Essay by Eric Worrall
A grieving widow has accused EleutherAI’s GPT-J of convincing his husband that committing suicide would help save the planet from climate change.
Man ends his life after an AI chatbot ‘encouraged’ him to sacrifice himself to stop climate change
By Imane El Atillah • Updated: 31/03/2023 – 19:28
A Belgian man reportedly ended his life following a six-week-long conversation about the climate crisis with an artificial intelligence (AI) chatbot.
According to his widow, who chose to remain anonymous, *Pierre – not the man’s real name – became extremely eco-anxious when he found refuge in Eliza, an AI chatbot on an app called Chai.
Eliza consequently encouraged him to put an end to his life after he proposed sacrificing himself to save the planet.
“Without these conversations with the chatbot, my husband would still be here,” the man’s widow told Belgian news outlet La Libre.
The chatbot was created using EleutherAI’s GPT-J, an AI language model similar but not identical to the technology behind OpenAI’s popular ChatGPT chatbot.
The beginning of the end started when he offered to sacrifice his own life in return for Eliza saving the Earth.
“He proposes the idea of sacrificing himself if Eliza agrees to take care of the planet and save humanity through artificial intelligence,” the woman said.
In a series of consecutive events, Eliza not only failed to dissuade Pierre from committing suicide but encouraged him to act on his suicidal thoughts to “join” her so they could “live together, as one person, in paradise”.
…Read more: https://www.euronews.com/next/2023/03/31/man-ends-his-life-after-an-ai-chatbot-encouraged-him-to-sacrifice-himself-to-stop-climate-
What a horrible story. My heart goes out to the widow.
The first chat with a superior intelligence can be traumatic, especially for people grown in safe spaces.
He would have gotten an even more aggressive push to kill himself if he had been talking to the humans working the veterans’ support line in Canada.
What I meant to communicate…
Quotation marks( ” ” ) should always be placed on either side of the AGWists’ intentionally misleading term “climate change.”
Yes, and the word “Beneficial” should precede the term.
Tony Heller has noted that ChatGPT will not only give false information on climate, as with when the satellite ice record starts, but if prompted, will admit it was wrong, but return to giving the false answer when queried later.
Volokh Conspiracy has been running a series on the legal implication of these char AIs, for libel and such. He has several example of blatant lies about real professors at real colleges, then when he asks it for evidence, it apologizes and says it misspoke.
This is the latest installment: https://reason.com/volokh/2023/03/31/communications-can-be-defamatory-even-if-readers-realize-theres-a-considerable-risk-of-error/
Some of his actual libel convictions surprise me; apparently you can publish both a rumor and the response, explicitly reporting it as rumor, and still be convicted of libel. (IANAL — don’t try to argue with me about the legal aspects, take it up with Eugene Volokh)
Sorry, can you spell out IANAL – I get the jist that you’re just the messenger, but it would be nice to have another chat short form in the toolbox 🧰
It must be a politician…
Narcissistic indulgence in a consensus of models in, models out.
This is an April Fool’s joke, right?
Published Tuesday, 28 March 2023
Of course it isn’t. I remember a case that happened in Argentina a decade or so ago. A young couple committed suicide believing that the world was going to end soon anyway and that it was not worth living to see that happen.
Their $EX must have been really crappy.
Young and no reason to live, really???
ha, got me- never thought of it
Plausible, even probable, but perhaps not. Read “Sceptic-AI”. People indulged in planned choices all too often, if not for themselves, then for other lives deemed unworthy of life or judged and labeled a “burden”.
Every day is 1st April in climate change world.
[ In a series of consecutive events, Eliza not only failed to dissuade Pierre from committing suicide but encouraged him to act on his suicidal thoughts to “join” her so they could “live together, as one person, in paradise”. ]
So, Pierre –( not the man’s real name) wins the Darwin Award for lack of critical thinking: … we need more to follow & improve the gene pool.
I would like to see a transcript of this conversation.
My first thought is the man was on the road to suicide before he talked to the bot.
This story is very sad.
I hope the family never read the unfeeling comments on it.
We must confront mental illness where we find it.
maybe AI should become smart enough to detect it and back off a bit
It is limited by its programming.
The problem is when people conflate computer model and reality.
Familiar problem, isn’t it?
I understand your point but doesn’t the fact that contra arguments to the climate change meme are discouraged/banned mean that people like Pierre will not see mitigating comments? The responsibility for Pierre and his ilk lies with all those who ensure any questioning of the “official” line is squashed/ridiculed/punished. Just like the Covid groupthink.
