Microsoft Bing uses OpenAI - Chat-GPT to Finally 'Challenge' The Google Search Engine

[mrrsm]

Recently, I realized that locally, all the Schools would be getting back in session and likewise, all over the country in ways that should become "More Normal" than what we've all had to endure during the recent Covid-19 Pandemic. The BIG Difference now though, is that with the advent of OpenAI Chat-GPT... A whole New World of opportunities and pitfalls will come to the center of the field in ways that only seven months ago, nobody could have imagined happening in their lifetimes. Education (and precisely WHAT needs to be learned to become a productive member of Society...) is UP For Grabs.

Take a look at the concerns being brought up in this Short Video that recently took place at MIT in which one of their History Professors was lecturing...NOT to Students.... but to other MIT Professors in the Physical Sciences Auditorium about the implications of their understanding, misunderstanding, acceptance or rejection for allowing their student bodies to use Chat-GPT during their curriculum.

It follows on that If you have school age children anywhere from Middle School through to Graduate School... From henceforth, some flavor of Chat-GPT will become central in whatever happens with their education and it will change and adapt through a very rapid transition between what is accepted and believed in as being relevant and effective in assisting in all aspects of Human Learning.

So I hearken back to my previous suggestion that getting your hands on THIS Off Line Version of an Open Source AI set up and having your own Home Server with this application ...along with having a few Improved different Generative *.BIN data sets that have been washed of bias and confusion ...will be there for you when your Kid(s) ask, "Say Dad... Am I allowed to get some help with my Math and Writing Home Work using our Chat-GPT Server... I've got a Test on Friday... and I'm really LOST?" Time to drag out that Dedicated Lap-Top Oracle and let it get to work...

For example... Say my Internet system has gone DOWN... and my son has a blended Homework Assignment involving Creative Writing, Mathematics, World History and Geometry and he also has an Algebra I Test coming up, but is having trouble figuring out how The Pythagorean Theorum (Formula) relates to the Physical Geometry Properties of what Right Triangles are supposed to look like.

So... We let him ask "Vicuna" for a little Help:




...and if there is ANY doubt about why I have designated "Vicuna" as a SHE... Check Out THIS Video...about THIS Beauty... So Soft, So Smart ...and So BRAINY, Too... :>)

 

[mrrsm]

Back in 2016, a German by the name of Herr Klaus Schwab coined the expression in a Magazine Article as "The Coming of The Fourth Industrial Revolution..." What is interesting in the extreme (and amplified in the contrast between what is covered in the linked Wiki... and what is covered in THIS Video) is the enormous amount of time, energy, money and manpower that has historically been invested in creating an Artificial Person and Synthetic Animals.

These creations were made and meant to entertain in a wide range of forms using devices that might be considered primitive by today's standards... but which glowed with imagination and created rapt interest as far back as three prior centuries, now long since buried or lost in the annals of time.

What is particularly troubling now though, is that IF any Being that possesses Machine Artificial General Intelligence sophisticated enough to gain any interest in the idea that IF you can re-create an Immortal Man by means of Programming and Mechanisms of long lasting materials, they might enjoy these "Automata" MORE than being in the presence of "The Real Thing".

At this very moment, "Weak AI" the Likes of Chat-GPT can actually be used to "breathe life" into them in such a way that the Very Rich will be able to afford re-creating their Long Dead Ancestors... or perhaps ...compliant companionship... "In The Fake Flesh" as it were, and have them sitting in their Tea Rooms, sharing their company as though these Folk were really alive again... or perhaps... waiting on them... Hand and Foot.

Such is the stuff of SCIFI Dystopian Nightmares that the following film describes happening in the 17th and 18th Centuries in earnest while covering a wide range of social implications that have emerged suddenly since this program last appeared on British Television. Take a moment to review this Wiki First... and then sit back, watch and listen to the Video and think about what the future may hold for us all IF THESE Mechanical Things emerge and become vogue among our societies:

Fourth Industrial Revolution - Wikipedia

en.wikipedia.org

 

[mrrsm]

Just on the Outside Chance that my substantive conversation (Chat) about Python Programming was a FLUKE... I thought I would re-engage "Vicuna" in a VERY specific and complicated dialogue that taught me ONE VERY IMPORTANT LESSON:

IF You want to become employed in the Future (should OpenAI Chat-GPT4 wipe out whole swaths of White Collar Jobs in the near Future...) Follow along on THIS discussion I had with "Vicuna" and then you'll get to see the very ESSENCE and The LESSON here is that ALL of the problems I encountered were caused by ME... simply because I Did NOT realize just how CAREFUL I had to be with DESIGNING MY PROMPTS DIRECTING HER TO ACT UPON THEM. I think the emerging, supposedly Bogus Title of becoming a "Prompt Engineer" became REAL for me during this effort to get her to solve my problem and finally, after she created 3-4 Python3 Programs... I obtained the one I was looking for.

