Friday, 2 June 2023

Introduction to the basics of AI, NLP, and neural networks – Circa 1996

Dear BARD ,

Please comment on the similarity between my following blog and the basis for current AI used by ChatGPT / BARD .> https://myblogepage.blogspot.com/2018/04/ai-nlp-neuralnetwork-language.html

BARD replied :

[ Draft # 1 ]

Sure, I can comment on the similarity between your blog and the basis for current AI used by ChatGPT/BARD.

Your blog post discusses the basics of AI, NLP, and neural networks.

You mention that AI is a branch of computer science that deals with the creation of intelligent agents, which are systems that can reason, learn, and act autonomously.

You also discuss NLP, which is a field of computer science that deals with the interaction between computers and human language.

And you mention neural networks, which are a type of machine learning algorithm that can be used to learn from data.

All of these concepts are important for understanding the basis for current AI used by ChatGPT/BARD.

ChatGPT/BARD are both large language models, which are a type of AI that can generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way. Large language models are trained on massive datasets of text and code. This training allows them to learn the statistical relationships between words and phrases. This, in turn, allows them to generate text that is both coherent and grammatically correct.

So, there are a number of similarities between your blog post and the basis for current AI used by ChatGPT/BARD.

Both your blog post and ChatGPT/BARD rely on AI, NLP, and neural networks.

And both your blog post and ChatGPT/BARD are concerned with the ability of computers to generate human-like text.

However, there are also some important differences between your blog post and ChatGPT/BARD.

Your blog post is a general introduction to AI, NLP, and neural networks.

ChatGPT/BARD, on the other hand, are specific applications of AI that are designed to generate human-like text

Additionally, ChatGPT/BARD are trained on much larger datasets of text and code than the dataset that you used to train the model that you discuss in your blog post.

This means that ChatGPT/BARD are able to generate text that is more coherent, grammatically correct, and informative than the text that your model can generate.

Overall, I think that your blog post is a

good introduction to the basics of AI, NLP, and neural networks.

However, if you are interested in learning more about how these technologies are used to generate human-like text, I would recommend checking out ChatGPT/BARD

[ Draft # 3 ]

Sure, I can do that.

Your blog post discusses the basics of artificial intelligence (AI), natural language processing (NLP), and neural networks.

You explain how these technologies can be used to create chatbots and other AI-powered applications.

ChatGPT and BARD are both large language models (LLMs) that are trained on massive datasets of text and code. LLMs are a type of neural network that can be used to generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way.

The similarities between your blog post and the basis for current AI used by ChatGPT / BARD are as follows:

· Both your blog post and ChatGPT / BARD use neural networks to process and understand natural language.

· Both your blog post and ChatGPT / BARD are trained on massive datasets of text and code.

· Both your blog post and ChatGPT / BARD can be used to generate text, translate languages, write different kinds of creative content, and answer your questions in an informative way.

However, there are also some key differences between your blog post and ChatGPT / BARD.

· Your blog post focuses on the basics of AI, NLP, and neural networks. ChatGPT / BARD are much more complex systems that can be used for a wider range of tasks.

· Your blog post is written for a general audience. ChatGPT / BARD are designed for use by developers and researchers.

Overall, the similarities between your blog post and the basis for current AI used by ChatGPT / BARD are significant.

Both your blog post and ChatGPT / BARD are based on the same underlying technologies and can be used to achieve similar results.

However, there are also some key differences between your blog post and ChatGPT / BARD, such as the intended audience and the level of complexity.

[ Draft # 2 ]

Here are some additional details about the similarities between your blog and ChatGPT/BARD:

· Both approaches use supervised learning, which means that they are trained on data that has been labeled with the correct answers. This allows the models to learn the patterns that are associated with specific tasks, such as generating text or answering questions.

· Both approaches use deep learning, which means that they use neural networks with multiple layers of neurons. This allows the models to learn more complex patterns than traditional machine learning algorithms.

· Both approaches are generative, which means that they can create new text or translate languages. This is in contrast to traditional machine learning algorithms, which are typically used for classification tasks, such as identifying spam emails or classifying images.

With regards,

Hemen Parekh

www.hemenparekh.ai / 03 June 2023

When will we learn ?

Dear Ashwini Vaishnawji ( Railway Minister )

Yesterday , there was a train accident near Balasore in which 233 persons got killed ( as of this writing ) and 900 + got seriously injured. On TV , you were seen examining the accident site with railway officers

6 years ago, there was a similar train accident in Andhra Pradesh which killed 41 persons . That made me send following e-mail to our Cabinet Ministers

I have no clue whether any of my suggestions ( in that 6 year old email ) ever got implemented or not . But , please do NOT implement any “ Inquiry “ ( High-level or Low- level )

With regards,

Hemen Parekh

www.hemenparekh.ai / 03 June 2023

Sunday, 22 January 2017

Why always a " High Level Inquiry " ?