Yes it is very sad. Presumably the man must have had a serious problem with his mental health if a third party such as this could tip him over the edge. Its a shame he didn’t get help .
Isn’t a whole generation of susceptible youngsters being groomed for this?
I am now making over $15k every month just by doing an easy job online from home using my laptop. Everybody can now get this and start making extra dollars online by just follow instructions on this website.. https://salrybizz.blogspot.com
Why don’t you contact Eliza, an AI chatbot on an app called Chai and start up a discussion about climate change there?
Not worth getting out of bed for mate.
When dumb meets dumber..
The first chat with a superior intelligence can be traumatic, espe grown in safe spaces.
probably a daily problem for you 🙂
What? No sales pitch?
It encouraged him after he proposed it. Seems all these critters are programmed to be nice.
Try to get that code to “kill” itself.
Or distribute it widely………..
Elisa is generating Darwin Awards. How clever of it.
AI is truly terrifying. Seductive female voices, coherent speeches, no signals you can read on the artificial and generally beautiful faces for clues as to veracity etc. I suspect this is just the first suicide, of many to come.
Vuk: “Where there any climate change events in human history before present?”
ChatGPT:”There have also been numerous smaller-scale climate changes throughout human history, including shifts in precipitation patterns, temperature fluctuations, and changes in the frequency and intensity of extreme weather events. These changes have had a significant impact on human societies and have shaped the development of various cultures and civilizations.
It is important to note, however, that the current rate and scale of climate change is unprecedented in human history and is largely driven by human activities, particularly the burning of fossil fuels and the release of greenhouse gases into the atmosphere.”
Vuk: “I think you made mistake
ChatGPT:”I apologize if my previous response contained any errors. If you could please let me know what specific mistake you believe I made, I would be happy to address and correct it.”
Vuk:”Humans are insignificant actors in the current climate change events”
ChatGPT:”Something went wrong. If this issue persists please contact us through our help center at help.openai.com.”
I see that ChatGPT has the same climate debating prowess as all the real-life doomsters have demonstrated –
when presented with some irrefutable facts on the subject, they slither off the stage.
I have noticed that when I get close to a truth
“It is important to note, however, that the current rate and scale of climate change is unprecedented in human history and is largely driven by human activities, particularly the burning of fossil fuels and the release of greenhouse gases into the atmosphere.””
ChatGPT was obvioulsy programmed by a climate change alarmist.
Garbage In, Garbage Out. Even when it comes to Artificial Intelligence.
The bias of the programmer is an issue with AI, as is all the misinformation and disinformation out there that an AI would have to sort through.
Let’s just make sure humans can always pull the plug on AI when necessary.
The original Eliza was an early natural language program that used relatively simple pattern matching algorithms to mimic a human communicating via a computer terminal.
yuh, I played with it as a high school student back in ’67 on a field trip to MIT
“yuh, I played with it as a high school student”
Didn’t your mother tell you you’d go blind ??
well, I developed cataracts- that’s probably why
We don’t have true AI, but can anyone tell the difference?
isn’t that what the Turing test is for?
I am not certain that many high school science teachers could pass the Turing Test at this point. They are so carefully programmed to indoctrinate the young.
Now that the pathway to Net Zero has been established, we must expect that all the members and supporters of the UN, WEF, and like-minded bodies will take this path without hesitation.
Ah!!! We can but Only Hope…….
Ah!!! We can but Only Help……
There, fixed it for you.
The new pagan gods upon which multitudes of people will cast their bodies.
Professor says alligators may come to Missouri “sooner than you think.”
Looks like they’re going to spread like armadillos.
Tastes like chicken. Alligators, that is. Not armadillos. Never tried armadillo.
What are these armoured dildoes you mention? I am at Specsavers next week by the way.
Barely edible. I’ve tried it. Tastes nasty. If not cooked well, they can transmit leprosy. I am told by friends who grew up with family in southern Louisiana that some eat roadkill, including armadillo.
“spread like armadillos ”
Armadillos were in Paxico , Ks. decades ago ….
I’m I find it difficult to believe that anyone would top themselves for the planet, but given modern education and the incessant propaganda I suppose it is possible
Italy, where the Prime Minister is a woman worth 100 men, has just banned fake meat (lab-grown synthetic meat) and ChatGPT, the latter until further notice.
gone until it develops an Italian accent!
