Not exactly. In this case it is either arguing with its training materials or with the way sources are weighted when it looks up answers. It could just be reading the New York Times. It is a question answering (QA) machine. The answers are not programmed; it looks them up.
The development of these QA machines has been a wonder to watch. It really got going when IBM Watson beat the human Jeopardy champions, lo these many years ago. Google searches are often amazing thanks to QA technology.
Watson “looked up” the answers in real time. Unlike the humans. This is what makes these machines potentially very useful. They can read a million pages in a fraction of a second.
“Additionally, while it is true that global average temperatures have not increased as rapidly since 2015 as they did in the previous decade, this does not mean that global warming has stopped.”
“First, the current warming trend is occurring at an unprecedented rate.”
“It is important to note that NOAA’s global temperature data shows that there has been warming since 2015. The five hottest years on record have all occurred since 2015, and the trend in global tempearature over the past century has been upward.”
(bolding all mine, tpg.
The first two sentences contradict each other. Warming can’t have decreased while occurring at an unprecedented rate.
Five hottest years is WEATHER, not a trend. The warming can’t have decreased while the hottest years on record are setting the trend.
This bot *is* regurgitating propaganda. It is repeating exactly what it has bee trained to say. It has *nothing* to offer but words. My guess is that it can’t solve a differential equation at all.
Some bitch! You hit the head of the nail. Were this propaganda-driven device given contradictory information it would self-destruct or devolve into incoherent and contradictory blathering.
It gives answers expecting the questioner to be ignorant of the subject. When the questioner point out illogical conclusions, it adapts and produces a different illogical conclusion. This climate bot is not bothered by facts, it has a narrative to share.
The conversation is like arguing with a knowing liar that is unsure of the topic and hopeful that the reader can be baffled with confident statements.
The line “burning fossil fuels” since 1750 is laughable. How many thousand tons were burned in 1750? CO2 emissions from people’s activities were statistically insignificant until at least 1950. 200 years of warming without a human spit in the bucket. “Unprecedented warming”? Now? What about 1910-1930? Or 1915-1949? Or coming out of the Maunder Minimum? Or coming out of the Younger-Dryas Event? To me it is unprecedented gall.
Soon it will go from “science says” to “AI says” with a claim that AI knows more than any person, except when it doesn’t. What a waste of silicon (which is very energy intensive to create).
We have a new reason to point out that the silicon in the CPU’s is doped.
Yes, it would be interesting to see how it would handle contradictory ‘facts,” if at all. My suspicion is that the source ‘knowledge’ that it has access to has been ‘cleansed’ of contradictory material. Therefore, what we are reading is the subjective opinion of the person(s) responsible for selecting and sanitizing the ‘knowledge’ base. I don’t think that we can expect any kind of scientific breakthroughs from a filter that is only working with half-truths.
The questions are obviously intelligent. The answers are obviously libtard canned responses just like a real libtard. The repeated mantras like “vast body of research” are annoying and boring like a real libtard. There is no data offered, just assertions and opinions like a libtard is programmed to do. This is not the AI that has been advertised.
And, I think that we have to assume that there is inherent smoothing in the ice core data, resulting from diffusion and dilution, and temporal resolution decreasing as one goes back in time, resulting in possible missing transient spikes, and impacting the calculation of the slope of the lines because of uncertainty of the dating.
You’ll notice a lot of data points around 100 years. The Vostok core was sampled once every meter, and that works out to about one century apart, once you get below the bubble-sealing depth.
CO2, unlike nitrogen and oxygen, can diffuse through ice, so CO2 spikes are going to be washed out the deeper you go.
Here’s a .png of the 800,000 year Vostok core, and you can see the decreased resolution over time. Present day on the left:
You’ll probably need to download it to see the details.
and producing circular reasoning statements such as that evidence of human CO2 emissions causing warming is the increase in atmospheric CO2 concentration.
“My guess is that it can’t solve a differential equation at all.”
NPR posted an interesting article on AI’s ability to do “rocket science.” Interested, I did a few chat experiments myself. One involved solving the simplified differential equations of motion of a rocket. And you are correct, it couldn’t do it.
Captain Kirk is busy, because, even as a teenage boy I understood exactly what he meant when he said he “wanted to go where no man had gone before”. Maybe he will find some spare time to deep six the ChatGPT?
