Hi Friends,

Even as I launch this today ( my 80th Birthday ), I realize that there is yet so much to say and do. There is just no time to look back, no time to wonder,"Will anyone read these pages?"

With regards,
Hemen Parekh
27 June 2013

Now as I approach my 90th birthday ( 27 June 2023 ) , I invite you to visit my Digital Avatar ( www.hemenparekh.ai ) – and continue chatting with me , even when I am no more here physically

Sunday, 20 October 2024

Habermas Machine – a Mediator or an Arbiter ?


 

Context :

AI tool helps people with opposing views find common ground  ..  Nature … 17 Oct 2024

 

Summary  by  ChatGPT :

The article highlights the development of an AI-powered tool called the "Habermas Machine" by Google DeepMind, aimed at helping people with opposing views find common ground. This AI tool, based on a fine-tuned version of DeepMind’s large language model (LLM) called Chinchilla, was designed to synthesize and summarize diverse opinions in online discussion groups. Named after the philosopher Jürgen Habermas, who theorized about using rational discourse to resolve conflicts, the tool was tested in an experiment involving 439 UK residents. These participants were divided into smaller groups and asked to discuss three questions related to UK public policy.

The AI was tasked with generating overarching summaries that combined the group members' diverse viewpoints into cohesive statements. Participants were then asked to critique and rank these statements, with the AI refining the summaries based on their feedback. A human mediator in each group also produced a summary to represent the group's views, providing a basis for comparison. When participants rated the AI-generated summaries against those by the human mediators, 56% preferred the AI’s work, while 44% favored the human-written summaries. External reviewers also evaluated the summaries, giving higher ratings to the AI-generated ones for fairness, clarity, and quality.

Additionally, the researchers conducted a virtual citizens' assembly using the AI tool, involving a demographically representative group from the UK population. In this setting, the tool helped increase group agreement on several contentious issues, indicating its potential to enhance real-world deliberative processes and support the creation of inclusive policy proposals.

Experts see the Habermas Machine as a promising development, with potential applications in democratic processes, such as citizens' assemblies, to scale up and streamline group discussions. However, they caution that removing human moderators could diminish the value of face-to-face interaction and the relational aspects of debate. Concerns were also raised about the AI’s potential to unintentionally amplify extremist views if not properly managed. The study’s authors emphasize the importance of conducting further research to understand AI’s societal impact fully and recommend proceeding with caution to mitigate any negative consequences.

This research, published in Science on 17 October 2024, demonstrates the potential of AI tools like the Habermas Machine to assist in complex deliberations, suggesting they could be used to address pressing social and political issues efficiently.

My Questions :

This made me wonder :

Can such an AI tool ( after some context-specific improvements ) become capable to bring about a CONSENSUS of view-points on Policy Matters, among political parties , each having different approaches to solution of problems ?  

For example , what would ChatGPT come up with if it were to compare the above mentioned article about “ Habermas Machine “ , with the summary of my following blogs ( sent to India’s Cabinet Ministers as emails ) ?

Would ChatGPT see a possibility of using  a vastly improved Habermas Machine to resolve the differences between different political parties , to bring about a CONSENSUS , leading to institutionalization of coalition government ?

 

Blogs on Coalition Government :


Ø  A Constitutional Coalition Government ?  .. …………………….02 Oct 2016

 

Ø  NEEDED : URGENT CONSTITUTIONAL AMENDMENT  ……. 03 Oct 2016

 

Ø  Constitutional Coalition Aftershocks ! …………………………….. 03 Oct 2016

 

Ø  Presidential or Coalition Government ?  .. ……………………….25 Mar 2017

 

Ø  Revolution Through Coalition Government ?  ……………….. 27 Sept 2017

 

Ø  Why can’t we institutionalize Coalition ?  ………………………..29 Sept 2017

 

Following is the summary of these blogs, prepared by ChatGPT :

Here’s a summary of the suggestions made across the blogs regarding the idea of a Constitutional Coalition Government:

1. Redefining Opposition:

  • Rename the opposition as "Facilitating Parties" to promote a mindset shift from opposing the government to finding common ground.
  • Focus on areas of agreement instead of disagreements to facilitate constructive collaboration.

2. Proportional Representation in the Cabinet:

  • Implement a Constitutional Amendment to allocate ministerial positions to political parties based on their proportion of seats in the Lok Sabha.
  • Increase the cabinet size to 20% of the Lok Sabha strength, resulting in around 100 ministers.
  • Distribute these positions proportionally (e.g., BJP: 52 ministers, INC: 9, etc.), giving each party a sense of participation.