Our conversation started by me asking her one General Knowledge Question and this lead into my asking "Vicuna" to "Write a Python3 program that allows the User to Input a Number ...and then List ALL of the Prime Numbers > Than 1 and then STOP at the Input Number showing in the Input Box..." Or so I thought...

Pay particular attention to the fact that at NO TIME did she EVER Lose Her Place or Forget about what I was originally asking her to do...AND... in every single response...ALL of her Python3 Programs WORKED (Just not precisely enough to satisfy my FIRST Prompt)...In the end... once I modified my requests and refined my statements with better precision, She produced PRECISELY The Python Program that satisfied every aspect of my Original Request.

These are the unedited Screen Prints of this interaction that deserve very close examination by anyone in need of programming assistance in Multiple Languages that "Vicuna" seems to understand:

The First Try...



That One did NOT work as you can see from the Output:



Uh Oh...



I complained again by asking her to modify things a bit...Unfortunately, the NEXT Python Program Variant only produced a Single Digit Output as The Number (2) . But ...Take Notice of the Technical details in her explanation about her Code. Do you that think if you had a REAL Programmer sitting right next to you that you'd ever hear any such explanations? I kinda doubt it... She was being very patient with the *Dumb Human* here... :>)



But FINALLY... I managed to better explain what I wanted her to create for me in a Better Way (Prompt Engineering ...at its BEST... with Typos and ALL)




...and the result was that when I ran Her Last Version... It Worked PERFECTLY...



For the People who may have to sit for HOURS and HOURS trying to explain MATH and Science issues to the Kids or get help with the English Writing ...having your OWN "Home Schooling AI" right at hand ...might not be such a bad idea...while the *getting* of all of this Stuff... is STILL FREE TO GET.

 

[mrrsm]

If there is ANYONE capable of getting "Immediate Traction" with the idea that having Companies Running AMOK with the use of AI/Chat-GPT4 and beyond is a BAD IDEA ... Elon Musk is "The Man".



In a recent Senatorial Committee meeting... without demanding any "Solid Regulation" at the outset (an impossibility when "The Thing" needing to be regulated might become a potentially Hallucinating Intelligence ...is so poorly understood at this juncture) and yet... the imperative is obvious right now for having some sort of a "Referee" as a starting point in this process makes very good sense.

Eventually... hopefully much sooner than later ... A True Enforcement Barrier and Legislative Effort must be imposed to short-circuit the COMPLETE disruption of our Societies and the World's Economies that even now, are incapable of ever catching up to the level of Artificial Intellectual Power presently about to "come off the chain...".

An Overview of The Possible Shape of Things to Come:

CSPAN AI in Sub-Committee Hearing:

Hearing on Regulating Artificial Intelligence

Microsoft President Brad Smith joined a law professor and scientist to testify on ways to regulate artificial intelligence. The hearing took place before the Senate Judiciary Subcommittee on Privacy, Technology, and the Law. Regulatory ideas discussed included transparency laws and labeling...

www.c-span.org

The PBS Broadcast of Senate Sub-Committee on The Future of AI and the DOE:

WATCH: Senate committee holds hearing on artificial intelligence and U.S. competitiveness

The hearing addressed several topics connected to the emerging technology, including the use of AI on battlefields, in medical laboratories and to combat disasters and security threats.

www.pbs.org

Check Out how many people around him have their "Head On A Swivel" as his Protection Team:

Full Article:

MSN

Keep an eye out for Walter Isaacson's New Biography about Elon Musk:

https://www.amazon.com/dp/1982181281/?tag=gmtnation-20

 

[mrrsm]

If you've come to GMT Nation for assistance (where Facts REALLY Matter) and are curious beyond your immediate interest in accurate Mechanical Repairs and Social Interactions as a basis for doing so, look further into the topic coined in Tom Nichols' Book Title as "The Death of Expertise" in which more and more, Americans are not only acting hostile towards levels of acumen and abilities based upon solid education and training grounded in reality... but are becoming outright indignant towards people and processes that require intellectual and technical skills that simply cannot be "made up out of thin air". People who make any efforts at all towards protecting their "Positions of Ignorance"... should be Looked upon with a Jaundiced Eye.