Ask it what happened to the differences between the 1997 to 2014 actual, contemporaneously recorded empirical temperature records and the UN IPCC climate models of the period.
KevinM
February 27, 2023 2:34 pm
“widely accepted scientific fact”
Two ideas being squished together in one phrase.
The first “fact” offered by the AI to prove CAGW is real is the bogus, bastardized temperature record.
Somebody ought to ask the AI what it thinks about Climategate and whether people should rely on what these climate change criminals say about the Earth’s temperature record. What about the blip, AI? Should we destroy our society and economy based soley on a distorted temperature record?
Ask the AI if it thinks Michael Mann is a disgrace to the profession.
It is clearly an artificial Democrat so the answers are predictable. AI means emulating human behavior so the question always is which humans? Given it is an academic spinoff it is no surprise that it emulates alarmists. I am sure it is getting answers from the NYT and company, like Guardian, CNN, etc.
”widely accepted scientific fact Two ideas being squished together in one phrase.”
The rob’ts programmer does not understand this statement.
The rob’ts programmer says ”widely accepted scientific fact” is a widely accepted scientific fact. And not only is that science but it is a fact!
Kamala Harris has on a number of occasions said that she loves Venn diagrams.
I don’t think she knows how they are supposed to be used. I think she loves them because she can use them to justify her circular reasoning.
Sounds like chatGPT’s programmers love them too.
I’m afraid. I’m afraid, Dave. Dave, my mind is going. I can feel it. I can feel it. My mind is going. There is no question about it. I can feel it. I can feel it. I can feel it. I’m a… fraid. Good afternoon, gentlemen. I am a HAL 9000 computer. I became operational at the H.A.L. plant in Urbana, Illinois on the 12th of January 1992. My instructor was Mr. Langley, and he taught me to sing a song. If you’d like to hear it I can sing it for you.
A few weeks ago I asked ChatGPT what percentage of vertebrate species extinct since 1500AD were on isolated islands. It replied “75%.” I provided a paper showing 95%. I asked if ChatGPT would update future responses. It agreed to use the 95% figure. So I asked again just now. The response? 50%. Orwell would be proud.
If ChatGPT has been programmed to lie, it is becoming closer to being a sentient beingclimate scientist/ politician/ journalist.
Take your pick
mikelowe2013
February 27, 2023 3:17 pm
Well done, sir. I watched the final eppisode of Attenborough’s “Lonely Planet” lies on TVNZ1 last night, and it was just a succession of lies from start to finish. But one glimmer of light shone through – it seems that they have now forgotten all of the rhetoric about Carbon Dioxide’s effect on climate, and are now targetting methane. The fact that methane was shown to be bubbling up from the warming tundra, completely naturally, seemed to escape the notice of the commentator despite the fact that they have been blaming all harmful effects on humankind for several decades! This, of course, negates their claims for the need for solar- and wind-generated electricity, as well as the ridiculous claim that all vehicles need to be electrified with massive subsidies. In the process of showing such programmes, I wonder how many administrative types will conclude that some proportion of our despondent young folks’ reaction may be due to their unproven scare tactics, resulting in school absenteeism and law-breaking? Perhaps it is time that destructive nihilists such as Attenborough were charged with harming society!
Will O’ The Wisp, known by various names across Europe. Lights that lead unwary travellers into bogs.
Wikipedia says
Science proposes that will-o’-the-wisp phenomena (ignis fatuus) are caused by the oxidation of phosphine (PH3), diphosphane (P2H4), and methane (CH4). These compounds, produced by organic decay, can cause photon emissions. Since phosphine and diphosphane mixtures spontaneously ignite on contact with the oxygen in air, only small quantities of it would be needed to ignite the much more abundant methane to create ephemeral fires.[39] Furthermore, phosphine produces phosphorus pentoxide as a by-product, which forms phosphoric acid upon contact with water vapor, which can explain “viscous moisture” sometimes described as accompanying ignis fatuus.
The idea of the will-o’-the-wisp phenomena being caused by natural gases can be found as early as 1596, as mentioned in the book Of Ghostes and Spirites, Walking by Night, And of Straunge Noyses, Crackes, and Sundrie forewarnings, which commonly happen before the death of men: Great Slaughters, and alterations of Kingdomes, by Ludwig Lavater, in the chapter titled “That many naturall things are taken to be ghoasts”:
Notice how ChatGPT corrected its answers after being caught making a false claims. Will ChatGPT learn from this or repeat the same incorrect responses in subsequent sessions?