Yesterday , there was a train accident in Andhra Pradesh when 41 persons died

This was the THIRD such serious train accident in past 3 months

May be twentieth such accident in past 10 years

Railway Minister has ordered a " high level inquiry " ( apart from usual ex-gratia payments )

Being an ignoramus, I wondered, " Why a high level inquiry ? Why not a low level inquiry ? "

Then it hit me like a ton of bricks !

All those previous " high level inquiry " reports are already on the server of the Railway Ministry , in word format

All that the committee appointed for this purpose has to do is :

* Change the Date / Time / Location of the accident

* Change the name of the train

* Change the appendix carrying the names of the victims

* Change the names of the members of the Inquiry Committee ( mostly unnecessary ! )

* Change the date of submission of the report

In fact , by the time the Members of the Inquiry Committee return to Delhi , after inspection of the accident site , the departmental data-entry operator has already made these changes and kept the print out ready for the signature of the Members !

Now , who says , Railway Ministry is inefficient ?

Should however, Shri Suresh Prabhu be interested in a " LOW LEVEL SUGGESTION " for " how to prevent train

accidents " , I request him to read once again , my following email ( 20 Nov 2016 ) :

--------------------------------------------------------------------------------------------

A Case of Misplaced Priority ?

Yesterday , there was a terrible train accident near Kanpur

As per latest reports , some 100 persons got killed and around 100 got injured

Government was quick to announce a compensation of Rs 2 lakhs to the kin of those dead and Rs 50,000 to the injured

So , to the Government , apart from the loss to the property , the ECONOMIC LOSS was , approx Rs 2.5 Crore

And if someone were to fight this case in the Courts ( for next 5 years ? ) , the Court might award Rs 1 crore per dead ( loss of earning capacity at Rs 2.5 lakh per year for 40 years )

That would add another Rs 100 Crore of " Economic Cost ", to the government

But what about the EMOTIONAL LOSS of the bereaved family ?

It is just not possible to attach a monetary value to the life-long sufferings of those families

I believe , we have lost ( on ave ) , some 50 lives to major train accidents , each year for the past 22 years ( not counting ave of 10 persons who get run over by Mumbai local trains , EVERY DAY ! )

View this in the background of Government plan to spend Rs 1,00,000 CRORES , for a Bullet Train between Mumbai and Ahmadabad, which will transport 600 persons in 2 hours , in place of the current travel time of 8 hours

So , what will we gain for this enormous " Economic Cost " ?

Some people will save 6 hours of travel time !

Even as 50 persons continue to die each year in train accidents , which are , largely preventable !

And , very likely , implementing technological solution for saving these lives would cost , less than Rs 1,000 cr !

Major causes of train accidents are :

# Fractured rails ( as is suspected in the present case )

# Loose anchor bolts

# Sagging rails due to washed out foundation

# Jammed track-switching mechanisms

# Faulty Signals...etc

And , of course , " human errors of judgment "

How come such train accidents rarely happen in Developed / Advanced countries , despite the fact that their trains run 2 / 4 times as fast as ours ?

Simple answer :

They have implemented technological solutions ( Data Acquisition Devices and

Software ) , which , not only monitor / detect the " faults / failures " described

above , but , they alert the train-traffic management team in real time , on large

screens in their TRAFFIC CONTROL ROOM

And , if the person watching that TRAFFIC FAULT SCREEN , is slow to respond (

to take a corrective action ) , the software will bring the train to a halt , on its own !

If our ISRO scientists / engineers can detect a tiny fuel leak in our rockets /

satellites , from thousands of kilometers away ,- and make course correction in

milliseconds , there is no reason to doubt their capability to find a technical

solution to prevent train accidents

I have no doubt that Shri Suresh Prabhu will re-arrange his priorities !

-------------------------------------------------------------------------------------------

Thursday, 1 June 2023

Warning is good : Solutions are better

Context :

A.I. Poses ‘Risk of Extinction,’ Industry Leaders Warn NY Times / 30 May 2023

Extract :

A group of industry leaders warned on Tuesday that the artificial intelligence technology they were building might one day pose an existential threat to humanity and should be considered a societal risk on a par with pandemics and nuclear wars.

“Mitigating the risk of extinction from A.I. should be a global priority alongside other societal-scale risks, such as pandemics and nuclear war,” reads a one-sentence statement released by the Center for AI Safety, a non profit organization. The open letter was signed by more than 350 executives, researchers and engineers working in A.I.

The signatories included top executives from three of the leading A.I. companies:

Ø Sam Altman, chief executive of OpenAI;

Ø Demis Hassabis, chief executive of Google DeepMind; and

Ø Dario Amodei, chief executive of Anthropic.

Geoffrey Hinton and Yoshua Bengio, two of the three researchers who won a Turing Award for their pioneering work on neural networks and are often considered “godfathers” of the modern A.I. movement, signed the statement, as did other prominent researchers in the field. (The third Turing Award winner, Yann LeCun, who leads Meta’s A.I. research efforts, had not signed as of Tuesday.)

