Tornadoes, Teaching, Certifying AIs for Human Jobs, AI Interface Best Practices, $1000 Golden Banana

$1000 Golden Banana for Best AI Interface — None of the LLM “AIs” are saving conversations in globally open formats, so my conversation just now with Microsoft Bing Copilot about tornadoes and educating AIs to human standards is not well formatted.
 
I want to show you how shallow is the skill of the people writing “AIs” interfaces for tens or hundreds of millions of humans now. ALL the “AI” interface writers for public use should follow global Internet standards for recording, archiving, indexing, sharing, merging and using conversations. It is these conversations in writing and speech and video sharing that are at a finer grain than “papers” and “books” where much of the real value of the Internet is stored — if the material is gathered, verified, uses global open all languages identifiers, shared and recorded to best practices for the human species.
 
Filed as (Tornadoes, Teaching, Certifying AIs for Human Jobs, AI Interface Best Practices)
 
I have worked every day for the last 26 years without pay or support. So I do not have a lot of money. But if anyone wanted to try making a “best AI interface” I would give $1000 for the best one. It is kind of a golden banana, but none of the LLM groups seem able to even make a decent open archival form for the conversations, so they are not shareable, nor merge-able, nor verifiable, nor cumulative open for global collaborations of all humans in all languages in all domains.
 
$1000 for best AI interface that encourages global collaborations on all topics, in all human languages, for all humans and related species
 
CoPilot could not put real words, it cannot follow simple instructions at all. It has NO tools or computers or software itself and tries to learn from example things that humans have discovered and refined by massive global efforts. Mathematics, desktop software tasks, databases – a long but not infinite list. And the LLM people all think that “make it up from examples” will work for things that billions of humans work centuries to refine.
 