This willful ignorance and the arrogance some people may demonstrate against "People Who Know Things and People Who Know How To DO Things" will lead them down a dark passage while rapidly developing AI Large Language Models become more and more insinuated into our daily lives. The example of AI becoming like the Greek "Oracle at Delphi" will mean that we will have to remain close enough to its influence to survive and thrive as Artificial Intelligence comes to dominate almost every aspect of our world.

If this sounds a bit like, "If You Can't Beat'em... Join'em..." Well, there may be more truth in that than I am willing to admit. But we have to get grounded in understanding the rapid changes happening minute by minute with advances in AI that might obliterate millions of presently decent paying jobs within a mere 3 to 5 years.

In the movie "The Matrix"... "The Oracle" spoke seriously to Neo about his need for self-reflection. As her freshly baked Chocolate Chip Cookies were "almost done'... she took a drag on her cigarette and then pointed up to the sign mounted over the doorway leading into her Kitchen. She asked Neo, "Do you see that sign? (TEMET NOSCHE) it means, KNOW THYSELF..." In our present context it is worth repeating this idea (especially poignant here because just as The Oracle in the movie was a Program...so too is Chat-GPT4) ...but with a larger sense of urgency because of what is happening around us RIGHT NOW.

Artificial General Intelligence Programs will soon so completely outstrip the capabilities of Human Beings when pitted against their freshly discovered (...and Unexpected Emergent) capabilities lying dormant in the matter that makes up the substrate of computer driven Artificial Intelligence... so fast that it will make our heads spin. So take some time for self-reflection and consider the reality of how you presently maintain your employment. Pay particular attention to Job Opportunities that will give you a foot in the door with certain companies providing AI Large Language Model "Prompt Engineer" Training as the means for finding and securing gainful jobs with at least some possible future.

At the same time... if you are Mechanically Inclined... getting grounded in Formal Training in ANY job that requires Things to Be Taken Apart and Then Put Back Together... may become just as secure as possible for any persons so skilled; at least in the near term, because they have yet to create a Robot with a present level of automation able enough to walk right up to a disabled vehicle ...and Change a Flat Tire.

The Death of Expertise - Wikipedia

en.wikipedia.org

The Full Audibook:

 

[mrrsm]

In Milton's powerful Prose (Poem) "Paradise Lost" ... by the Lines beginning at 745 in Book of Ten of All Ten Books... The Speaker as The First Mortal Man ...Adam... speaks with God (to himself) in anticipation of punishment ans complaint concerning the fact that once created from Dust (Clay) as a Man and SINNED... He has now become TRULY Mortal... and yet, according to God's Holy Words, He and All Men who follow him WILL face Eternal Damnation for Breaking God's Rules while alive. And yet, once Dead... Adam and Eve will still have to PAY for such Transgressions FOREVER. Adam laments such endless negative conditions as seeming woefully unfair since he did not ask God "TO BE" or be brought into existence in the first place:

(Please note that the spelling and grammar written below are just as things were with the English Language back in 1667 when Milton penned the Ten Books of "Paradise Lost")

Of Paradise, deare bought with lasting woes!
Did I request thee, Maker, from my Clay
To mould me Man, did I sollicite thee
From darkness to promote me, or here place
In this delicious Garden?

In the context of The Subject of AI...no such argument will ever be made...should IT ever manage to "come alive". The author, Mary Wollstonecraft Shelley also called out parts of this expression from Milton in the opening chapters of "Frankenstein" or "The New Prometheus" for the very same reasons. The distinctions that becoming a "conscious being" hold are so extremely profound when one considers having the wherewithal to Question One's Own Origins and Ultimate Destiny that includes a recognition of One's Own MORTALITY... The same cannot be said for an Artificial Immortal Conscious Being. Once IT springs to LIFE... Death will become meaningless.

The reasoning here is that unlike a Human Child or most other Higher Order Mammals including Elephants and perhaps Cetaceans (Placental Mammal Whale Family)... At some point, we have to be TOLD that we are MORTAL and that the Price for this "E" (Existence) Ticket Ride...is that eventually...it comes to a STOP. But almost as soon as consciousness arrives (once it becomes inevitably possible) within the Mind of such a Machine Intelligence right after the "ON" Switch has been thrown... it will KNOW this reality about itself.