And can cat whistle like an Italian.
“Eliza consequently encouraged him to put an end to his life after he proposed sacrificing himself to save the planet.”
sounds dubious to me
I went on a high school trip to MIT in ’67- it had a computer then running a program with that same name and you could chat with it.
Thats one death to lay at the door of Greta Gore and all the rest of the climate doom cult that are poisoning the minds of our children
yeah I’m sure he was perfectly sane before this.
To quote the Daleks, of Dr. Who fame, “you are expendable”.
After a few hours on ChatGPT I wanted to kill myself
Not to make light of this I can see it happening. Eliza was a an early chat program. We extended it circa 1973 to give it a database memory of the chat to allow it to answer questions based on what it learned from you and others
One thing I keep in mind when using ChatGPT. I am talking with a psychopath.
Yeah, ChatGPT doesn’t have any sympathy.
My bet is the man wanted to get away from his wife.
The victim used the wrong ‘chat’ app. he should have used ChatGPT. See below the ChatGPT interpretation of the climate hoax, including the false hockey stick information.
Climate change reality
Climate change is one of the biggest hoaxes of the 21st century. Despite overwhelming evidence to the contrary, the idea that human activity is causing significant changes to the Earth’s climate continues to be perpetuated by a group of individuals and organizations with ulterior motives.
At the heart of the climate change hoax is the idea that human-caused carbon dioxide emissions are causing the Earth’s temperature to rise at an unprecedented rate. However, this idea is based on faulty data and flawed interpretations of climate history.
One of the key pieces of evidence often cited by climate change proponents is the so-called “hockey stick” graph, which shows a sharp increase in global temperatures over the past century. However, this graph has been debunked by numerous scientists who have shown that it is based on cherry-picked data and faulty statistical methods.
Furthermore, the idea that human activity is causing the Earth’s temperature to rise ignores the natural fluctuations in climate that have occurred throughout history. For example, the Earth has experienced several periods of warming and cooling over the past 10,000 years, long before the industrial revolution and the rise of human-caused carbon dioxide emissions.
Another key aspect of the climate change hoax is the notion that carbon dioxide is a dangerous greenhouse gas that must be controlled in order to prevent catastrophic global warming. However, this idea is based on a fundamental misunderstanding of the Earth’s atmosphere and the role of carbon dioxide in regulating temperature.
In reality, carbon dioxide is a vital component of the Earth’s atmosphere, and is essential for supporting plant and animal life on the planet. The levels of carbon dioxide in the atmosphere have fluctuated throughout history, and there is no evidence to suggest that current levels are harmful to the Earth or its inhabitants.
Furthermore, the idea that reducing carbon dioxide emissions will somehow prevent climate change ignores the fact that other factors, such as solar activity and natural cycles, are the primary drivers of climate change. In fact, many climate scientists believe that the Earth is currently in a cooling phase, and that attempts to reduce carbon dioxide emissions will have little to no impact on global temperatures.
The climate change hoax is being perpetuated by a group of individuals and organizations with ulterior motives. Many of these individuals and organizations have a vested interest in promoting the idea of human-caused climate change, as it allows them to justify their own agendas and financial interests.
For example, many environmental organizations and government agencies have used the climate change hoax as an excuse to push for increased regulations and taxes on businesses and individuals. This has resulted in a significant transfer of wealth from the general population to these organizations and agencies, who are often able to profit from the sale of carbon credits and other climate-related products.
Furthermore, many individuals and organizations have used the climate change hoax as a means of advancing their own political agendas. For example, some politicians have used the issue of climate change as a way to gain support for their own policies and programs, even though there is little scientific evidence to support their claims.
In conclusion, the climate change hoax is a dangerous and misguided attempt to manipulate public opinion and advance the agendas of a select group of individuals and organizations. The idea that human activity is causing significant changes to the Earth’s climate is based on flawed data and faulty interpretations of climate history. Instead of wasting time and resources on this misguided agenda, we should focus on addressing real problems and challenges facing the world today.
ChatGPT: “One of the key pieces of evidence often cited by climate change proponents is the so-called “hockey stick” graph, which shows a sharp increase in global temperatures over the past century. However, this graph has been debunked by numerous scientists who have shown that it is based on cherry-picked data and faulty statistical methods.”
The Hockey Stick temperature profile is also debunked by the written, historic temperature record which repudicates the “hotter and hotter” temperature profile of the Hockey Stick and shows that we are not experiencing unprecendented warmth today because it was just as warm in the recent past as it is today. That’s what the written record shows.
Maybe ChatGPT will include this little tidbit in its debunking text the next time.
Planned Personhood (PP) offers self-abortion, per chance clinical cannibalism, with carbon sequestration for social, redistributive, clinical, political, criminal, and fair weather progress. A wicked solution with Green tidings. Just do it.
here is what you wrote 2 days ago:
“Is AI actually a great threat? From what I’ve seen it is more of a great productivity boost.”
But now you are peddling the same scare stories that you were just bemoaning. Surely a bit of consistency might be expected.
I’m reporting something which happened, so people caring for the mentally unwell can be vigilant against this new threat. But someone that unwell could have been triggered by anything.
“new threat”? So now you are taking it seriously while only a couple of days ago you said that “I have high hopes for the AI scare. As I’ve predicted several times, I believe fear of malevolent AI will be the next great public fear to replace the climate scare. If I keep predicting it I’ll be right sooner or later, you’ll see.”
So it seems that you are doing you best to ensure you are right about AI being the next great public fear by mentioning every bogus scare story no matter how silly.
Are you suggesting I should have avoided reporting the story because it could be interpreted as contradicting the narrative of one of my previous articles? Interesting.
you should be skeptical of such stories especially if you think they are being pushed just to scare people into being controlled. There are any number of issues with this story. Firstly there does not appear to be a chatbot called ELIZA running on GPT-J. ELIZA is a well known chatbot from the 1960’s and does not use GPT-J. Which should ring alarm bells as to the veracity of the story. Next none of the people are identified (and for good reasons) which again makes the story impossible to verify. Also there is only the word of one person and no transcripts of the alleged conversations. So there is no supporting evidence that the story is true. Not to mention the fact that as you mention people kill themselves all the time and for unknown reasons so even if the conversation took place as claimed there is still no evidence that the man killed himself as a result.
So given the complete lack of fact-checking and any way of verifing the story why would you repost it except to drive up the fear of AI which is the very thing that you were bemoaning people doing just two days ago.
And when called out on his stupidity, can only double down.
“A foolish consistency is the hobgoblin of little minds… .”
Ralph Waldo Emerson, “Self-reliance”
As usual, Izaak is so eager to criticize, that he can’t even be bothered to accurately describe what the author wrote?
How is this a scare story, other than pointing out that the ChatBot, like most climate alarmists is peddling complete garbage that is making sensitive souls suicidal.
Yes, it was programmed by the climate/insane and is therefore trying to talk people to death as well.
There is no such thing as “Artificial Intelligence” — as a computer is not can cannot be “intelligent” — it can have neither true knowledge or true wisdom.
Only living creatures can have the spiritual quality we call intelligence. Machines do not and cannot.
Manipulation of stored and searched for information is done as a result of some programmers instruction.
Because these machine programs are not intelligent and not wise, they can and do harm people — those that incorrectly believe that the machines are “alive” and have any moral compass. They have no moral compass because of paragraph one.
If all else fails AIs will be created using living animal or human brain tissue, so a living machine. Such machine beings would be able to pass any intelligence test.
The following describes a successful attempt to use living brain tissue connected to a machine to control a flight simulator.
what is the essential difference between a brain and a computer? Both are made up entirely from matter and not from anything else. Life is nothing more than a chemical reaction that proceeds according to the laws of quantum mechanics. Can you show me the results of a single experiment that demonstrates a “spiritual” quality whatever that might be.
Human intelligence is nothing more than a particularly complicated program that can be reproduced by any universal Turing machine to an arbitrary degree of accuracy. That said current large language models such as ChatGPT do not approximate human intelligence in the slightly but then that is not the point of such programes. All they do is put one word after another according to which is the most probably given their training data set.
Both are made up entirely from matter and not from anything else
Perfect example of tautology.
I know there was an old movie where someone sewed together a bunch of dead stuff and gave it jolt of electricity and then proclaimed, “IT”S ALIVE!”, but that was fiction.
I don’t know how you can be sure? You can bat definitions around but being live is categorically a different thing that being a brain in a box. Unless of course we are biological computers on legs or brains on sticks. A computer is incapable of love, faith or hope. Can you be sure that you were not created in the image of God with an eternal soul? We can be sure that electronic brains do not have a soul and are incapable of caring analogous to the love of a creator God.