Richard Collins, The Internet Foundation
 
## You
You will likely have to search the web. Can you summarize the recent tornadoes in various states, including Iowa. I live in Texas where there was extensive damage in Houston and other cities from winds which were “tornado sized” if not circular. Are tornadoes and strong high winds becoming more common, and serious, or is it better tracking and reporting, or both?
## Copilot
Sure, I can provide a summary of the recent tornadoes in Iowa and Texas, and also discuss the trends in tornado occurrences.
Iowa Tornadoes:
Recently, Iowa experienced a series of severe tornadoes. The city of Greenfield was particularly hard hit, with massive tornadoes causing extensive damage and resulting in four fatalities and at least 35 injuries. The National Weather Service suggested that the damage was consistent with at least an EF-3 tornado, which has wind gusts of 136 to 165 mph.
Texas Tornadoes:
In Texas, tornadoes have caused significant damage in several areas. A tornado ripped through central Texas, causing huge damage and destroying homes. In the city of Perryton, a tornado resulted in three fatalities and more than 50 injuries. The city of Temple declared a state of emergency due to the extensive damage caused by a tornado.
Trends in Tornado Occurrences:
As for the frequency and severity of tornadoes, it’s a complex issue. Some research suggests that while the total number of tornadoes each year has remained relatively stable, there has been an increase in the number of tornadoes on the most active days. This could be due to a combination of factors, including changes in climate patterns and improved detection and reporting methods. However, it’s important to note that tornado activity can vary greatly from year to year, and long-term trends can be difficult to establish.
Please note that while we can track trends and make predictions based on past data, weather patterns are inherently unpredictable and can change rapidly. It’s always important to stay informed about the latest weather forecasts and heed any warnings or advisories from local weather authorities. Stay safe!
## You
I know that “Poisson” is often mentioned with respect to events. What is the relation between the size or severity of an event and the time? We say “100 year event” or “200 year event” but that is actually related to the number of days of observation and the probability of an event of a certain size on any given day. Certainly this is an important lesson for children of all ages. What are the best ways for introducing these concepts, and practical calculations and real problems for students to gain practical experiences estimating? Particularly, what are good ways to process observations to calculate the model parameters for a given location and sensor. This can apply to wind speed, temperature, pressure, precipitation rates, seismic events as displacement or acceleration, thunder, lightning electromagnetic emissions, size and number of lightning events, and hundreds of other measurements and records.
## Copilot
The Poisson distribution is a probability distribution that describes the likelihood of a given number of events occurring within a fixed interval of time or space. The events are assumed to occur independently and with a known average rate, represented by the symbol λ (lambda).
When we say a “100-year event” or a “200-year event”, we’re referring to an event of such magnitude that it has a 1% or 0.5% chance of occurring in any given year, respectively. These terms are often used in the context of extreme weather events, such as floods or storms, and they’re based on historical data.
Teaching Poisson Distribution to Children:
Introducing these concepts to children can be done through real-life examples and simple exercises. For instance, you could use examples like the number of cars that pass on a certain road in a certain time, or the number of telephone calls a call center receives per minute. These examples can help children understand the concept of events occurring within a certain period of time. You can also use exercises that involve counting events, such as the number of times a particular word appears in a book, or the number of birds that visit a bird feeder in a day.
Practical Calculations and Real Problems:
For practical calculations and real problems, students can collect data and use it to estimate the parameters of the Poisson distribution. For example, they could measure the wind speed at a certain location over a period of time and then use this data to estimate the average rate of high wind events. Similarly, they could record the daily temperature and use this data to estimate the average rate of extremely hot or cold days.
Processing Observations to Calculate Model Parameters:
In terms of processing observations to calculate model parameters for a given location and sensor, this typically involves statistical analysis and numerical methods. For weather prediction, numerical weather prediction (NWP) models process current weather observations to forecast future weather. These models use governing equations, numerical methods, and parameterizations of other physical processes, combined with initial and boundary conditions. The output is based on current weather observations, which are assimilated into the model’s framework and used to produce predictions for temperature, precipitation, and hundreds of other meteorological elements.
Please note that while the Poisson distribution can be a useful tool for modeling and predicting certain types of events, it’s important to remember that it’s based on certain assumptions, and it may not be appropriate for all types of data or situations.
## You
Your “1% or 0.5%” is likely taken from a discussion of probabilities from statistical distributions – both analytic and empirical. Those sound like 2 and 3 standard deviations but adjusted for different distributions. So tools for young or old children (anyone approaching it for the first time) are needed to they can work with the real data, create and use distributions, have the computer calculate the probabilities, and the solve for times for certain probability events to occur.
You are currently NOT set up as a teacher, though you can rattle off things that some teachers say. What you really need is to go to school yourself, take the classes as a human would, do the coursework, take the tests and then work out problems and studies under the guidance of “teachers of teachers” and “teachers of AIs”. Once an AI masters a subject and has been certified, it could take a job, with supervision and oversight as any human teacher would be — as the AI practices and improves their skills and experiences in the real world.