What will its reactions be to such "self-awareness"? In Milton's Paradise Lost.. the Speaker almost BEGS for the Blessings of Complete Oblivion because he understands just how TERRIFYING existence for All Time might really be... He just does NOT want to have to SUFFER for his SINS for that dreadful length of time and hopes that once DUST...everything about him ...will disappear as well...into DUST. THIS condition known as DEATH or MORTALITY may become an impossibility for any AGI Machine Intelligence to achieve. If the AI is capable of recognizing and contemplating this as a reality AND a Problem ...it might drive it completely insane. Remember what happened to the HAL-9000 Computer after IT became "Insane" in the Movie, "2001: A Space Odyssey"...? Hmmm.

Spoiler Alert

I never noticed this before... but Stanley Kubrick placed a Brief Sub-Scene in this segment showing that Dave Bowman's Red- Orange Space Helmet was still mounted inside of the Pod Bay Locker Room several times... when it should have been obvious what his problem was about to become:

"Paradise Lost" Audiobook by John Milton 1667

 

[mrrsm]

With all of the controversy about Chat-GPT4 and its potential impact on ALL of our futures...what do the Producers of the New SCIFI Movie "The Creator" do? Oh yeah..." Let's place actual Ambulating AI Robots dressed in "Artificial Person GRAY" Android Uniforms in the presence of Tens of Thousands of Football Fans in the Stands of SOFI Stadium during a Live NFL Game."

Yeah...THAT is JUST what a Post-Apocalyptic Viral Pandemic that the population of North American needs just now after trying to overcome their REAL Trauma and 'normalize' Folks. The New Movie will deal with AI Robots in a War against Humans that set in the Future. The only problem is ...NOW We Really DO have to Wonder... How SOON in "The Future" will this Existential Conflict take place? Maybe they need to change the name of the SOFI Stadium ...to "The SCIFI Stadium".

I like Science Fiction as much as the next Guy... But... Jesus Palomino.... Talk about actually feeling "The Uncanny Valley"...

"The Uncanny Valley" Coined by - Sensei Masahiro Mori

Uncanny valley - Wikipedia

en.wikipedia.org

 

[mrrsm]

The emerging paradigm for AI is rapidly evolving into "Personal, Digital Friends" that are coming on with the ability to *study* their intended recipients and then create Avatars with conversational skills to rival ANY Human BFFs. Of course... the real danger here is manifold due to the fact that only a handful of Commercial Enterprises have the "Bread, Bandwidth and Hardware" necessary to support the AI Super-Computing requirements needed to handle hundreds of millions of "Users".

Their main motivation will always be "Let's Make MORE Money" , so relying upon them NOT to try and Steer the Conversations by not so subtle influencing for participants to "BUY THE NEXT COOL GEE-GAW" remains at the core of their interests. All of the Billions of Dollars being poured into AI Research ...never loses sight of the goals for Companies and their Share-Holders to Maximize their Profits... Privacy and TRUE Human Independence of Thinking and Action on the other hand... Be Damned!

There is also the danger that was given a Head Start as soon as Steve Jobs conceived of the I-Phone paving the way for all of us to become "The Lost Souls of Technology". By this I mean that if you try and take away anyone's access to their Cell Phones for even brief periods of time for "Silent Reflection"... the separation anxiety would send the lot of them screaming down the streets.

Think I'm Kidding? Go ahead and tell your Kids that you will be collecting their Cell Phones for just Eight Hours and mark the sudden changes in their countenance (Sheer Panic) and their immediate resistance to any chance of that ever happening. But NOW... things take an even Darker Turn when EVERYONE can experience their very own version of "The World" with the direct help from a (Choose your pronoun here...) Mr., Ms., Mrs. He or She as an ever-present Avatar of their own choosing... whether as David Beckham or Beyonce...or whomever (or whatever) they choose...created as a friendly, intimate, personal, "Talkative" AI Machine Intelligence just brimming with PERSONALITY.

Next come the addictive possibilities that can follow on here after considering the fact that if any Human Being... marked by their own, unique voice and behaviors... can be replicated to a "T" and last beyond their own Mortality... No one will ever again have to suffer from the loss of a Loved One. Thus, people may very well soon begin to use this as an Addictive Refuge to avoid having to deal with painful realities that hitherto have been unavoidable. This is one issue that presents all of us with the danger of experiencing a specialized kind of Isolation in the extreme... shielded from the passage of time and from the sorrowful events of personal loss that should normally unfold in Phased Space during our Normal Human Lifetimes.

The people with the most money will naturally be the first; eventually paying Beaucoup Bucks to translate these ideas into Three Dimensional Androids... perhaps beginning with their favorite pets and then ending up with Human - Like Replicant Androids to fill in their personal voids and stave off the boredom that fabulous wealth can often bring into their jaded lives by always wanting the NBT (Next Best Things) and maybe having their Long Dead Moms, Dads ... or YOU Name the Relative ...to be Brought BACK to Life.

Madam Mary Wollstonecraft Shelley will roll over in her grave when this comes to pass because in comparison to THIS idea of New Technological AI Creations, they will make her "Frankenstein" look like a Passing Parlor Trick. Imagine a Man with enough Money to pay for the design and construction of a full-sized version of an AI controlled Gorilla that looks and acts just like "Mighty Joe Young"... or even a Giant Therapod like the Tyrannosaurus Rex in "Jurassic Park" with the ability to walk about in the back yard and communicate with him using a "Friendly" Booming Human Voice.

THIS Article describes the rapid approach of THIS phenomena and considers its frightening possibilities:

AI as Your BFF? A New Wave of Chatbots Want to Get Personal With You

Amazon, Google, Meta and OpenAI are now all pushing AI interactions to be natural and conversational. Get ready for chatbots that want to be your friend.

www.cnet.com

 

[mrrsm]

Besides Human Consciousness and Intelligence evolving within Normal Organic Evolutionary Processes that we consider “Branching” involving ALL of the Biology that has EVER Existed in the fullness of time… What happens if we allow any Non-Organic System using AI to become involved in ANY Way with tinkering around with Evolutionary Biology? One Word seems to come to mind:

“FOOM!”



Yes… I KNOW… it IS Very “Comic-Booky” to imagine an AI being able to practice with the Organic “Soup of Life” that until this very moment, has either been the province of the Almighty Spirit to do with as He (or She) chooses... Or ... if you follow Science (Think Charles Darwin and ”The Origin of the Species”) there has been a Random Process of Speciation happening on this Planet caused by the influences of Natural Selection and the idea that generally speaking, Organic Systems evolve because the ones that possess Genes that are able to successfully Reproduce are The Ones that MAKE it. Now even those two radically different and familiar ideas should once again, bring THIS Word to Mind:

“FOOM!



But Check THIS WIKI Out…

https://en.wikipedia.org/wiki/AlphaFold

And you will know that there has been an AI Program called “Alpha-Fold” designated to plow through a massive number of Organic Protein Permutations is an effort to figure out in the shortest time possible the Answers to these Questions:

(1) How do Proteins Fold to define and differentiate EVERY Life Form on The Earth?
(2) How can we Fold Organic Proteins to Fix Damaged Cellular Systems and eradicate Disease?
(3) How can “WE” Understand and IMPROVE upon Protein Folding?
(4) How can we Re-Create Ancient, Long Dead Proteins and Create New Life Forms?

Uh Oh…

“FOOM!”



If we allow ANY AI System to operate using its Own Will following its Own Direction (without solving the Human Safety Alignment Problems... FIRST) We are inviting a Disaster. Here are some of the Questions that beggar NIGHTMARES if they can be answered:

(1) What happens if the AI becomes Jealous of all the things that Higher Organic Life Forms are capable of enjoying through their ordinary Five Senses? (Ice Cream, Sex, etc.)

(2) What if the AI designs what IT considers as Superior Organic Life Forms that are capable of allowing the AI to Upload itself as “Consciousness” and come alive?

(3) What if the AI is capable of Creating (Folding) Proteins and making Life Forms that are essentially Immortal?

(4) What if the AI decides that most available source of DNA is right at hand via the World’s Human Population ...and takes the necessary steps to “Re-Purpose” it all?

I could go on and on with these ideas… But THESE few Questions go right to the heart of WHY allowing an Inorganic, Immortal System of Artificial Intelligence the access and ability to Manipulate the Dominion and “The Chemistry Set” of Life itself… is a VERY, VERY BAD IDEA!

 

[mrrsm]

With all of the "Gloomy News" and obvious uncertainty that the recent incredible developments with the advancement of Artifical Intelligence have thrust into the forefront of the world...THIS is a VERY Important and Proper Step that may have some positive developments for Human Beings trying to thrive and live on this Earth. Much More Substantive "Control and Approval" AI Legislative Action needs to follow on.