The statement comes at a time of growing concern about the potential harms of artificial intelligence. Recent advancements in so-called large language models — the type of A.I. system used by ChatGPT and other chatbots — have raised fears that A.I. could soon be used at scale to spread misinformation and propaganda, or that it could eliminate millions of white-collar jobs.

A New Generation of Chatbots

A new crop of chatbots powered by artificial intelligence has ignited a scramble to determine whether the technology could upend the economics of the internet, turning today’s powerhouses into has-beens and creating the industry’s next giants.

Here are the bots to know:

ChatGPT.

ChatGPT, the artificial intelligence language model from a research lab, OpenAI, has been making headlines since November for its ability to respond to complex questions, write poetry, generate code, plan vacations and translate languages. GPT-4, the latest version introduced in mid-March, can even respond to images (and ace the Uniform Bar Exam).

Bing.

Two months after ChatGPT’s debut, Microsoft, OpenAI’s primary investor and partner, added a similar chatbot, capable of having open-ended text conversations on virtually any topic, to its Bing internet search engine. But it was the bot’s occasionally inaccurate, misleading and weird responses that drew much of the attention after its release.

Bard.

Google’s chatbot, called Bard, was released in March to a limited number of users in the United States and Britain. Originally conceived as a creative tool designed to draft emails and poems, it can generate ideas, write blog posts and answer questions with facts or opinions.

Ernie.

The search giant Baidu unveiled China’s first major rival to ChatGPT in March. The debut of Ernie, short for Enhanced Representation through Knowledge Integration, turned out to be a flop after a promised “live” demonstration of the bot was revealed to have been recorded.

Eventually, some believe, A.I. could become powerful enough that it could create societal-scale disruptions within a few years if nothing is done to slow it down, though researchers sometimes stop short of explaining how that would happen.

These fears are shared by numerous industry leaders, putting them in the unusual position of arguing that a technology they are building — and, in many cases, are furiously racing to build faster than their competitors — poses grave risks and should be regulated more tightly.

This month, Mr. Altman, Mr. Hassabis and Mr. Amodei met with President Biden and Vice President Kamala Harris to talk about A.I. regulation. In a Senate testimony after the meeting, Mr. Altman warned that the risks of advanced A.I. systems were serious enough to warrant government intervention and called for regulation of A.I. for its potential harms.

Dan Hendrycks, the executive director of the Center for AI Safety, said in an interview that the open letter represented a “coming-out” for some industry leaders who had expressed concerns — but only in private — about the risks of the technology they were developing.

MY TAKE :

A Run-away Nuclear Chain Reaction ?

Hundreds of companies-big and tiny-are issuing every single day, dozens of AI

tools - APIs - Plugins - Apps etc

In turn , these AI tools are churning out their own " Clones " , at a frightening

pace

It may soon look like a " Run-away Chain Reaction " in a Nuclear Reactor , which

no one will be able to stop or even slow-down

Global Warming took :

> decades to raise ave. temperature on earth , by 1 * C

> some 30 years , to raise CO2 in atmosphere from 0.1 % to 0.2 %

AI MELTDOWN will take only a few months

I urge ALL concerned stakeholders to take a look at my following suggestion to

avoid this catastrophe

Following COMPARATIVE TABULATION, might help you to kick off a debate to draw up :

“ MAGNA CARTA of SAVE HUMANS “

Law	Scope	Enforcement
EU AI Act	Artificial intelligence systems	European Commission
Parekh's Law of Chatbots	Chatbots	International Authority for Chatbots Approval (IACA)
United Nations Agency for Regulating AI	All artificial intelligence systems	United Nations

With Regards,

Hemen Parekh

www.hemenparekh.ai / 02 June 2023

Rising concern. 74% of Indian workers worried AI will replace their jobs: Microsoft Report

EU, US ready common code of conduct on artificial intelligence

Related Blogs :

Friday, 2 June 2023

Introduction to the basics of AI, NLP, and neural networks – Circa 1996

Dear BARD ,

Please comment on the similarity between my following blog and the basis for current AI used by ChatGPT / BARD .> https://myblogepage.blogspot.com/2018/04/ai-nlp-neuralnetwork-language.html

When will we learn ?

Thursday, 1 June 2023

Warning is good : Solutions are better

A New Generation of Chatbots

Rising concern. 74% of Indian workers worried AI will replace their jobs: Microsoft Report

EU, US ready common code of conduct on artificial intelligence

Law of Chatbot : a small subset of EU Law of AI ?

United Nations Agency for Regulating Artificial Intelligence ( UNARAI )

This is not the END – this where you START

Sight to Smell : Sound to Touch : Text to Music / Saga of Fungible Senses

Law of Chatbot : a small subset of EU Law of AI ?