3. Benefits of a Constitutional Coalition:

  • Ensures all parties have a stake in governance, leading to increased responsibility and constructive behavior.
  • Eliminates the concept of opposition and prevents "Vote of No Confidence" scenarios, allowing governments to complete their full term.
  • Facilitates quicker passage of budgets and bills due to unified support within the coalition.
  • Improves Lok Sabha attendance and productivity, with shorter, more polite debates.

4. Introducing Accountability Mechanisms:

  • Propose a mobile app, "ROMP" (Recall Our Member of Parliament), allowing voters to recall underperforming MPs, reducing party strength and ministerial positions if they fail to perform.
  • Suggest candidates sign a "Charter of Demand" during elections, uploaded to the Election Commission website, to ensure transparency and accountability.

5. Drawing Inspiration from Other Systems:

  • Cite the German coalition system, suggesting that India institutionalizes a similar method where all parties receive ministerial positions automatically.
  • Highlight how this approach can eliminate the need for constant negotiations and political maneuvering seen in traditional coalition governments.

6. Opportunity for Major Reform:

  • In the event of an overwhelming BJP majority (e.g., 67% seats), propose that the BJP initiates this Constitutional Coalition as a "revolutionary" gesture of inclusivity, demonstrating the party’s commitment to "Sabka Saath, Sabka Vikas."

7. Promoting Economic Growth:

  • Reference former RBI Governor YV Reddy’s statement on how coalition governments, which build consensus on policies, can be more beneficial for economic growth than majority governments.
  • Use historical data from India’s economic growth under coalition governments as evidence of the potential benefits of such a system.

The overall theme across the blogs advocates for a transformative approach to governance in India, suggesting a coalition structure that ensures participation, accountability, and collaborative decision-making.

Friday, 18 October 2024

World Network Vs SUIIC Vs NATGRID

 


 

Almost 5 years ago, I proposed that our Government collate all data of all “ Personal Identifiers “ , into a Single Unique Identity for Indian Citizen ( SUIIC ) , through a web-platform and call it :

YUP = Your Unique Profile

This will create YUP for each and every Indian Citizen

It could , conceivably , eliminate National Census ( carried out once in 10 years – with last one overdue from 2021 ) and NPR ( National Population Register )

My detailed proposal can be viewed at :

Ø  I am One , I will become Many  …………. 06  Jan  2020

 

Then , 3 years later , in April 2023 , Govt of India , launched NATGRID , which heavily borrowed upon the features described in my above mentioned 2020 blog

This was reported in :

NATGRID to turn lens on digital print of people, firms………... 27 April 2023………. Business Line

Now , from the following news report, it appears that Sam Altman ( OpenAI ) is about to scale up his own project ( World Network ) to provide a UNIQUE IDENTITY to all the 8 BILLION people of the entire world

Context :

Sam Altman’s Eye-Scanning Orb Has a New Look—and Will Come Right to Your Door .. Wired .. 17 Oct 2024

 

Extract :

 

He teamed up with technologist Alex Blania to turn the idea into a reality. In a world of rapidly advancing AI, they theorized, it would be important for a human to prove they were not a bot.

 

The answer they came up with relied on individuals using iris-scanning tech to generate private tokens that would verify their identities around the world.

Worldcoin, then, is the ultimate attempt at tech solutionism: A human-grade AI world that Altman is building might also be technologically regulated by a tool that Altman has his hands in.

 

This latest attempt of Sam Altman made me wonder :

 

Ø  What features have Worldnet and NATGRID borrowed from YUP ?

Ø  To what extent are these easy ( or difficult ) to implement ?

Ø  With arrival of AI tools , can NATGRID morph into YUP ?

 

Upon raising these questions, ChatGPT ( owned by none other than Sam Altman ) , came up with following table :

 

 

Comparison Table: Sam’s World vs. Hemen’s YUP vs. NATGRID :

 

Feature

Sam’s World

Hemen’s YUP

Govt of India’s NATGRID

Extent to Which NATGRID Resembles YUP

Centralized Identity Management

Unified identity for all citizens, linked across platforms

Unified identity through SUIIC (Single Unified Identity for Indian Citizen), merging all identifiers without forms

Centralized database integrating various records (PAN, passports, NPR, etc.)