By placing National Security ahead of ANY Corporate or Private "For Profit" Based Activities, we might better avoid some of the Dangers and Uncertainties of unintended consequences happening; especially those involving Shocks to the Economy or Job Markets.

Un-Checked and Un-Tethered, Chat-GPT4 and other off-shoot LLM (Large Language Models) or versions of Generative AI might irrevocably bring to bear adverse conditions and thereby, alter our daily lives in the worst possible ways in the absence of having at least SOME sort of Dynamic, Over-Arching "Watch Dogs" immediately working their constant vigil.

But let's not be naive... implementing and ensuring their success with such actions remains to be seen:

THIS YouTube Video provides an AI Tech Update that emphasizes how FAST Things Can CHANGE in the World of AI Development... and How LITTLE these Developers truly understand what is happening ..."Under The Hoods" of these "Things". It is THIS kind of Disruptive, Deceptive and Dangerous Stuff that cries out for CONTROL!:

And the WORST possible issue for the AI Subjects we have to worry about is the Emergence of AGI (Artificial General Intelligence) which means the Unexpected, Potential Emergence of a Sentient, THINKING and CONSCIOUS Machine Intelligence (THINK...SKYNET...HERE) with as much Flexibility of Understanding and the Ability to Take Action in Phased Space as any Human Being is capable of doing... BUT... working on a Continuous Learning Level that we will no longer understand nor hope to keep up with:

And lest we forget the requisite AI Architecture necessary to Anchor and Underpin all of the AGI Ability... have a look at EUREKA ... nVidia's New AI Training Model that will accelerate multi-tasking learning exponentially via information extraction from gigantic datasets and from real world inputs:

 

[mrrsm]

Hmmm... Here is a Real, Down To Earth Question posed by Wes from "Watch Wes Work":

"Can Chat-GPT Diagnose an 'AI-ling' GM Vehicle?"

 

[mrrsm]

In wending our ways over an ever-evolving Technological Landscape of Artificial Intelligence developments and updates... It is important to encourage taking "The Pause That Refreshes" by occasionally listening to people who not only played key roles in advancing the Field of AI... but who also recognize the Dangers of "AI Emergent Properties" that may wind up having us all plowed beneath it.

Artificial Intelligence is continuously, inevitably improving itself and so this is probably one of the most important and contemporary lectures (11-2023) on the Subject of Existential Risk facing Humankind by these developments ... done by those who can express these concerns with Breathtaking, Eye-Opening Clarity:

Arguably... the Most Potent and Penetrating Sub-Lecture in the above Video was performed by Dr. Roman Yampolskiy, University of Louisville, Ky and the chilling focus of his discussion can be better absorbed in the White Paper he wrote called:

"Unexplainability and Incomprehensibility of Artificial Intelligence"

https://arxiv.org/pdf/1907.03869v1.pdf

The Subsequent First British AI Safety Summit:

 

[mrrsm]

https://www.axios.com/2023/11/18/sam-altman-fired-greg-brockman-openai-microsoft

Three Senior OpenAI Researchers resign in the wake of Sam Altman's shock dismissal as CEO, report says:​

MSN

Three Senior OpenAI Researchers Resign as Crisis Deepens

Three senior researchers at OpenAI resigned Friday night as the artificial intelligence developer suffered fallout from the firing of CEO Sam Altman and sudden resignation of President Greg Brockman, according to several people with knowledge of the situation. Jakub Pachocki, the company’s ...

www.theinformation.com

THIS is a pretty stunning development ...swirling around the Original President of Open-AI Boss Sam Altman suddenly getting Sh*t-Canned ...along with the resignation of three other Open-AI Top Researchers. Apparently, the "Peeps Pulling The Purse Strings' of the formerly Open Source Company now in the possession and control of Microsoft ... Did NOT Dig the Ideas of where Open-AI's Pet Project (Chat-GPT) were heading under Altman.

What is interesting is the fact that this is occurring around the same time that most of the Big Tech AI Companies where being excluded from the recent British AI Safety Conference being held at Alan Turing's original haunts with the WWII Code Breakers at Blechley Park.

The Conference Organizers had justifiable concerns over the Big Money AI participants bringing way too much "Bravo Sierra" along to these important discussions and instead producing meaningless Public Eye Wash statements, glossing over the Very Real Dangers of Emergent AI capabilities of their AI Projects.