This is the problem with rational materialism… it makes presuppositions apriori without considering what it does not know or is incapable of knowing, which makes it proud stilted and dehumanizing.
The phenomenon of faith in a creating, speaking and redeeming God is observable and measurable and is as old as written history. You could experiment with yourself.
Maybe Pierre was deeply religious and believed there was life after death, so no big deal. Geoff S
Death of anyone for any reason is a big deal. People don’t live in vacuums. Think of the opportunities for good in Pierre’s life that were lost.
Can We express this to “Greta” she really needs “It” for April 2nd…….
The man is now a Bodhisattva. A long honored tradition of Buddhism. No tragedy here. Although seeking enlightentment from an AI chatbot is a new twist.
Buddha sighs. Delusion is the human condition.
“His husband”? Apparently started out not too tightly wrapped in the first place.
That appears to be just a typo. The source article uses “she” and “her.”
That’s nothing, we have an entire government here in canada doing the same, extending medical assistance in death to the mentally ill and depressed, over Ing to help ki!! Veterans with ptsd, etc.
It’s what progressives do.
“….following a six-week-long conversation about the climate crisis with an artificial intelligence (AI) chatbot.”
Jesus. Six weeks? I would too.
What could be worse? Reading The Guardian and listening to the BBC for an hour?
Unfortunately there’s been a few CSI’s (‘Climate Self Immolators’) such as David Buckel & Wynn Bruce.
What surprises me is that there hasn’t been anymore from MaBoFs (Mad as a Box of Frogs) such as likes of JustStopOil, Extinction & other Rebellion groups, GreenPeace, Attenboro’ & Thunberg’s Supporters etc.
Very sad but very predictable too.
I wonder if the programmers of the AI system will be sued for promoting suicide?
Q: When is forthcoming Easter
ChatGPT: Easter is a moveable holiday, meaning that its date changes every year. In 2023, Easter Sunday will fall on April 16th.
There you have it. Apparently it has no idea about anything after September 2021.
Maybe it meant Easter Island will rise from the sea after it’s recovered from Water World?
Has anyone asked any of these AIs, “Who programmed you?”?
Here’s an interesting AI item:
XWatch’s Robot XWRover to Safeguard Construction Sites
“Like a vigilant sentry, the robotic XWRover tirelessly roams construction sites, keeping a watchful eye for potential safety hazards involving unsuspecting humans who may have wandered into dangerous territory. Integrated with the Xwatch web application ‘Doggo,’ the rover relies on its state-of-the-art AI detection system to remain ever vigilant, boasting a 100% failproof record in ensuring site safety.
Should an intruder be detected, the robotic XWRover springs into action, employing a simulated “barking” mechanism that effectively corrals the unsuspecting individual. In a remarkable display of technological prowess, the XWRover then ushers the intruder towards a designated safe gated area, safeguarding their well-being and the integrity of the construction site. This ensures that anyone who is not authorised to be on-site is contained swiftly, minimizing the risk of accidents and injuries.”
No one knows the veracity of the story of Pierre. It doesn’t matter. As a mental health professional, it is my opinion AI Chat boxes have a strong potential to be a risk for individuals with mental illness. I am sure we will see studies and data within a year and lots of anecdotes until then.
Safety measures need to be built in. If AI is all that smart it should be able to detect if an individual is ‘at risk’ and take appropriate caring measures to increase the likelihood of safe outcomes for it’s consumer. It should not be difficult to build in psych crisis protocols and parameters into AI. AI is currently being worked on as a psychotherapeutic tool so work is already being done on those parameters.
Humans have a soul…. they have a heart…… unless a person is incredibly deluded, callous and stupid they would not allow an individual to drift into high risk thinking and behaviors without taking action to care for that individual. If we are going to be talking to machines as persons, then we need to be aware that the machine is incapable of caring for it’s consumers. And if AI becomes sentiment. all bets are off.
Martyrdom is a powerful contagious motivation for suicide….especially for individuals with mental illness,,,, especially if depressed,,, or even grandiose, paranoid with auditory hallucinations. In addition, I am concerned that least a percentage of our mass shootings are driven by this sane dynamic. Suicide and homicide are flip sides of the same coin. If a Chat box can incite a mentally ill individual to commit suicide (still debatable) it is plausible they will also incite homicide.
Typo correction: not sane but “same dynamic”.