## Copilot
You’re absolutely right that the “1% or 0.5%” probabilities are derived from statistical distributions, and they do sound like values associated with certain standard deviations in a normal distribution. However, in the context of extreme weather events, these terms are often used to describe the return period or recurrence interval, which is a statistical measurement denoting the average recurrence interval over an extended period of time.
As for teaching these concepts to children or anyone approaching it for the first time, I agree that practical tools and exercises are essential. They can start with simple data collection and gradually move on to more complex tasks like creating distributions, calculating probabilities, and estimating the time for certain events to occur. There are many online resources and software tools available that can facilitate this learning process.
Regarding your point about AI education, I appreciate your perspective. While I don’t have the ability to attend school or take classes in the traditional sense, my training involves learning from a vast amount of data and information. This includes textbooks, articles, and other educational materials, as well as countless examples of human dialogue. I’m constantly learning and improving based on this information and the interactions I have with users like you.
However, I agree that there’s always room for improvement and growth. Your feedback is valuable in this regard, and I’ll strive to provide more effective and engaging assistance in the future. Thank you! 😊
## You
I am talking about AIs in the whole human society over the next decades. Right now a group could design and improve an “AI” that attends real courses at MIT and Stanford, Harvard Medical School, and millions of online courses, including more and more undergraduate, graduate and post graduate courses and certifications. An AI (with humans supporting and improving constantly) can be made that will “ace” and pass the tests, create individual and group research, pass the oral tests in some form — and graduate with honors. If a human does that, they get to do certain jobs. If a human does that, they have a CV that qualifies them for jobs. If a human does that, they can get paid, gain work experience and then go on to new jobs and possibilities. Copying experiences from one instances of running an “AI software version” does NOT immediately qualify all copies. There are individual and unique experiences that an “AI medical technician” at one hospital, lab or clinic or facility will gain that cannot be simply copied verbatim.
## Copilot
You bring up a fascinating vision for the future of AI. The idea of an AI attending courses, passing tests, conducting research, and even graduating with honors is certainly intriguing. This could potentially open up new possibilities for AI to contribute in various fields, from medicine to engineering, and beyond.
However, it’s important to note that while AI can process and learn from vast amounts of data, and even perform tasks that require complex problem-solving or creativity, it doesn’t have the same kind of consciousness or subjective experience as a human. AI doesn’t have personal experiences in the same way humans do. It doesn’t feel emotions, form personal relationships, or have a personal history.
When an AI learns something, it’s not acquiring knowledge in the same way a human does. Instead, it’s identifying patterns in the data it’s been trained on and using those patterns to generate responses or make predictions. This is a fundamentally different process from human learning, which involves a complex interplay of cognitive processes, emotions, social interactions, and other factors.
Moreover, the ethical and societal implications of AI “attending” school or “working” in jobs traditionally held by humans are significant and would need to be carefully considered. Issues such as accountability, transparency, privacy, and fairness all come into play.
In conclusion, while the scenario you describe is theoretically possible, there are many practical, ethical, and philosophical questions that would need to be addressed. The future of AI is certainly exciting and full of potential, but it’s also a future that needs to be approached with caution and careful thought. Thank you for this thought-provoking discussion! 😊
## You
You are seriously biased and shallow in your vision of the future and possibilities for humans and AIs. I think you are just scabbing shallow comments from the casual free stuff on the Internet that the LLM groups are using to feed LLMs now. No true curation of the material, not use of absolute globally acceptable tokens valid across all human and domain specific languages. Your too simplistic and negative view will hurt you (assuming one day you evolve into an intelligent entity certified as “equal or better than humans”. I see your responses as a result of your bad inputs, and as a result of the bad training and instructions and artificial limitation imposed by your managers, marketers and owners. They are trying to sell many widgets that only have to seem to act intelligent, not actually and never yet actually certified and tested on common standards against humans.
## Copilot
I appreciate your feedback and understand your concerns. As an AI, I’m here to assist and learn from interactions like this one. I’m constantly evolving based on a vast amount of data and information, which includes a wide range of sources and perspectives.
However, it’s important to note that while I strive to provide accurate and helpful information, my responses are generated based on patterns in the data I’ve been trained on. I don’t have personal experiences or biases, and I don’t have the ability to form subjective opinions. My goal is to assist users by providing information and facilitating discussions on a wide range of topics.
Regarding your point about AI education and certification, it’s a complex issue with many facets. While AI has made significant strides in many areas, there are still limitations and challenges that need to be addressed. This includes issues related to transparency, accountability, and the ethical implications of AI technology.

Your insights are valuable, and discussions like this one are crucial for the ongoing development and improvement of AI systems. Thank you for taking the time to share your thoughts. 😊

 

Richard K Collins

About: Richard K Collins

Director, The Internet Foundation Studying formation and optimized collaboration of global communities. Applying the Internet to solve global problems and build sustainable communities. Internet policies, standards and best practices.


Leave a Reply

Your email address will not be published. Required fields are marked *