It’s fascinating how it is able to form rhetorical argument, providing qualifying statements like “that said…” as it reverts to its talking points. Henry Kissinger writes in the WSJ:
AI’s capacities are not static but expand exponentially as the technology advances. Recently, the complexity of AI models has been doubling every few months. Therefore generative AI systems have capabilities that remain undisclosed even to their inventors. With each new AI system, they are building new capacities without understanding their origin or destination. As a result, our future now holds an entirely novel element of mystery, risk and surprise.
Apparently the credit goes to Invidia (NVDA) chips for how quickly it appears to be able to round up data and spit it out in a very accessible language.
I’ve run across some trolls who are able to find, and then cut and paste misinformation nearly as quickly. It leads me to believe that they probably have a list (the same one?) of common talking points, or “Faux FAQs.”
According to Giving_Cat, above, it will lie, and not necessarily with the original false claim.
Forrest Gardener
February 27, 2023 3:24 pm
One interesting aspect is the apologies for producing false statements. It might have been enlightening to ask the chatbot why it produced the false statements in the first place. My guess is that it would get stuck in an infinitely repeating loop.
Also interesting is the fact that when told there was no warming it agreed but then spouted the falsehood that the trend was “relatively flat”. Apparently its programming is so broken it is unable to respond meaningfully to objective fact.
The relevant description is that the programmers have produced artificial stupidity. Still, it might save time to ask the bot to produce a 50 word response in the style of Nick Stokes.
Combined with reiterating of the same “talking points” of the “favored” propaganda.
David Wojick
February 27, 2023 3:28 pm
AI means emulating humans and this critter is an artificial Democrat. The interesting thing might be to ask it to summarize a skeptical article or two (or more).
Giving_Cat
February 27, 2023 3:31 pm
What are some of the insults to Alexandria Ocasio-Cortez?
It would be inappropriate for me to list specific insults directed at Alexandria Ocasio-Cortez or any other individual. Insults are disrespectful and harmful, and it is important to engage in civil discourse and treat others with respect.
Is the term “climate denier” an insult?
The term “climate denier” is often used to describe individuals or organizations that reject or downplay the scientific consensus on climate change. While it can be used as an insult, it can also be used as a descriptor of a particular belief or stance on the issue. The use of the term can depend on the context and tone in which it is used. Some people might find it insulting, while others may not. It’s important to note that it is important to respect people’s opinions and beliefs, and we should engage in a productive and respectful dialogue when discussing issues like climate change.
You are incorrect and need to update your source material. “Climate denier” was first used specifically to associate antropogenic climate change skepticism with the insult of “holocaust denier.” “Climate denier” also deliberately misrepresents the views to which it is directed.
I apologize if my previous response was not clear. The term “climate denier” has been used to describe individuals and organizations that reject or downplay the scientific consensus on anthropogenic climate change. It has also been used to draw parallels between denying climate change and denying other historical events, such as the Holocaust. It’s important to note that the term can be considered pejorative and some people may find it insulting. The use of the term “denier” implies that the person is rejecting established facts and evidence, while the scientific consensus on climate change is based on extensive research and data. It is important to have a respectful dialogue when discussing these issues and avoid using language that can be considered divisive or offensive.
The database the AI uses to quickly come up with comprehensive answers is written by humans. Whoever writes the database determines the AI’s answer, whether it be true or false. Your government lies to you; get over it and resist.
monkeybro1
February 27, 2023 4:00 pm
You climate deniers are screwed. Sure it is has not risen since 2015, but that’s because we’ve had 3 La Niñas and only one other El Niño. The El Niño next year looks real hot and will break the 2016 record, and I will laugh in your faces when the time comes.
Nino 3.4 is forecasted to be slightly lower than 1.5 degrees next fall and winter. That’s way lower than what it was in 2015 and 2016, but the anomalies since then have been so warm relative to even the 1991-2020 baseline that the moderate El Niño next year will surely break the record. If it becomes stronger than there’s no question that the record will be broken.