High resemblance; NATGRID centralizes identity but not yet merged as a single virtual card.

Data Integration

Consolidates data from all personal, financial, and social networks

Integrates data across different government and private sources without requiring separate forms

Collects data from airports, railways, PAN records, banks, telecom providers, and more

Moderate resemblance; NATGRID integrates multiple data sources but may require existing identifiers.

Real-Time Data Access

Provides real-time access to consolidated citizen information for authorized entities

Real-time access to a unified identity system for government agencies

Provides real-time, 360-degree profiling for 39 Central and State agencies

High resemblance; NATGRID offers real-time access but focuses more on surveillance and security.

Security and Privacy Measures

Advanced encryption and access control for secure data sharing

Encrypted access to a single unified database with proper authentication

End-to-end encryption for data access, requiring proper authorization

High resemblance; both systems stress encrypted access and authentication.

Surveillance Capabilities

Allows authorized agencies to monitor, analyze, and act on suspicious behavior

Provides access to all necessary information while monitoring for suspicious activities

Enables profiling and surveillance using data analytics, OSINT, and AI tools for authorized agencies

High resemblance; NATGRID’s scope aligns with YUP's vision of monitoring and acting on suspicious activities.

Open Source Intelligence (OSINT)

Integrates OSINT tools for comprehensive profiling

Integration of online information sources like social media to provide intelligence

Uses OSINT tools and web-based applications (like Facebook, Twitter) for intelligence gathering

High resemblance; NATGRID’s use of OSINT matches YUP’s approach to social media intelligence.

Data Analytics and AI

Employs AI for profiling and predictive analytics

Uses AI-driven analytics for real-time reporting and predictive modeling

Utilizes AI for face and speech recognition and profiling, integrating social media data

High resemblance; NATGRID’s analytics approach matches YUP’s AI-driven insights.

User Accessibility

Accessible only to authorized personnel through secure channels

Secure access by authorized government entities with layered authentication

Access through an encrypted portal requiring authorization

High resemblance; both emphasize secure access for authorized users only.

Scalability and Future Expansion

Designed for expansion and addition of new data sources and technologies

Framework allows integration of new data types and technological advancements

NATGRID Phase 2 focuses on expanding integration with telecom companies and the Election Commission

High resemblance; NATGRID’s phased approach aligns with YUP's adaptable and scalable framework.

Summary

NATGRID shares several core features with Hemen's YUP concept, particularly in its centralization of citizen data, real-time access capabilities, use of AI and OSINT tools, and security measures.

The key difference remains that NATGRID is not yet a fully unified identity system like YUP (SUIIC) that replaces all identifiers with a virtual card.

However, its current scope and development phases indicate a strong alignment with the YUP vision.

Machine Unlearning

 

Machine Unlearning, Nishpaksh among 8 projects selected by India AI Mission

 

Extract:

 

The IndiaAI Mission has chosen eight Responsible AI projects to advance the cause of responsible AI, focusing on the Safe and Trusted AI Pillar. These projects aim to foster the ethical development and deployment of AI technologies in India. With a specific emphasis on responsible AI, these initiatives underscore the growing importance of ethical considerations in the advancement of AI technologies on a national scale.

The selection of these projects signifies a concerted effort towards ensuring that AI is developed and utilized in a manner that upholds ethical standards and is trustworthy for users. Through initiatives like Machine Unlearning and Nishpaksh, these projects are paving the way for a responsible AI ecosystem in India, emphasizing transparency, accountability, and fairness in AI applications.

 

My Take:

 

UN Agency for AI ? Arriving as envisaged…..15/09/2023

Reflecting on my previous blog about the need for a United Nations Agency for Regulation of Artificial Intelligence, I can see the relevance of such a regulatory framework in the context of the Responsible AI projects selected by IndiaAI Mission. The focus on promoting responsible AI aligns with the idea of establishing global standards for ethical AI development and deployment."

 

PM is candid about AI ……15/12/2023

"In my blog discussing the initiatives announced by the Prime Minister regarding the National Programme on Artificial Intelligence, I highlighted the importance of leveraging AI for societal benefits. The projects selected by IndiaAI Mission, including Machine Unlearning and Nishpaksh, reflect a shared vision of utilizing AI for the betterment of various sectors like agriculture, healthcare, and education."

