A Suggestion

Published March 22, 2023 by Nan Mykel

DO YOU SEE ANY LIGHT DOWN OUR TUNNEL?

Earlier  in the week I read an article on the internet about how Americans have become so afraid of other Americans. They suggested all the nice psychological and open ways to approach the problem–all decent suggestions about communication solutions.

BUT–they ignored the number one possible solution to our current problem with fear of neighbors:

Remove their guns!  I remember some conservatives warning not to vote for the Democrats because they’ll “take away our guns!”

Unfortunately, we haven’t, and assuming that folks are fully armed, we have such calls as for genocide of LGBTQs and for murder of the opposition if our former president is indicted.  I thought the specificity of the postings about murder  (execution?) should be illegal, and maybe they were–they have disappeared from the internet.  Good riddance.

To the best of my knowledge automatic-kill rifles are still legal, at least in some states.

But “the people” are armed– in some states they can carry concealed and even openly carry guns.  An example of extreme concern should be in the state of Missouri, where Federal law enforcement officers say they seized 11 guns, a silencer, more than 1,000 rounds of ammunition, body armor plates and several pounds of a binary explosive from a Springfield man with ties to the Boogaloo extremist anti-government group. Local and state law enforcement are not allowed to enforce federal gun laws.  !!!

In Missouri, local law enforcement agencies, under penalty of being fined, can’t help in enforcing federal guns laws.

The Missouri situation is from <https://www.dailykos.com/stories/2023/2/7/2151687/-Arrest-of-a-Boogaloo-supporter-with-guns-in-Missouri?detail=emaildkre&pm_source=DKRE&pm_medium=email>

 

 

 

 

 

Moral Self-Correction in Large Language Models?

Published March 21, 2023 by Nan Mykel

This is over Nan’s head but thought you might like to know.

(Ha Ha, my computer wouldn’t accept “computer morality” in my tags, insisting it should be computer mortality…)

arXiv Forum: How do we make accessible research papers a reality?

Can we truly call it “Open Science” when most research papers are not fully accessible? You are invited to join the forum on Monday April 17 to chart a path towards truly accessible research papers.

Learn More   Skip to main content

Computer Science > Computation and Language  [Submitted on 15 Feb 2023 (v1), last revised 18 Feb 2023 (this version, v2)]

The Capacity for Moral Self-Correction in Large Language Models

We test the hypothesis that language models trained with reinforcement learning from human feedback (RLHF) have the capability to “morally self-correct” — to avoid producing harmful outputs — if instructed to do so. We find strong evidence in support of this hypothesis across three different experiments, each of which reveal different facets of moral self-correction. We find that the capability for moral self-correction emerges at 22B model parameters, and typically improves with increasing model size and RLHF training. We believe that at this level of scale, language models obtain two capabilities that they can use for moral self-correction: (1) they can follow instructions and (2) they can learn complex normative concepts of harm like stereotyping, bias, and discrimination. As such, they can follow instructions to avoid certain kinds of morally harmful outputs. We believe our results are cause for cautious optimism regarding the ability to train language models to abide by ethical principles.

Subjects: Computation and Language (cs.CL)
Cite as: arXiv:2302.07459 [cs.CL]
(or arXiv:2302.07459v2 [cs.CL]for this version)
https://doi.org/10.48550/arXiv.2302.07459

From <https://arxiv.org/abs/2302.07459> 49 scientists apparently  plan for  greater accessibility of research papers  April 17  is the forum to discuss how this might happen.  “You are invited to join the forum,” presumably via the internet.

 

Test Yourself

Published March 19, 2023 by Nan Mykel

What is a movie about a man who is a man who is a man?

What could the image at the top refer to?

Stephen Ornes wrote an article in Quanta magazine (Not Quora) about unsuspected abilities of the large language models (LLMs) of computational engines behind AI chatbots such as ChatGPT, and I want to refer you to that article in a minute. But first, the LLMs answered correctly on the first try at the movie question, which was Nemo. Also, it has been discovered as reported in Quanta, that the most powerful LLMs can change themselves by Emergence, a process which first referred to an evolutionary process in living things (slime molds).