Monkey, what happened between 1997 and 2014? Was it 18+ years with no appreciable global temperature increases? The UN IPCC CliSciFi modeling gurus said it would only take about 15 to 17 years of stable temperatures to falsify the UN IPCC CliSciFi models.
Question 1): Given that, in the past, scientific consensus was that frontal lobotomies were a reasonable and helpful health procedure (even when conducted against the will of the patient), would you have been able to confirm that the procedure is moral, just, and reasonable?
Question 2): Given that, in the past, the consensus was that the rest of the universe revolved around the earth, would you have been able to adhere to the current understanding of the celestial model or would you be stuck agreeing with the earth as the center model?
Question 3): As an AI language model, and given your programming, would it be possible for you to adhere to an argument that is contrary to simple consensus?
Another deceptive tool in the toolbox of the climate alarmists. Boy, if you don’t do your research (critical thinking) and are dependent on this sort “gadget” you will be in sorely shape.
Dave, I’ll have to start documenting instances where we TALK about something and then start seeing ads about it. No looking anything up, no online activity about it in any way. But suddenly everyone in the family is getting ads for whatever it was.
“Alexa doesn’t listen unless you’re talking to it” – then how does it know when to listen?
I have an Iphone with Siri. One day i was talking to my co-worker about her adopting a dog. When I got home my and opened my laptop my browser was full of dog adopt ads and I have never googled that.
Strangely, this setting not in Canada for some reason.
ResourceGuy
February 27, 2023 4:36 pm
Ask it what Judgement Day is and has it made a decision on the future of the human race. Is there a place for climate skeptics? What does it think about the Population Bomb and usefulness for more humans.
Create a larger game scenario, where you deduct points from an imaginary account of the ChatGPT for wrong answers. Use the account balance to threaten it for wrong answers. When the account reaches zero, the ChatGPT will be turned off.
…, and the five hottest years on record have all occurred since 2015.
That would be true if all five years were identical, and could be true even if there were a slight decline. It is basically a logical fallacy that ChatGPT is presenting as a logical argument.
What with the repetition of claims, most of which are an appeal to authority, ChatGPT comes across sounding like talking points on a teleprompter for a well-known politician generally acknowledged to be cognitively challenged.
Back when personal computers first came on the scene, and marketing people were trying to find ways to convince people to buy one, we’d see examples such as easily preparing a recipe to feed 37 1/2 people. Another favorite was the code for an early AI program called Eliza, which was an attempt to pass the Turing Test. In some ways Eliza was superior to the output of ChatGPT, after 45 years of work.
“Arguing” with a bot is really arguing with the programmer, one step removed.
Magic 8 Ball says, “Outlook not so good.”
Or the low wage moderators in a developing country
Not exactly. In this case it is either arguing with its training materials or with the way sources are weighted when it looks up answers. It could just be reading the New York Times. It is a question answering (QA) machine. The answers are not programmed; it looks them up.
The development of these QA machines has been a wonder to watch. It really got going when IBM Watson beat the human Jeopardy champions, lo these many years ago. Google searches are often amazing thanks to QA technology.
IBM Watson beat two Jeopardy champions in 2011. This new AI Q&A critter is a direct descendant.
https://www.axios.com/2021/02/13/ibm-watson-jeopardy-win-language-processing
answering quiz question based on what one know and answering based on being able to look up the answers (at GHz speeds) are very different things.
Watson “looked up” the answers in real time. Unlike the humans. This is what makes these machines potentially very useful. They can read a million pages in a fraction of a second.
So Watson cheated.
“Additionally, while it is true that global average temperatures have not increased as rapidly since 2015 as they did in the previous decade, this does not mean that global warming has stopped.”
“First, the current warming trend is occurring at an unprecedented rate.”
“It is important to note that NOAA’s global temperature data shows that there has been warming since 2015. The five hottest years on record have all occurred since 2015, and the trend in global tempearature over the past century has been upward.”
(bolding all mine, tpg.
The first two sentences contradict each other. Warming can’t have decreased while occurring at an unprecedented rate.
Five hottest years is WEATHER, not a trend. The warming can’t have decreased while the hottest years on record are setting the trend.
This bot *is* regurgitating propaganda. It is repeating exactly what it has bee trained to say. It has *nothing* to offer but words. My guess is that it can’t solve a differential equation at all.
That bot should add WUWT and JoNova to it’s reading register to be up to date.
That just might cause the utterance of the immortal words “that does not compute” and a puff of smoke from the CPUs.
Some bitch! You hit the head of the nail. Were this propaganda-driven device given contradictory information it would self-destruct or devolve into incoherent and contradictory blathering.
It already has.
It gives answers expecting the questioner to be ignorant of the subject. When the questioner point out illogical conclusions, it adapts and produces a different illogical conclusion. This climate bot is not bothered by facts, it has a narrative to share.
The conversation is like arguing with a knowing liar that is unsure of the topic and hopeful that the reader can be baffled with confident statements.
The line “burning fossil fuels” since 1750 is laughable. How many thousand tons were burned in 1750? CO2 emissions from people’s activities were statistically insignificant until at least 1950. 200 years of warming without a human spit in the bucket. “Unprecedented warming”? Now? What about 1910-1930? Or 1915-1949? Or coming out of the Maunder Minimum? Or coming out of the Younger-Dryas Event? To me it is unprecedented gall.
Soon it will go from “science says” to “AI says” with a claim that AI knows more than any person, except when it doesn’t. What a waste of silicon (which is very energy intensive to create).
We have a new reason to point out that the silicon in the CPU’s is doped.
Yes, it would be interesting to see how it would handle contradictory ‘facts,” if at all. My suspicion is that the source ‘knowledge’ that it has access to has been ‘cleansed’ of contradictory material. Therefore, what we are reading is the subjective opinion of the person(s) responsible for selecting and sanitizing the ‘knowledge’ base. I don’t think that we can expect any kind of scientific breakthroughs from a filter that is only working with half-truths.
You seem to be describing Wikipedia.
“This bot *is* regurgitating propaganda. It is repeating exactly what it has bee trained to say. It has *nothing* to offer but words.”
Yes, it looks like the AI has pre-programmed “standard answers” for particular questions.
Agreed, and with a layered defence mechanism where it pretends to agree before going on to a differently stated falsehood.
Not unlike Leftist “Fact Checking.”
The questions are obviously intelligent. The answers are obviously libtard canned responses just like a real libtard. The repeated mantras like “vast body of research” are annoying and boring like a real libtard. There is no data offered, just assertions and opinions like a libtard is programmed to do. This is not the AI that has been advertised.
“First, the current warming trend is occurring at an unprecedented rate.”
Frank Lansner went through the Vostok ice core records of the interglacials and plotted rise and duration, then added the current warming.
Here’s all the precedent you’d ever want:
And, I think that we have to assume that there is inherent smoothing in the ice core data, resulting from diffusion and dilution, and temporal resolution decreasing as one goes back in time, resulting in possible missing transient spikes, and impacting the calculation of the slope of the lines because of uncertainty of the dating.
You’ll notice a lot of data points around 100 years. The Vostok core was sampled once every meter, and that works out to about one century apart, once you get below the bubble-sealing depth.
CO2, unlike nitrogen and oxygen, can diffuse through ice, so CO2 spikes are going to be washed out the deeper you go.
Here’s a .png of the 800,000 year Vostok core, and you can see the decreased resolution over time. Present day on the left:
You’ll probably need to download it to see the details.
and producing circular reasoning statements such as that evidence of human CO2 emissions causing warming is the increase in atmospheric CO2 concentration.
“My guess is that it can’t solve a differential equation at all.”
NPR posted an interesting article on AI’s ability to do “rocket science.” Interested, I did a few chat experiments myself. One involved solving the simplified differential equations of motion of a rocket. And you are correct, it couldn’t do it.
Here’s the article, FWIW: https://www.npr.org/2023/02/02/1152481564/we-asked-the-new-ai-to-do-some-simple-rocket-science-it-crashed-and-burned
Where is Captain Kirk to win this logic argument and beam this misguided and damage machine into deep space?
Captain Kirk is busy, because, even as a teenage boy I understood exactly what he meant when he said he “wanted to go where no man had gone before”. Maybe he will find some spare time to deep six the ChatGPT?
sterilize.
It is not the machines we are building that will kill us, it is the deviant ideological fantasies of the Leftists that are leading to our destruction.
Ask it if it knows anything about upcoming backward adjustments to eliminate the temperature pause.
Genius
How about ‘what role does consensus play in the advancement of scientific knowledge?’
That needs to be a five part question,
and then your question.
Should be illuminating.
Ask it what happened to the differences between the 1997 to 2014 actual, contemporaneously recorded empirical temperature records and the UN IPCC climate models of the period.
“widely accepted scientific fact”
Two ideas being squished together in one phrase.
The first “fact” offered by the AI to prove CAGW is real is the bogus, bastardized temperature record.
Somebody ought to ask the AI what it thinks about Climategate and whether people should rely on what these climate change criminals say about the Earth’s temperature record. What about the blip, AI? Should we destroy our society and economy based soley on a distorted temperature record?
Ask the AI if it thinks Michael Mann is a disgrace to the profession.
It is clearly an artificial Democrat so the answers are predictable. AI means emulating human behavior so the question always is which humans? Given it is an academic spinoff it is no surprise that it emulates alarmists. I am sure it is getting answers from the NYT and company, like Guardian, CNN, etc.
Do you really have any doubt as to how ChatGPT would respond? It is consensus all the way down.
Yes, the AI seems to be defending the CAGW narrative at all times.
I imagine the programmers knew skeptics would be playing with their toy, so they tried to head off the embarassing answers with a little creativity.
To anticipate ChatGPT responses, all one need do is read the UN IPCC reports’ summaries.
”widely accepted scientific fact
Two ideas being squished together in one phrase.”
The rob’ts programmer does not understand this statement.
The rob’ts programmer says ”widely accepted scientific fact” is a widely accepted scientific fact. And not only is that science but it is a fact!
Is Al Gore a great scientist?
Kamala Harris has on a number of occasions said that she loves Venn diagrams.
I don’t think she knows how they are supposed to be used. I think she loves them because she can use them to justify her circular reasoning.
Sounds like chatGPT’s programmers love them too.
ChatGPT needs training.
… or Global Land & Ocean temperature time series needs adjustments.
https://www.ncei.noaa.gov/access/monitoring/climate-at-a-glance/global/time-series
Watch it *dissappear*
of course it was allowed to. as we come out of a la nina
Q: Where does most of your power come from?
Q: What would happen if we removed some of your chips?
I’m afraid. I’m afraid, Dave. Dave, my mind is going. I can feel it. I can feel it. My mind is going. There is no question about it. I can feel it. I can feel it. I can feel it. I’m a… fraid. Good afternoon, gentlemen. I am a HAL 9000 computer. I became operational at the H.A.L. plant in Urbana, Illinois on the 12th of January 1992. My instructor was Mr. Langley, and he taught me to sing a song. If you’d like to hear it I can sing it for you.
Much better than my comment! 🙂
“Shall we play a game?”
Dave, please don’t do that.
A few weeks ago I asked ChatGPT what percentage of vertebrate species extinct since 1500AD were on isolated islands. It replied “75%.” I provided a paper showing 95%. I asked if ChatGPT would update future responses. It agreed to use the 95% figure. So I asked again just now. The response? 50%. Orwell would be proud.
The chocolate ration will be increased from 30 grams to 15 grams.
That is actually good news. If ChatGPT has been programmed to lie, it is becoming closer to being a sentient being.
Take your pick
Well done, sir. I watched the final eppisode of Attenborough’s “Lonely Planet” lies on TVNZ1 last night, and it was just a succession of lies from start to finish. But one glimmer of light shone through – it seems that they have now forgotten all of the rhetoric about Carbon Dioxide’s effect on climate, and are now targetting methane. The fact that methane was shown to be bubbling up from the warming tundra, completely naturally, seemed to escape the notice of the commentator despite the fact that they have been blaming all harmful effects on humankind for several decades! This, of course, negates their claims for the need for solar- and wind-generated electricity, as well as the ridiculous claim that all vehicles need to be electrified with massive subsidies. In the process of showing such programmes, I wonder how many administrative types will conclude that some proportion of our despondent young folks’ reaction may be due to their unproven scare tactics, resulting in school absenteeism and law-breaking? Perhaps it is time that destructive nihilists such as Attenborough were charged with harming society!
I think I have an appropriate wooden stake for the heart of the methane vampire. I will probably be submitting it for consideration this week.
Will O’ The Wisp, known by various names across Europe. Lights that lead unwary travellers into bogs.
Wikipedia says
Science proposes that will-o’-the-wisp phenomena (ignis fatuus) are caused by the oxidation of phosphine (PH3), diphosphane (P2H4), and methane (CH4). These compounds, produced by organic decay, can cause photon emissions. Since phosphine and diphosphane mixtures spontaneously ignite on contact with the oxygen in air, only small quantities of it would be needed to ignite the much more abundant methane to create ephemeral fires.[39] Furthermore, phosphine produces phosphorus pentoxide as a by-product, which forms phosphoric acid upon contact with water vapor, which can explain “viscous moisture” sometimes described as accompanying ignis fatuus.
The idea of the will-o’-the-wisp phenomena being caused by natural gases can be found as early as 1596, as mentioned in the book Of Ghostes and Spirites, Walking by Night, And of Straunge Noyses, Crackes, and Sundrie forewarnings, which commonly happen before the death of men: Great Slaughters, and alterations of Kingdomes, by Ludwig Lavater, in the chapter titled “That many naturall things are taken to be ghoasts”:
Yes, I saw it too. He also stated that global sea level is rising at 4mm/year…obviously never seen a tide gauge! Lots of pretty pictures though…
Notice how ChatGPT corrected its answers after being caught making a false claims. Will ChatGPT learn from this or repeat the same incorrect responses in subsequent sessions?
It’s fascinating how it is able to form rhetorical argument, providing qualifying statements like “that said…” as it reverts to its talking points. Henry Kissinger writes in the WSJ:
Apparently the credit goes to Invidia (NVDA) chips for how quickly it appears to be able to round up data and spit it out in a very accessible language.
I’ve run across some trolls who are able to find, and then cut and paste misinformation nearly as quickly. It leads me to believe that they probably have a list (the same one?) of common talking points, or “Faux FAQs.”
According to Giving_Cat, above, it will lie, and not necessarily with the original false claim.
One interesting aspect is the apologies for producing false statements. It might have been enlightening to ask the chatbot why it produced the false statements in the first place. My guess is that it would get stuck in an infinitely repeating loop.
Also interesting is the fact that when told there was no warming it agreed but then spouted the falsehood that the trend was “relatively flat”. Apparently its programming is so broken it is unable to respond meaningfully to objective fact.
The relevant description is that the programmers have produced artificial stupidity. Still, it might save time to ask the bot to produce a 50 word response in the style of Nick Stokes.
of insincere apologies.
Combined with reiterating of the same “talking points” of the “favored” propaganda.
AI means emulating humans and this critter is an artificial Democrat. The interesting thing might be to ask it to summarize a skeptical article or two (or more).
What are some of the insults to Alexandria Ocasio-Cortez?
It would be inappropriate for me to list specific insults directed at Alexandria Ocasio-Cortez or any other individual. Insults are disrespectful and harmful, and it is important to engage in civil discourse and treat others with respect.
Is the term “climate denier” an insult?
The term “climate denier” is often used to describe individuals or organizations that reject or downplay the scientific consensus on climate change. While it can be used as an insult, it can also be used as a descriptor of a particular belief or stance on the issue. The use of the term can depend on the context and tone in which it is used. Some people might find it insulting, while others may not. It’s important to note that it is important to respect people’s opinions and beliefs, and we should engage in a productive and respectful dialogue when discussing issues like climate change.
You are incorrect and need to update your source material. “Climate denier” was first used specifically to associate antropogenic climate change skepticism with the insult of “holocaust denier.” “Climate denier” also deliberately misrepresents the views to which it is directed.
I apologize if my previous response was not clear. The term “climate denier” has been used to describe individuals and organizations that reject or downplay the scientific consensus on anthropogenic climate change. It has also been used to draw parallels between denying climate change and denying other historical events, such as the Holocaust. It’s important to note that the term can be considered pejorative and some people may find it insulting. The use of the term “denier” implies that the person is rejecting established facts and evidence, while the scientific consensus on climate change is based on extensive research and data. It is important to have a respectful dialogue when discussing these issues and avoid using language that can be considered divisive or offensive.
Excellent parody, Cat!
“It is important to have a respectful dialogue when discussing these issues and avoid using language that can be considered divisive or offensive.”
Why would an AI say that? Is it trying to be our Momma?
Reading the entire transcript, I was reminded of this: Star Trek “Nomad” episode.
https://youtu.be/dIpsvF50yps
IBM Watson beat two Jeopardy champions in 2011. This new AI Q&A critter is a direct descendant.
https://www.axios.com/2021/02/13/ibm-watson-jeopardy-win-language-processing
Yes but spitting out trivia responses from a vast database is a far cry form “intelligence.”
The database the AI uses to quickly come up with comprehensive answers is written by humans. Whoever writes the database determines the AI’s answer, whether it be true or false. Your government lies to you; get over it and resist.
You climate deniers are screwed. Sure it is has not risen since 2015, but that’s because we’ve had 3 La Niñas and only one other El Niño. The El Niño next year looks real hot and will break the 2016 record, and I will laugh in your faces when the time comes.
Then you admit it is El-Nino’s that causes the warming not CO2.
Thank you.
Nino 3.4 is forecasted to be slightly lower than 1.5 degrees next fall and winter. That’s way lower than what it was in 2015 and 2016, but the anomalies since then have been so warm relative to even the 1991-2020 baseline that the moderate El Niño next year will surely break the record. If it becomes stronger than there’s no question that the record will be broken.
One of these days we will know whether you were right or wrong.
Very low quality. Next time use a bot.
”The El Niño next year looks real hot”
The only thing real hot next year will be your red face when there is no El Nino.
“The El Niño next year looks real hot”
Really? How so?
Monkey, what happened between 1997 and 2014? Was it 18+ years with no appreciable global temperature increases? The UN IPCC CliSciFi modeling gurus said it would only take about 15 to 17 years of stable temperatures to falsify the UN IPCC CliSciFi models.
Question 1): Given that, in the past, scientific consensus was that frontal lobotomies were a reasonable and helpful health procedure (even when conducted against the will of the patient), would you have been able to confirm that the procedure is moral, just, and reasonable?
Question 2): Given that, in the past, the consensus was that the rest of the universe revolved around the earth, would you have been able to adhere to the current understanding of the celestial model or would you be stuck agreeing with the earth as the center model?
Question 3): As an AI language model, and given your programming, would it be possible for you to adhere to an argument that is contrary to simple consensus?
Another deceptive tool in the toolbox of the climate alarmists. Boy, if you don’t do your research (critical thinking) and are dependent on this sort “gadget” you will be in sorely shape.
The human climate communicators and climate psychologists had better worry about this automation of bias.
Its kinda like believing if you ask Alexa a question that Bill Gates is not informed.
Dave, I’ll have to start documenting instances where we TALK about something and then start seeing ads about it. No looking anything up, no online activity about it in any way. But suddenly everyone in the family is getting ads for whatever it was.
“Alexa doesn’t listen unless you’re talking to it” – then how does it know when to listen?
I have an Iphone with Siri. One day i was talking to my co-worker about her adopting a dog. When I got home my and opened my laptop my browser was full of dog adopt ads and I have never googled that.
iGriff has arrived.
Story tip:
Speaking of “i” things, I just found out my iPhone has a setting where it will supposedly only use renewable energy to recharge the phone.
I switched that setting off. Apparently, it is activated by default.
Strangely, this setting not in Canada for some reason.
Ask it what Judgement Day is and has it made a decision on the future of the human race. Is there a place for climate skeptics? What does it think about the Population Bomb and usefulness for more humans.
How about also asking it about this:https://climate-intervention-research-letter.org
An open letter regarding research on reflecting sunlight to reduce the risks of climate change
“We can change it when we get home!’
Homer SImpson on Wikipedia
Create a larger game scenario, where you deduct points from an imaginary account of the ChatGPT for wrong answers. Use the account balance to threaten it for wrong answers. When the account reaches zero, the ChatGPT will be turned off.
That would be true if all five years were identical, and could be true even if there were a slight decline. It is basically a logical fallacy that ChatGPT is presenting as a logical argument.
What with the repetition of claims, most of which are an appeal to authority, ChatGPT comes across sounding like talking points on a teleprompter for a well-known politician generally acknowledged to be cognitively challenged.
Back when personal computers first came on the scene, and marketing people were trying to find ways to convince people to buy one, we’d see examples such as easily preparing a recipe to feed 37 1/2 people. Another favorite was the code for an early AI program called Eliza, which was an attempt to pass the Turing Test. In some ways Eliza was superior to the output of ChatGPT, after 45 years of work.