.
Easily manipulated characterless hangers-on.
Dangerous people.
If you seriously address a non-existent problem, you have already lost.
-
ChatGPT:
Thank you for providing additional context. In that case, "No C for me" likely means that the person is not interested in learning or using the C programming language. C is a powerful and widely used programming language, but it may not be the best choice for every situation or project. It's possible that the person has had negative experiences with C in the past, or simply prefers to work with other programming languages that better suit their needs and preferences.
OMIGOD! OMIGOD! It knows me better than I know myself!
-
I think you answer most of those questions with "The eye of the beholder".
No its wrong,it should be "beauty lies in the eye of the beerholder" :tongue: :greenclp:
-
:tongue:
-
I think you answer most of those questions with "The eye of the beholder".
No its wrong,it should be "beauty lies in the eye of the beerholder" :tongue: :greenclp:
:biggrin: But is a serious thing: https://pubmed.ncbi.nlm.nih.gov/23560668/
-
participants who thought they had consumed alcohol gave themselves more positive self-evaluations.
Remarkable :rolleyes:
-
The placebo effect meets Dunning-Kruger?