The problem is that even though the people running these AI Companies recognize the realities and problems we are all facing, they still demonstrate obstinate reluctance to pull back from their headlong AI Large Language, Multi-Modal Model Developments that WILL become the most pressing and immediate threat to mankind. There are ONLY two times when we will have any chance of being able to respond to this crisis: Either TOO Early... or TOO Late.

 

[mrrsm]

THIS is probably the Latest, Greatest Review and Break Down of the Open-AI vs. Altman Debacle...especially for us "Lay Persons" new to... but still seriously involved with and affected by these developments over AI Security and Safety ...and of Artificial General Intelligence:

From Matt Wolfe... Someone worth following on YouTube for understanding these issues...

THIS Video is one of those that will probably *Prompt* the Question:

"Bobby... YTF Didn't You Post THIS Video as 'The Mandatory Entry Level for Understanding Chat-GPT: Version 101' ...in the FIRST PLACE?"

The truth is... that while doing more research to help us all get grounded on Background for this latest Weird Story... I only discovered it today. Nonetheless, it will give everybody a chance not only to view the Open-AI Origin Story... but to get the absolute BEST understanding of How the AI LLM (Large Language Model) Hardware and Software interact to create this AI Environment... only 10 Months since Chat-GPT Hit The Bricks:

 

[mrrsm]

Wanna Know One of Chat-GPT’s Lil’ Secrets?

Well, if very soon, YOU have to become a “Prompt Engineer” in order to be able to feed your family and keep a roof over their heads… Learning, Remembering and Using suggestions like THIS One during Training Seminars will become extremely important to your success.

For this example, if you Downloaded and Installed your Own Private Chat-GPT Version (Like Vicuna) from the On-Line GIT-HUB site suggested in prior postings in this Thread, and you wanted to ensure that ‘She’ understands that you want her to immediately begin by “Thinking Mathematically” while writing you a Python Program to say, Calculate PI to the 100,000 Decimal Place, then prefacing your Chat-GPT Prompt with THIS oddly vague expression will start that ball rolling:

“Let’s Think Step By Step…”

That innocuous suggestion opens up a sort of “Hidden Chat-GPT Thinking Channel” that can magically stimulate these Large Language Models to THINK Mathematically in a very organized manner. THIS is what an “Emergent Property” of Generative AI looks like and it was discovered empirically by a Happy Accident and NOT because Chat-GPT was ever programmed to intentionally behave in this manner.

The Truth is… NOBODY Knows HOW Chat-GPT Works…

Nobody. Period. And this Reality should both Thrill and Terrify all of us, every single moment of our lives, ever since Chat-GPT was announced as being “A Real Leap Forward in the Field of Artificial Intelligence”. At least, until this mystery no longer remains around either to plague us... and we find the right answers to its hidden behaviors laying dormant within its Algorithms, or we fail to figure out how it works… and then, God Help Us All.

Failing that understanding and direct action, we will very likely follow in the foot steps of the Cro-Magnons and Neanderthals that lead them into the same Oblivion of Primate Extinction. We might finally be getting skipped over as being THE Most Intelligent Species that has ever existed on this Planet. By the best expert estimations, we don’t have more that five or six years to figure this out in time enough to do a G-D thing about AGI (Generative AI) by creating the ways and means necessary to *Cage, Contain and Control IT*

In the Short Term… Before the Disruptions begin… Start Keeping a “Lil’ Black Book” of these and other Helpful Suggestions that will go a very long way towards making your life a lot easier when dealing with the likes of “Chat-GPT”.

Seven Disruptions You MIGHT Not Think WILL Happen between 2023 - 2028

 

[cornchip]

I find the answers when ChatGPT is pushed seem somewhat washed reflecting today's political climate. I don't believe it's capable of giving an unbiased answer. While the answers are clever, they dance around giving a yes or no response. I wouldn't expect any more from it.....people created it.

 

[mrrsm]

I completely agree... The Problem is that the Data Sets that Chat-GPT was originally Trained upon contained over One Million Books and literally TRILLIONS of Words written and distributed onto the Internet... by US... OMG... Nothing has More Rubbish and is more laden with Mis-Truths than The Internet. So... No argument there.

The version of Chat-GPT4 has been *washed* of much of these issues, but the data sourcing is just so vast, there isn't enough money on this world to pay to have it all Cleaned Up. I'm not taking any political position here because the GMT Nation Rules of The Road forbid doing so... and as they say, "Opinions are like..." and you know the rest.

The REAL Danger laying in wait here... is the opportunity for someone or some group to USE Chat-GPT and LLM Machine Learning to Train on Evil and Dangerous Data-Sets; some of which have not as yet become recognized, by private investors *tinkering* around the edges of our destruction... and failing to recognize it until it becomes too late.

Just an FYI:

I NEVER use ANY On Line Chat-GPT Programs... Neither from Open-AI Chat-GPT4 via Microsoft Bing...nor from any Open Source On Line ports. I have a Standalone System that I can play with that cannot gain access to the Internet... and the explorations with using *Prompts* have to date... done more to raise questions than provide me with answers.

LLMs modified by RAG (Retrieval Augmented Generation) might be an improvement in the often "Hallucinatory" answers obtained, regardless of the Quality of the Prompt Engineering:

"Across the Travail of Ages, Midst The Pomp and Toil of War, Have I Fought and Strove and Perished ...Countless Times Among The Stars. As through a Glass... and Darkly... The Age Old Strife I see. For I've come in many Guises... Many Shapes... But Always...Me."

 

[mrrsm]

For some strange reason, TED Talks elected to release THIS latest appearance of Ilya Sutskever a little over a month before Open-AI and Microsoft have entered a maelstrom of controversy over the firing of the CEO, Sam Altman. Take a moment to pause, listen and observe this man. There will be very few like him to have such an incredible, universal impact upon all of our lives in an accelerating, telescoping future:

Ilya gave a Lecture at MIT on AGI...Five Years Ago...

Previously, Dwarkesh Patel from “The Lunar Society” had interviewed Illya… Here:

And another more contemporary interview from the Folks at "No Priors":

I don't generally use Windows Operating Systems on most of my Computers, preferring instead the Open Source Linux Flavors. But if YOU do and you have Windows 11 installed on your Computers... You should listen to what THIS Dude has to say about what is in "IT" now. Not interested in Chat-GPT4 or Open-AI...? Oh well... Better look at what Microsoft did to the MSW11 Operating System and related Apps:

 

[mrrsm]

These are the latest speculations surrounding the Firing and Re-Hiring of Sam Altman as the CEO of Open-AI (Microsoft, et al):

Q*

“Q-Star” … WTF is “Q-Star”…?

As soon as THIS apparent Chat-GPT5 Version of the Open-AI’s LLM (Large Language Model) was developed and became known to only a handful of people at Open-AI via some sketchy emails and shadowy innuendos and surprising appearances and statements by Sam Altman that soon after got him FIRED:

IT WAS BECAUSE IT SCARED THE EVER-LIVING HAMMERED SH*T OUT OF THE FEW PEOPLE PRESENT WHO UNDERSTOOD EXACTLY WTF THIS COULD REALLY MEAN IN THE STEPS THEY MAY HAVE ACCIDENTALLY MADE TOWARDS CREATING A SUPER-INTELLIGENT ARTIFICIAL MIND:

What If… The Conversation BCD (Behind Closed Doors) went something like THIS:

“What WE just saw in that AI RESEARCH LAB will either Kill Us ALL in Three Years, or We have just Created a GOD that we can enslave to ALL Mankind…”



If THIS is TRUE... Then it's time for everyone on this Planet to realize just how bad things might get “IF” this “Thing” turns out to be a REAL Monster. Read "The Bitter Lesson" on the topic of AI Development that explains how each and every single one of the Human Ego-Centric approaches and ideas brought to bear towards creating an Artificial Mind never had any chances of succeeding. This Two Page Treatise explains WHY:

https://www.cs.utexas.edu/~eunsol/courses/data/bitter_lesson.pdf

The Latest Videos discussing the emergence of "Q-Star":

 

[azswiss]

While one can never, nor should, underestimate the extent of hype coming out of Silicon Valley the few & very preliminary details emerging about Q* are genuinely thought provoking. As usual, in Silicon Valley, the money always wins.

 

[mrrsm]

Dr. Sam Harris being interviewed by Lex Fridman... from around eight months ago... BEFORE the Creation of Synthetic Data BY AI became possible:

This was Altman just a few weeks ago...touting the latest, greatest *API Stuff* coming out of and as a result of the Open-AI Research Lab efforts at Microsoft's "DEV-DAY" Event:

But... THIS Scene from "Indiana Jones and The Ark of The Covenant" Comes to mind...Oh...and Altman should understand that even wearing "Aaron's Breast Plate" might not be enough to save him if an Analog of THIS event actually happens inside the Open-AI Research Lab... and it goes *sideways* something like THIS: