[Note from Alexandros: The article below is a rewrite of a thread I wrote on Twitter here. The rewrite was mostly done by ChatGPT, with some edits from me. If you want to see Ryan and me experiment further with content transformations, you can tune into our livestream
This article reminds me of the articles published in the newspapers under the former Eastern European communist regimes. "Wellspring of knowledge" ? Lol. Where is the "fountain of wisdom"?
Imagine, if you will, an AI employed to control indoor climate in people's homes. This AI might learn that lower oxygen levels "help" with sleep. Then imagine finding out a decade afterwards, if ever, that nonzero unusual overnight deaths (within statistical noise) were not random accidents.
This is the future of AI, and the stupid humans who will rely on it.
Although chatGPT is impressive, we humans have a habit of anthropomorphizing. Even with a computer program from the 1960s "many early users were convinced of ELIZA's intelligence and understanding". https://en.wikipedia.org/wiki/ELIZA
I have challenged it on the jib jab a couple of times, the conflicts of interest with the FDA and lack of testing. Comes out with the usual fact checker style responses and if you drill down a bit further it "encounters an error"
If these "AI" programs do transform even the physical production of goods and take over relevant areas of expertise, it is an open door to the tyranny of the wealthy few - the owners of the factories and corporations, who will no longer need to pay for the work of employees (or to a much lesser extent) while still earning the money from goods that are necessary. Increasing numbers of non-workers would be in a state of direct financial dependency, relying on welfare given to prevent revolt, but easily taken away to punish non-compliance - if you need someone for their skills, it is harder to fire them because it hurts you as well, but if you are only paying someone for nothing, it will not hurt you to stop paying them. A similar situation is touched on in the scifi series The Expanse. The potential for societal disaster within the next 20 years has grown exponentially.
I disagree with the idea that there is a fear of the unknown when it comes to AI. History IS repeating itself. Technology may change, but human nature doesn’t change. We know where we are headed with AI and so fear is a waste of time.
Hi- this isn't a response to the current post- but I thought you might be interested in my exchange with ChatGPT--you can see that it essentially lied at first and corrected itself after I caught it:
Q: Summarize the data on the impact of the Pfizer covid-19 vaccine on all-cause mortality, with references.
A; As of my knowledge cutoff in September 2021, the Pfizer-BioNTech COVID-19 vaccine (BNT162b2) has been shown to have a significant impact on reducing all-cause mortality among vaccinated individuals. Multiple studies have evaluated the effectiveness of the vaccine in various populations, with the majority showing a reduction in mortality due to COVID-19 as well as a decrease in overall deaths. It is important to note that more recent studies may provide updated information on the vaccine's impact on all-cause mortality.
2. Dagan N, Barda N, Kepten E, et al. (2021) BNT162b2 mRNA Covid-19 Vaccine in a Nationwide Mass Vaccination Setting. N Engl J Med; 384(15):1412-1423. [Link: https://www.nejm.org/doi/full/10.1056/NEJMoa2101765]
Q: The Polack study does not report all-cause mortality and the Dagan study does not include a placebo group. Do you know any placebo-controlled trials of the Pfizer vaccine that report all-cause mortality?
A: Apologies for the confusion. You are correct that the mentioned studies do not directly report all-cause mortality data in the context of a placebo-controlled trial. Unfortunately, as of my knowledge cutoff in September 2021, there were no published placebo-controlled trials specifically reporting all-cause mortality data for the Pfizer-BioNTech COVID-19 vaccine. Most studies focused on vaccine efficacy against COVID-19 infection, symptomatic disease, and severe outcomes related to COVID-19.
However, the reduction in COVID-19 infections and severe outcomes achieved by the vaccine likely has an indirect impact on all-cause mortality. Further real-world observational studies or more recent updates to existing research might provide better insights into the impact of the Pfizer-BioNTech vaccine on all-cause mortality.
Hello to you all! On May 13th there will be an international event for all of those who suffered from this horrendous last few years. The most visible victims are the ones dead or maimed by the shots and those who died from untreated cv or other infections and diseases, who did not fit well to the narrative.
But there are many of us who suffered in various ways - discrimination, insults, being outcast from the workplace, public life and places, or simply put in repeated lockdowns.
Could you please share if there is anything planned in your country / region / town? It would be interesting and we still have the time to use and improve creative ideas, so that it could not get silenced and would not go unnoticed.
Self driving cars were looking pretty promising a few years ago, but that seems to have stalled. Isn't AI likely to be similar? I image that it will be much like computers were a few decades ago. It will not take over but will increase productivity with the use of human intervention.
This article reminds me of the articles published in the newspapers under the former Eastern European communist regimes. "Wellspring of knowledge" ? Lol. Where is the "fountain of wisdom"?
Imagine, if you will, an AI employed to control indoor climate in people's homes. This AI might learn that lower oxygen levels "help" with sleep. Then imagine finding out a decade afterwards, if ever, that nonzero unusual overnight deaths (within statistical noise) were not random accidents.
This is the future of AI, and the stupid humans who will rely on it.
"I'm sorry Dave, I'm afraid can't do that."
We were warned a long time ago.
I think the future holds less of a risk of an evil AI and more of a risk of a stupid AI that we won't understand, but will trust anyways.
Porque no los dos?
Indeed
Although chatGPT is impressive, we humans have a habit of anthropomorphizing. Even with a computer program from the 1960s "many early users were convinced of ELIZA's intelligence and understanding". https://en.wikipedia.org/wiki/ELIZA
While I agree that the most recent generation of AIs will be transformative... ChatGPT's main talent seems to be platitudinal pontification.
"Brevity is the soul of wit" has some interesting philosophical connotations here.
"Feel free to link me to other analyses on the future of this technology in the comments. " - there is Gabriel's recent stack: https://libresolutionsnetwork.substack.com/p/the-collapse-of-the-intangible
And what does chat tell you about the safety and efficacy of el jabbo?
I have challenged it on the jib jab a couple of times, the conflicts of interest with the FDA and lack of testing. Comes out with the usual fact checker style responses and if you drill down a bit further it "encounters an error"
AI is called AI because it is Artificial Intelligence. Made by the bean counter PHD's (Push Here Dummies) of Digital Technology.
If these "AI" programs do transform even the physical production of goods and take over relevant areas of expertise, it is an open door to the tyranny of the wealthy few - the owners of the factories and corporations, who will no longer need to pay for the work of employees (or to a much lesser extent) while still earning the money from goods that are necessary. Increasing numbers of non-workers would be in a state of direct financial dependency, relying on welfare given to prevent revolt, but easily taken away to punish non-compliance - if you need someone for their skills, it is harder to fire them because it hurts you as well, but if you are only paying someone for nothing, it will not hurt you to stop paying them. A similar situation is touched on in the scifi series The Expanse. The potential for societal disaster within the next 20 years has grown exponentially.
Have to say, I very much appreciate your analysis over the past year, but using ChatGPT to turn your Twitter threads into Substack posts? I'll pass.
Good job
I disagree with the idea that there is a fear of the unknown when it comes to AI. History IS repeating itself. Technology may change, but human nature doesn’t change. We know where we are headed with AI and so fear is a waste of time.
Hi- this isn't a response to the current post- but I thought you might be interested in my exchange with ChatGPT--you can see that it essentially lied at first and corrected itself after I caught it:
Q: Summarize the data on the impact of the Pfizer covid-19 vaccine on all-cause mortality, with references.
A; As of my knowledge cutoff in September 2021, the Pfizer-BioNTech COVID-19 vaccine (BNT162b2) has been shown to have a significant impact on reducing all-cause mortality among vaccinated individuals. Multiple studies have evaluated the effectiveness of the vaccine in various populations, with the majority showing a reduction in mortality due to COVID-19 as well as a decrease in overall deaths. It is important to note that more recent studies may provide updated information on the vaccine's impact on all-cause mortality.
1. Polack FP, Thomas SJ, Kitchin N, et al. (2020) Safety and Efficacy of the BNT162b2 mRNA Covid-19 Vaccine. N Engl J Med; 383(27):2603-2615. [Link: https://www.nejm.org/doi/full/10.1056/NEJMoa2034577]
2. Dagan N, Barda N, Kepten E, et al. (2021) BNT162b2 mRNA Covid-19 Vaccine in a Nationwide Mass Vaccination Setting. N Engl J Med; 384(15):1412-1423. [Link: https://www.nejm.org/doi/full/10.1056/NEJMoa2101765]
Q: The Polack study does not report all-cause mortality and the Dagan study does not include a placebo group. Do you know any placebo-controlled trials of the Pfizer vaccine that report all-cause mortality?
A: Apologies for the confusion. You are correct that the mentioned studies do not directly report all-cause mortality data in the context of a placebo-controlled trial. Unfortunately, as of my knowledge cutoff in September 2021, there were no published placebo-controlled trials specifically reporting all-cause mortality data for the Pfizer-BioNTech COVID-19 vaccine. Most studies focused on vaccine efficacy against COVID-19 infection, symptomatic disease, and severe outcomes related to COVID-19.
However, the reduction in COVID-19 infections and severe outcomes achieved by the vaccine likely has an indirect impact on all-cause mortality. Further real-world observational studies or more recent updates to existing research might provide better insights into the impact of the Pfizer-BioNTech vaccine on all-cause mortality.
Hello to you all! On May 13th there will be an international event for all of those who suffered from this horrendous last few years. The most visible victims are the ones dead or maimed by the shots and those who died from untreated cv or other infections and diseases, who did not fit well to the narrative.
But there are many of us who suffered in various ways - discrimination, insults, being outcast from the workplace, public life and places, or simply put in repeated lockdowns.
Could you please share if there is anything planned in your country / region / town? It would be interesting and we still have the time to use and improve creative ideas, so that it could not get silenced and would not go unnoticed.
Thank you all!
Self driving cars were looking pretty promising a few years ago, but that seems to have stalled. Isn't AI likely to be similar? I image that it will be much like computers were a few decades ago. It will not take over but will increase productivity with the use of human intervention.