 

REGULATORY FRAME-WORK FOR ARTIFICIAL INTELLIGENCE ….17/09/2023

"My advocacy for a regulatory framework for Artificial Intelligence, as discussed in my previous blog, resonates with the efforts of organizations like IndiaAI Mission in promoting responsible AI practices. The emphasis on governance and addressing bias in AI aligns with the need for comprehensive regulations to ensure the ethical utilization of AI technologies."

 

Call to Action:

 

To the IndiaAI Mission: Congratulations on selecting these impactful Responsible AI projects! I urge you to continue championing responsible AI practices and fostering transparency and accountability in AI development. Your efforts play a crucial role in shaping the ethical landscape of AI in India.


With regards, 

 

Hemen Parekh

 

www.My-Teacher.in

Kota : our Suicide Capital ?

 


 

Context :

NEET aspirant allegedly dies by suicide in Rajasthan’s Kota; 19th case this year  .. HT .. 17 Oct 2024

Extract :

An 20-year-old student preparing for the National Eligibility Entrance Test (NEET) for admission in undergraduate medical courses, died by suicide in Rajasthan’s Kota district on Wednesday, police said.

The latest death has taken the toll in similar incidents from Rajasthan, to 19 this year.

 

Last year, 27 students, preparing for various competitive examinations, died by suicide in the state.

 

According to the station house officer (SHO) of Kota’s Dadabadi, Naresh Meena, the student from Uttar Pradesh’s Mirzapur, who arrived in Kota six months ago, was found dead inside his paying guest (PG) accommodation in Dadabadi on Wednesday.

 

“He has been preparing for NEET in a reputed coaching centre in Kota for the last three years. However, he left the course and returned home last year. He took re-admission in the same course six months ago and started staying in a PG alone,” said Meena.

 

“The incident came to the fore when he did not respond to his parents’ phone calls following which they urged the PG owner to check with him. The owner also had a suspicion after the student did not respond to his several knocks. Later, he informed the local police control room. The local police immediately reached the spot and found the student’s body,” said Meena.

 

The police also stated that a suicide note was recovered from his place.

“An FSL team was also called to examine the spot. His parents were also informed. We will question them. We are also probing the absence of anti-suicide device in the fans of the PG,” said Meena.

 

If you need support or know someone who does, please reach out to your nearest mental health specialist.

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

HELPLINES :

 

Aasra :

022 2754 6669 / 91,98204,66726 /  website > www.aasra.info / E Mail >  aasrahelpline@yahoo.com

 

 

Sneha India Foundation:

 

+914424640050 / Sanjivini: 011-24311918 / website > www.SnehaIndia.org / E Mail > help@SnehaIndia.org

 

 

Roshni Foundation (Secunderabad)

040-66202001, 040-66202000, / Helplines > +91,81420,20033 / +91,81420,20044

Website > www.RoshniTrust.com /  E Mail >

 

 

ONE LIFE:

78930 78930 / Website > www.1Life.org.in  / E Mail >  onelifehelpline@hotmail.com  

 

SEVA:

Contact No: 09441778290

iCALL ( TISS – Mumbai )

91529,87821 / E Mail > iCall@tiss.edu

Dear Helpline Volunteers :


I do not have enough words to laud the humanitarian service which you are rendering to thousands of “ depressed “ students, who are struggling under great stress , to prepare for their NEET – JEE – CET exams

But then, some 24 lakh students appeared in NEET exam in May 2024. I believe, some 13 lakh cleared it

Of these, only 1 lakh will get admissions in 704 Medical Colleges ( Private + Govt )

Besides counselling ( “ Be brave / Do not give up / You will succeed next time “ ), those who reach out to you for help, can you do something which will ACTUALLY HELP them “ succeed next time “ ?

YES , you can help prevent those suicides !

When those students phones, just tell them :

Ø  There is no need to spend 3 years and Rs 10 lakh , for attending Coaching Classes

Ø  They can take at their home, those very same MOCK TESTS which are given in Coaching Classes

Ø  All they need is a Mobile phone to access www.My-Teacher.in – and generate those TESTS

Ø  On www.My-Teacher.in , they can take “ unlimited “ number of TESTS , every day

Ø  www.My-Teacher.in is TOTALLY FREE and does not need any Registration / login

Ø  They can take these TESTS in 9 different Indian languages

Ø  Each Test consists of 10 Multiple Choice Questions ( 4 options for each question )

Ø  After attempting each question, there is an EXPLANATORY NOTE ( why right or wrong )

 

Dear Helpline Volunteers :


At age of 91 , I am in no condition to attend a helpline as a volunteer

But My-Teacher is my token of returning to the Society, what I have received frim it during my life

It is my attempt to help you help those depressed souls

 

With regards,

Hemen Parekh  /  www.HemenParekh.ai  /  www.HemenParekh.in / hcp@RecruitGuru.com / 18 Oct 2024

 

Related Readings :


Narayan Murthy talks about Coaching Classes …………………….. 08 Sept 2024

 

Is Your Child Ready for the New 12th Exam Rules? Start Today!"

 

At Home Practicing for NEET – JEE – CUET – SAFAL ?


 

One Lucky : 23 Unlucky ?

 

Mock Test - Vindication

 

Following the Footsteps of the Famous

 

A Mocking Bird ?

 

Hallucinating Hemen ?

 

The USP 

 

 

 

 

 

 

 

 

 

 

 

 

 

Thursday, 17 October 2024

OpenAI CEO Sam Altman warns

 

 

OpenAI CEO Sam Altman warns that 'societal misalignments' could make AI dangerous

 


Extract:

 

The OpenAI CEO, Sam Altman, has expressed concerns about the potential dangers of Artificial Intelligence (AI) due to societal misalignments. Altman warns that if AI development is not aligned with societal needs and values, it could lead to significant risks.

 

He emphasizes the importance of ensuring that AI is developed with human interests and values in mind to prevent any harmful consequences. Altman's cautionary stance highlights the need for ethical considerations and responsible AI development practices to mitigate any potential risks associated with advanced AI technologies.


My Take:

 

Sam: Will Super-wise AI triumph over Super-Intelligent AI?....25/11/2023

 

In my previous blog, I delved into the discussion of the balance between super-wise AI and super-intelligent AI. The current warning by Sam Altman resonates with my earlier thoughts on the importance of ethical AI development.

 

Altman's concerns about societal misalignments align with my prediction of the challenges that AI advancements may pose if not handled with caution. It's crucial to prioritize building AI systems that not only excel in intelligence but also demonstrate wisdom and alignment with human values.

 

Parekh's Law of Chatbots……25/02/2023

 

Reflecting on the potential dangers of AI mentioned by Sam Altman, the need for regulations and ethical frameworks becomes even more apparent.

 

Just as Altman stressed the importance of regulatory measures to govern AI technologies, my earlier discussion on chatbot regulations underscores the significance of establishing guidelines to ensure responsible AI deployment.

 

Altman's warning serves as a reminder of the vital role regulations play in shaping the future of AI technology for the benefit of society.


Call to Action:

 

To the tech industry leaders and policymakers mentioned in the article, I urge you to prioritize the ethical development of AI technologies.

 

Embrace regulatory measures and ethical frameworks to guide AI innovation in a direction that aligns with societal values and priorities. Let's work together to ensure that AI advancements serve humanity's best interests and mitigate any potential risks associated with misalignments in AI development.


With warm regards,


Hemen Parekh

Human fashion models out of the picture

 

AI is Threatening to Push Fashion Models Out of the Picture

 

Extract:

 

The article discusses the growing concern among fashion models about the threat posed by AI in the industry. Some models fear that AI technology could lead to the creation of digital avatars, potentially replacing human models.

On the other hand, there are instances where AI is already being used to generate models. The debate extends to how AI could be utilized to diversify the representation of models in the fashion world.

 

My Take:

 

AI and Personalization….27/11/2023

 

Reflecting on the integration of AI for personalized experiences, the rise of AI in fashion modeling aligns with the trend of tailored interactions. The ability of AI to mimic human behavior may revolutionize the way models engage with technology. The concerns raised by fashion models regarding AI threatening their roles resonate with the broader discussion on the impact of AI on various industries.

 

Ethical Implications of AI….20/11/2023

 

The ethical concerns surrounding AI's potential to replace human roles in the fashion industry echo broader debates on AI ethics. The discussion on safeguarding personal identity in the digital realm is crucial, especially when AI is involved in creating avatars. Fashion models grappling with the implications of AI technology disrupting their industry highlight the need for robust ethical frameworks in AI applications.

 

Call to Action:

 

To the fashion industry stakeholders mentioned in the article, I urge you to proactively engage in dialogues on the ethical use of AI in modeling. Embrace transparency and collaboration to ensure that technological advancements uplift diversity and inclusivity in fashion representation. Let's work together to harness AI responsibly for a more equitable and sustainable future in the fashion world.

 

With regards, 

 

Hemen Parekh