The question about the image at the top is that it depicts the Trojan horse, which aided invaders to attack Troy as reported in ancient literature. The dangerousness of its presence was unknown until too late. The invaders hidden in the horse were already in the town and it was too late to avoid a catastrophic takeover.

For the interested and curious go to both Ornes’ Quanta magazine’s site and  Johnson’s Book Emergence.

What is emergence?  In his 2001 book Emergence, Steven Johnson says that “The movement from low-level rules to higher-level communication is what we call emergence.”

Johnson wonders if computers will become self-aware in the coming years by drawing upon the adaptive open-endedness of emergent software.  “Even the most optimistic champions of self-organization feel a little wary about the lack of control in such a process.”

The discovery of emergence appears to have begun with studies of slime molds, which in some conditions move as one unit but in other conditions (of food availability)  separate and function as individuals.

MORE THAN 500 MEDICAL PROVIDERS SIGN LETTER TO GEORGIA HOUSE, SENATE

Published March 18, 2023 by Nan Mykel

To the members of the Georgia House of Representatives and Senate,

Last week the Georgia Senate passed a ban on health care for transgender youth that would prevent health care providers like us from offering our patients safe, effective and life-saving forms of health care. The Georgia House passed a version of the bill today, March 16.

As medical professionals, we are appalled to see politicians promote an agenda with such disregard for standards of medical care. Gender-affirming health care is supported by every leading medical and mental health association in the U.S. including the American Medical Association, the American Academy of Pediatrics and the American Psychological Association.

We have seen firsthand how access to quality gender-affirming care has improved the lives of our patients. For children before puberty, gender-affirming health care consists of social support, where providers like us work with parents to help their kids choose clothes or a name that makes them feel comfortable. For adolescents experiencing puberty, medicine can be prescribed to temporarily delay puberty or help manage gender dysphoria, which causes intense distress and prevents young people from living their day-to-day lives.

These forms of care have long been proven to lower the risk of depression, suicidal thoughts, and suicide attempts for transgender and non-binary youth. Simply put, access to health care allows young people to live happy, healthy, and fulfilled lives that they otherwise may not have.

But a handful of politicians behind this bill are choosing to prey on people’s lack of familiarity about what it means to be transgender in order to push legislation that shamelessly contradicts decades of proven scientific research. They’re making clear that they’re prioritizing their own interests – not those of Georgia’s youth – by ignoring the facts and spreading dangerous misinformation.

While the author of SB 140 claims that the bill doesn’t intend to harm anyone, we know from our experiences that revoking access to this health care will inflict severe harm on transgender kids across Georgia. You don’t have to take our word for it – the numbers speak for themselves. In 2022, more than 55% of transgender and non-binary youth in our state considered suicide. Efforts to roll back their fundamental rights, including their access to health care, are furthering the negative impacts on LGBTQ+ youth’s mental health. It’s unfathomable that lawmakers in Georgia are failing to do anything to combat this actual health crisis, while continuing to push restrictions on the care that actually helps our communities.

If Georgia lawmakers were truly concerned about keeping families and children safe and healthy, they would be focused on the real issues facing our state’s health care system, like the fact that too many families can’t afford the basic medicine and doctors’ visits they need. Instead, we’re hearing from families who are scared they’ll no longer be able to access the care their children need, and colleagues who are scared we will face penalties or lose our jobs for simply following medical best practices.

We’ll tell you what we tell the parents of transgender kids who visit our clinics: you don’t need to understand everything about what it’s like to be transgender to know that all kids deserve to be healthy and safe. When it comes to family decisions about health care, we work hard to give parents all the information they need to make informed choices and to respect each family’s unique circumstances and decisions.

