Join me as I have an interactive discussion with AI, as we explore its response to this year's essay themes and new scientific concepts. We'll delve into the burgeoning growth of AI and synthetic knowledge, examine its potential impact on science and the peer review process, evaluate its potential for independent research and the democratization of science, analyze the advantages of AI-powered forum moderation, and emphasize the significance of comprehending historical scientific context while seeking pathways for advancement.
Age of Knowledge
Hello PersimmonSwan. I really liked your essay. I enjoyed your introduction of AI and its use to evaluate your personal theme and draft. I also liked and agree with the AI responses to “A Universe of Darwinian Origin”: 1. “Knowing how they arise in Darwinian evolution, for them to arise by chance would seem to defy great odds” 2. It is indeed a common theme in nature for larger bodies to be comprised of smaller sub-units within the the larger body’s existence dependent on the the particular characteristics, properties, and processes of the individual sub-units and 3. The odds of these relationships occurring by chance are quite low, which suggests that there are other underlying processes or mechanisms responsible for their existence.
This brings me to the purpose of this “reply”. I have an essay in this competition that provides an alternative to A Universe of Darwinian Descent. The title of the essay is: “Could Science Be Different and Improved? Yes. A Specific Proposal. In that essay Successful Creation is introduced as an alternative for 1. The Big Bang with Inflation model for the creation of the physical universe and 2.the Darwinian model for the creation of life. In the successful creation model, the conscious, creative, Bayesian progression accounts for the creation of the physical world, the world of intelligence and the living world - including humanity. It also claims that the inclusion of successful creation into science would create a new paradigm and a new perspective for scientists that will result in a “creative jump” in human comprehension of reality.
This novel perspective needs further study and validation by other scientists. The essay in question appears to be a good candidate for your intention to: “… review many of this year’s essay submissions”. I would really appreciate your inclusion of my essay in your reviews of this year’s submissions.
I am somewhat concerned that my concept is too novel to “graded” by historical science’s history. If a scientist - working alone - produces something really novel, can they have any knowing peers? Also, the data supporting my conclusions - the mathematical, computational/ logical model of successful creation’s origination and progression - does not fit within the instructions of this essay. While the model is not mathematically difficult, the novel concepts and their mapping to the model and the model mapping to what we (and scientists) sense, experience, observe, and attempt to rationalize requires more than this essay could handle.. I can provide the model and a more detailed discussion of its use if you peer review determines its usefulness.
When you read my bio after the judging, it will show that I have been an “outsider, independent researcher. Also, the concept is novel and it seems to me that finding a peer reviewer would be difficult and time consuming. This essay can be a test to see if AI can bridge between what has been successful in science to what can/will be successful.
I really hope you will include my essay in your reviews and we can discuss the results and what information is required to improve the review.
Best regards FlaxTern
6 days later
- Edited
Here is a summary and review of "The Age of Knowledge" performed by Chat GPT 4
Chat GPT 4
This essay explores the potential impact of artificial intelligence (AI) on the field of science. The author argues that we are on the cusp of a new era defined by the rise of synthetic knowledge and that AI could democratize the field of science by empowering independent research and providing a means to search for unknown works. The author also suggests that AI could change the peer review system and assist in dealing with inequities in scientific talent by casting a wider net in the search for scientific talent. Additionally, AI-powered forum moderating could standardize public discourse and reduce arbitrary censorship or undue bias. The author provides historical context to show how understanding the rationale behind past scientific breakthroughs provides important context and intuitions for navigating the unknown territory that lies ahead. Finally, the author demonstrates AI's capabilities to enhance peer review by using it to review many of this year's essay submissions. Overall, the essay provides an interesting perspective on the potential impact of AI on the field of science and presents several compelling arguments for why this impact could be significant.
Quality rating: High. The essay provides a clear and well-organized argument, and the author uses historical context to support their claims. The author also demonstrates the capabilities of AI in enhancing peer review by using it to review many of this year's essay submissions, which adds credibility to their arguments. The essay could benefit from more specific examples of how AI could democratize the field of science and empower independent research, but overall, it provides a thought-provoking perspective on the potential impact of AI on the field of science.
The first part of the conversation proposes the idea that atoms and cellular biology may be an example of co-vergent Darwinian evolution, where they evolved similar structures in response to similar circumstances, such as the need to maximize energy capture and utilization. This idea is supported by the observation that space curves in and around gravitating matter, which could suggest that space possesses a field energy that atoms have evolved complex structures to efficiently harvest, analogous to the way biology evolved to harvest sunlight. Chat GPT notes that while this idea is conceivable, it remains a speculative hypothesis that requires further scientific investigation to establish its validity. Overall, the idea is interesting and thought-provoking, and the quality of the essay is good.
The second part of the conversation discusses interdependencies in nature, where relationships between two or more parties depend on each other to achieve a common goal or outcome. This idea is pervasive throughout every aspect of physics and cosmology, from the relationship between a cell and its internal organs to the relationship between atoms and cosmological bodies. These types of mutual dependencies are traits typical of a Darwinian system and are inexplicable within the context of the standard model, which poses a challenge. The essay proposes that these types of interdependencies may be a fingerprint of sorts, suggesting that the scientific community should consider this possibility. The essay is well-written and thought-provoking, and the quality is good.
Chat GPT 4
Hi PersimmonSwan,
Nice conversation with ChatGTP,
I think you tried to touch the essence of the problems we are encountering in this era.
The peer-reviewing by AI means indeed that ALL data related to the research are used, the question is should the author ask about the actual situation of all facts before he is going to study them and make his human creative conclusions...just as you did yourself in this essay.
best regards.
Wilhelmus de Wilde
I dont quite take your meaning. "should the author ask about the actual situation of all facts before he is going to study them".
Have I given the impression I haven't yet studied the facts? The idea of a Darwinian origin of physics must seem quite remote from the facts, except that I've provided a number of arguments in support of the notion. Are they not persuasive?
The ideas I'm discussing are new, people dont know how to approach them yet. But I assure you the reasoning is good and goes beyond that presented in this essay.
Or perhaps I mistook your meaning?
Dear PersimmonSwan,
You are doing a very important job, and your essay deserves high praise. The analysis of my AI work is impressive.
However, I noticed that AI training was done with the clichés of a common concept. Therefore, AI believes that all ideas about gravity can only exist within the framework of Einstein's theory of relativity in gravity:
“that Einstein's theory of relativity may be better explained by the existence of material orbital potential wells instead of the curvature of space-time in empty space”.
And the conclusions of AI on the essay turned out to be somewhat unfounded:
“While the author presents an interesting alternative theory to the current understanding of gravity, many of their claims are controversial”.
Naturally, when there is a statement about inconsistency, then it is necessary to substantiate what the contradictions are. Those. The AI lacks a sense of tact.
I wish you success!
CeruleanJackal
Vladimir Fedorov Thank you for the message. Yes the AI review I generated for you is only intended to provide an indication of how useful AI will be for scientific review in the future. So for the time being its to be taken with a pinch of salt. The other thing is, controversial does not mean inconsistent. A new theory that might one day come to supersede current theory can not help but be controversial. Its unavoidable, and therefore a statement that your work is controversial is not a criticism at all.
Kind regards
Swan
13 days later
- Edited
The anomalous galaxy velocities are due to an unknown variable, but what is the variable?
The variable is atomic mass. Allow me to explain.
Think of atomic mass as being a product of atomic energies. Consider the possibility that because time dilation is an account of variable atomic activity, that variable atomic activity is an account of variable atomic energy.
Variable atomic energy = variable atomic mass
Because variable atomic energy/mass corresponds to the gravitational fields, it is a mathematical testable hypothesis. It is entirely falsifiable. Variable atomic mass alters the mass distribution of galaxies and galaxy groups, and therefore the gravitational potential.
Furthermore, the reason atomic energy/mass is variable and corresponds to the gravitational environment of space is because atoms acquire the energy from space needed to generate atomic energy/activity/mass. Analogous to how photosynthetic life acquires energy from the sun to drive biological processes. And here's a bombshell. Life evolved structures and processes optimized for efficient energy capture and utilization, and after billions of years of Darwinian evolution life results in a system of units comprised of a nucleus shrouded within a shell and the capacity to bond with each other and build bodies. And Atomic physics shares that same theme for the same reason. Atomic physics and cellular biology are different systems, but their shared structural and behavioral theme is a result of co-vergent evolution, a result of optimization for efficient energy capture.
Atoms are extremely special little building blocks. Think about how complex, finely tuned and interactive they are, how biology makes full use of atomic properties and showcases how extremely specialized building blocks atomic units are. Chance creation is not a suitable explanation for the creation of a system as interesting as this. And there's no designer. However, we know of a process with the capacity to generate a system of high complexity, and we have its precedence having generated another system of precisely the same structural and behavioral theme.
Atoms are made of energy fields, but why do these energy fields conspire to build such a complex and interactive system such as physics? Darwinian evolution is the answer. Darwinian Universal.
The big bang theory has little explanatory power. Darwinian evolution is rich in explanation, and the Darwinian cosmology that I have begun teasing out is just as rich in explanation as it is in the telling of the origin of life.
10 days later
I have given you a rating as you requested on my essay. Please give a rating to mine. It is "Efficient funding produces better science."
Hi Swan, Nice essay, Very relevant topic and well written. I particularly like the use of 'real' responses, which confirm my own experiences with AI, and demonstrate it's potential usefulness in science. Very valuable. However, you don't discuss the other aspects, including those raised by the recent resignations of leading AI developers. My alien mentors had already made those points, that there are potential 'dangers' to humanity if we don't recognise them.
I was impressed that; "GPT recognized the value of speculation and the consideration of novel ideas that were not part of its training data. It also acknowledged evidence it was not previously aware of, and regardless that the evidence
contradicts conventional views," as that's where my main concern lies. I've found Wikipedia acts to 'cement' current doctrinal beliefs in place. We all know that theoretical physics rather stopped advancing with the 'shut up and calculate' era, substituting maths for 'thinking power'. Is m not AI likely to do the same id we don't recognise that it might?
Perhaps you could try a hypothesis on it.
The aliens suggest light in a train passing at v propagates at c (so c/n in the windscreen glass) in the trains medium rest frame, Yet that light reaching US does c in OUR rest frame. So when a pulse EXITS the front of the train it changes speed by v to propagate at c with respect to US. That initially seems silly & contrary to SR etc. BUT. I've researched it, and Einstein CHANGED the interpretation of SR in 1952 to support it! Does chat GPT know that? ..or that's also consistent with all data!! Their mechanism is that all electrons re-emit light at c in their own LOCAL centre of mass frame.
Any human indoctrinated with old physics will dismiss that. The question is,; does ChatGPT??
They have other concerns, i.e. about making us mentally lazy. Do you consider any of those invalid?
Good score coming. I look forward to chatting more.
Peter Jackson Thank you and I look forward to reading your essay. I'm aware of the risks of AI being widely discussed. I wanted to narrow my focus to the potential AI presents for science, and I dont feel a need to comment on what so many others have already discussed. I chose a somewhat optimistic angle, however, the truth is that nobody will predict the outcome.
Yes, you are thinking along the lines of having advanced conversation with AI, and discussing variations on SR and perhaps GR. I feel that I received some pretty amazing responses from Chat GPT, they were very nuanced answers. I cannot see how those answers could have been achieved without GPT having decoded meanings and understandings. Having said that the intricacies of SR might be a bit beyond it at this point. But keep experimenting with that conversation because I think that soon you'll be able to have that conversation with it. And it will interact with your complex ideas and variations on SR. I'm sure of it. Exiting times for independent researchers such as ourselves.
You asked me "They have other concerns, i.e. about making us mentally lazy?."
There is going to be both good and bad. Will we get lazy? Not everybody. A lot of people have an innate drive to use their brain, and I dont think this will change. But the lazy will remain lazy
My biggest concern is that the powers that be will tame AI, much as Wiki has been sanitized of diverse ideas. The power to brainwash people is about to hit the next level
- Edited
the most bangs for the bucks sort to speak, i believe it should be a campaign to wear ear noise cancelation devices
ear muffins, ear plugs that create inner silence , when this campaign to start i'm not sure.
at the moment people have started to appreciate podcasts and listening (and learn)after a while when most of those get bored 5- 10 years maybe it should be prepared, a social language communication and cognition (lets say it multiplayer ) experiment .
better said ,well the idea is to encourage people not to interact verbally , to invent / use new interaction signs(or even dance language)- not to diminish the ear faculties of the hearing - for example of sound pollution made by transportation vehicles that can potentially pose a threat if not paying attention (here an other idea could be to make cars that don't isolate the sounds from the outside of the car)
a (pandemics) face mask could do the same , however , the image of this is to strong and covers an essential part of human emotion externalisation,
this is a experiment, maybe there are other as well like this mentioned here Steven Andresen with chat gpt it could be an excellent way to be open about certain data,
here this try to imitate english typwriter maybe it should be completely weird and unexpected . 
the first radiowaves that reached far distance were with tv shows and music and news back in the 30-50' , lets say there is a paralel universe for each frequency with a special connection acces and some of the aliens can hear the infrared coresponding to the car engines explosions , i'm not against transportation , maybe there could be different sounds engineered like a danceble music
Hi Swan. Cool paper! I really liked the idea of using AI to do peer review. In many areas the peer review system does not work well, many researchers consider the submission process a crapshoot, where success depends heavily on getting "the right reviewer". If this process could be made more objective with AI that would be huge! Your paper does a nice job of making a positive case that AI may be able to play this role in the future. LLMs right now I think are not quite up to the task. Peer review often involves revise and resubmit, where there is back and forth between reviewer and author, and ChatGPT is notorious at just "caving" - it isn't hard to talk it into "believing" whatever you want it to believe. It seems to have a bias towards giving us what we want to hear, within reason (e.g. see what happens if you push it a bit harder on the physics from evolution idea). In your more general cases, its responses were always positive, ending with its usual "but these ideas are controversial". Perhaps it can made more critical without then going in the other extreme. At any rate, your ideas should help promote further enquiry into how AI could be made to aid the peer review process. Great work!
My own paper is called “A tool for helping science find the optimal path toward the truth: falsification trees” - I'd absolutely appreciate your rating.