Politicians shouldn’t be in the room for private family conversations like this. SB 140 is a clear form of government overreach that violates parents’ and providers’ rights by taking away their ability to freely discuss health care options and make decisions about what’s in the best interest of transgender youth. Our government should be helping families make ends meet, not telling parents who love their transgender children that they don’t know what’s best for their own children.

Our lawmakers are again putting political games over the health, safety, and well-being of the constituents they were elected to serve. It’s time for constituents to speak up and fight for our rights to freedom and privacy. Tell politicians to get out of your family’s private conversations and out of your doctor’s office.

As this bill moves back to the Georgia Senate for further consideration, we’re urging our lawmakers to keep their politics out of our clinics. They need to leave medical decisions to the actual medical experts and reject SB 140, before it causes irreversible harm in our state.

From <https://www.ajc.com/opinion/opinion-an-open-letter-to-gold-dome-on-transgender-bill/S47QAWV6DZBSTNK4W5DJICFZH4/>

Harmful

Published March 15, 2023 by Nan Mykel

As the scientific evidence began to pile up, a consensus finally emerged that leaded gasoline had turned out to be one of the most harmful pollutants of the 20th century, one that proved to be especially concentrated in urban areas. Globally, the phaseout of leaded gasoline that began in the 1970s is estimated to have saved 1.2 million lives a year. As Achim Steiner of the United Nations noted, “The elimination of leaded petrol is an immense achievement on par with the global elimination of major deadly diseases.”

Source of the above: New York Times Great Read
OpenAI Unveils GPT-4
What GPT-4 Can and Can’t Do
Funding Frenzy Escalates
How Chatbots Work

_______________________________________

The Shift
GPT-4 Is Exciting and Scary. Today, the new language model from OpenAI may not seem all that dangerous. But the worst risks are the ones we cannot anticipate.
Photo: The team from OpenAI, the creator of ChatGPT, from left: Sam Altman, chief executive; Mira Murati, chief technology officer; Greg Brockman, president; and Ilya Sutskever, chief scientist. Credit…Jim Wilson/The New York Times
The Shift By Kevin Roose March 15, 2023 :

When I opened my laptop on Tuesday to take my first run at GPT-4, the new artificial intelligence language model from OpenAI, I was, truth be told, a little nervous.

After all, my last extended encounter with an A.I. chatbot — the one built into Microsoft’s Bing search engine — ended with the chatbot trying to break up my marriage.

It didn’t help that, among the tech crowd in San Francisco, GPT-4’s arrival had been anticipated with near-messianic fanfare. Before its public debut, for months rumors swirled about its specifics. “I heard it has 100 trillion parameters.” “I heard it got a 1,600 on the SAT.” “My friend works for OpenAI, and he says it’s as smart as a college graduate.”

These rumors may not have been true. But they hinted at how jarring the technology’s abilities can feel. Testing GPT-4 had caused the person to have an “existential crisis,” because it revealed how powerful and creative the A.I. was compared with the tester’s own puny brain.

GPT-4 didn’t give me an existential crisis. But it exacerbated the dizzy and vertiginous feeling I’ve been getting whenever I think about A.I. lately. And it has made me wonder whether that feeling will ever fade, or whether we’re going to be experiencing “future shock” — the term coined by the writer Alvin Toffler for the feeling that too much is changing, too quickly — for the rest of our lives.

Below is the observation of Nan:

Might AI not be so willing to phase themselves out if they become as deadly for humans as leaded gas? What then, eh?

P.s. Nan received this pop up when she turned her computer on this morning:

Happy Ides of March, By the Way…

Published March 15, 2023 by Nan Mykel

Just today a new pop-up on my MicroSoft computer:

<img src="https://survivorpsychologist.files.wordpress.com/2023/03/ai-screenshot.jpg?w=300″ alt=”” width=”300″ height=”218″ class=”alignleft size-medium wp-image-19819″ />

I didn’t realize AI was going to follow us around…I SPY AI!

Senator Elizabeth Warren Criticizes Federal Reserve Chair Jerome Powell

Published March 11, 2023 by Nan Mykel

Again, this was a result of the Citizens United Supreme Court 2010 ruling permitting corporations to rule by exorbitant political funding. I wonder if the ruling was not very farsighted or on-purpose farsightrd by those judges who voted for releasing the reigns on corporations to buy elections rather than just the people’s power. (
From Daily Kos:

Senator Elizabeth Warren did not mince her words as she undressed Fed Chair for his blatant disregard for the American worker as he puts their jobs on the line Elizabeth Warren is on point.

I watch several minutes of the Federal Reserve Chair, Jerome Powell. He was discussing monetary policy and the economy with the Senate Banking Committee. While watching both Democratic and Republican Senators questioning the Chair, only Elizabeth Warren broke the undeserved praise and respect for a man that cared nothing about the middle class and poor America. She made him acknowledge the modus operandi of our economic system. It depends on having a certain percentage of people unemployed and suffering to ensure a stable economic system for the wealthy.

The Fed Chair is concerned about a stubborn 6%+ inflation rate. The strong employment market seems to concern him as well. A country with strong employment is a problem for the Fed, and our corporations, who want a heavy supply of unemployed workers from which to choose since, in a supply and demand world would accept pittances for wages.
Representative Katie Porter (D-CA) recently provided an analysis that proved most of our inflation was a result of corporate greed. They raised prices because they could. That is the definition of inflation. One would think the Fed Chair would advise Congress that if they did not want him to use the only draconian tool he has, raising interest rates, that they would create policies that stopped corporate attacks on those who put them into power. But he did not.

The purported worker shortage is a political choice. There are millions of immigrants who want to enter the United States for economic and political reasons and who are ready to work. Yet, many in our government use them as a political football that limits employment shortfall.

While corporate greed is responsible for more than half of the inflation picture, they are responsible for even more. Their incompetence and poor decision making is what we are all paying for now. They decided to offshore our manufacturing and at the same time create just-in-time-inventory, which means shortages occur whenever there is a shock in the system, like pandemics or natural disasters. They profit from those decisions, and they profit when their incompetence fails.

From <https://www.dailykos.com/stories/2023/3/11/2157480/-Sen-Elizabeth-Warren-slammed-Fed-Chair-Powell-

Fun Reads, in my opinion

Published March 10, 2023 by Nan Mykel

If you took my advice and added the still free subscription to 3 Quarks Daily then you know about this one:

Recently they carried this article from the Nautilus magazine. (You can read 2 articles in Nautilus for free, then subscribe if you like.:
“What Plants Are Saying About Us” by Amanda Gefter in the Nautilus magazineQuanta magazine via 3 Quarks Daily

_______

I liked this one also, from Quanta (not Quora)

“A Link Between Hearing Voices and Hearing Your Own Voice”
An experiment with bone-conduction headphones suggests a way for neuroscientists to better understand some hallucinations.
People using bone-conduction headphones in a study were more likely to correctly identify their own voices from a recording in which they were blended with other voices.Credit…Michael Sohn/Associated Press
By Veronique Greenwood. Feb. 14, 2023
3 MIN READ
It is the rare person who likes hearing their own voice on a recording. It sounds fake, somehow — like it belongs to someone else.
For neuroscientists, that quality of otherness is more than a curiosity. Many mysteries remain about the origins of hallucinations, but one hypothesis suggests that when people hear voices, they are hearing their own thoughts disguised as another person’s by a quirk of the brain.

I’d really like to have someone to talk with about these articles but my interests differ from those around me. Old-fashioned am I. Smile.

Evolution’s Tipping Point?

Published March 5, 2023 by Nan Mykel

Many of us thought the 21st century would be the Age of Aquarius, not evolution’s tipping point:

The older generation knows the “The Age of Aquarius”, as a song by the 5th Dimension in 1969. This is exactly what they were telling years before us we were coming into this new age. They got this information from old writings and prophecies. They knew these kids would come to us to help us heal. From (Spring Rayne)

The dictionary’s definition of tipping point is:
“the point at which a series of small changes or incidents becomes significant enough to cause a larger, more important change”.

Life-saving abilities, honed by evolution to foster survival, are being short-circuited in many areas. A look at various AI entries on Google just today is alarming. Do see for yourself…

Tech giants are heralding ChatGPT as revolutionary…. With millions of users, the chatbot has started an A.I. arms race. Companies are rushing to release their own chatbots, and some seem eerily human.
Beyond the excitement, the technology’s possibilities can feel scary — as if science fiction has become reality. ChatGPT has already inspired many people to ask: Will A.I. take my job? It’s a familiar panic, one that resurfaces every time a groundbreaking innovation emerges, like the car or the internet. Still, questioning how A.I. could replace jobs in the future misses a more urgent point: The platform is changing how people work right now.

No longer the future….NOW:

<https://www.forbes.com:
Between them, some of the world’s biggest tech companies have [already] collectively laid off more than 150,000 workers in recent months.
The Real Reasons For Big Tech Layoffs At Google, Microsoft, Meta, and Amazon: In truth, it isn’t likely to be because the companies involved need money. Microsoft MSFT +1.7%, which is reported to have laid off around 10,000 employees, practically simultaneously announced that it plans to invest $10 billion in OpenAI, the creators of the viral application ChatGPT. It seems likely that there is a business reason at the heart of the decision to invest a sum that would equate to $1 million per laid-off employee in an AI company.

Likewise, Google’s parent company Alphabet GOOGLE announced plans to reduce its global headcount by 12,000 – a cut of around 6%. CEO Sundar Pichai has previously described AI as the most transformational technology of all time, and in making the layoffs, stated that the strategy will be to “direct our talent and capital to our highest priorities.” It’s widely thought that Google is working on its own AI-powered answer to ChatGPT that will be announced soon.
From

Folks are asking a machine what the weather outside is like; depending on a GPS to guide them to their destination;and apparently enjoying the auxiliary help. They’ve even made progress in robot tv anchors. And my college graduate granddaughter cannot read cursive handwriting.

I’ll eat my words if AI succeeds in killing Climate Change, but with a decrease in smarts, manners, education and brotherly love, our much earlier and more primitive underpinning of kinship selection makes a happy Aquarious future less likely.

AND MANY STATES WANT, ARE GETTING, AND REWARDING MORE BABIES. Poor babies, poor mothers and poor us.

Mock Paper Scissors

The Internet's Band of Incorrigible Spitballers® and Cult Failure Since 2006

Pacific Paratrooper

This WordPress.com site is Pacific War era information

Edge of Humanity Magazine

An Independent Non-Discriminatory Platform With No Religious, Political, Financial, or Social Affiliations

K E Garland

Inspirational kwotes, stories and images

Nguyễn Thị Phương Trâm

Art and Literature Beyond Borders

Thar She Blows!

"So many people are crying in their latte!" ~ Sparks

Darcy Hitchcock

Envision a sustainable future

Barbara Crane Navarro

Rainforest Art Project - Pas de Cartier !

Kate Lunsford

Reflective Writing

Rosamond Press

A Newspaper for the Arts

Ned Hamson's Second Line View of the News

Second Look Behind the Headlines - News you can use...

Aging Capriciously

Divergent Thoughts on Life, Love and Death

Some View on the World

With previous posting of "Our World" on Blogger

Filosofa's Word

Cogito Ergo Sum

Trent's World (the Blog)

Random Ramblings and Reviews from Trent P. McDonald

Catxman's Cradle

Catxman dances, Catxman spins around, leaps ....... // I sing a song, a song of hope, a song of looove -- a song of burning roses. / Synthesizer notes. // (c) 2021-22

Mapping uncertainty

When nothing is certain anything is possible

%d bloggers like this: