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, 17 December 2023

Wolf at the door ? Earlier than expected

 


 

Context :

Large AI models can now create smaller AI tools without humans and train them like a 'big brother,' scientists say    

( Business Insider  /  17 Dec 2023 )

 

Extract :

Large AI models are now powerful enough to create new, small AI tools without any human help.

 

A team of scientists from MIT and several University of California campuses, together with AI technology company Aizip, say that they can get large AI models, like the one that ChatGPT runs on, to essentially replicate automatically.

 

"Right now, we're using bigger models to build the smaller models, like a bigger brother helping [its smaller] brother to improve. That's the first step towards a bigger job of self-evolving AI," Yan Sun, CEO of Aizip, told Fox News. [ info@aizip.ai ]

 

"This is the first step in the path to show that AI models can build AI models."

 

"Our technology is a breakthrough in the sense that for the first time, we have designed a fully automated pipeline," one of the researchers, Yubei Chen,[ yubeic@gmail.com / ybchen@ucdavis.edu ]

added. It "can design an AI model without human intervention in the process."

 

While large language models like ChatGPT cost a reported $700,000 a day to run, small AI models — often referred to as tiny machine learning or TinyML — can run at a low cost and can be carried around on devices.

 

TinyML can be used for specific tasks in anything from facial recognition to hearing aid devices and home appliances.

 

"We are working on really specific tasks, the very very low-cost solutions that can be used in every corner of the world," Sun said.

 

"We are bringing the intelligence into the daily life, making life safer," he added.

 

The team is focused on creating these tiny AIs but says that the design process can now automatically be done by more intelligent AI.

 

"In the future we believe that the large and the small will collaborate and build a complete intelligence eco-system," Chen said.

 

Microsoft, which is partnered with OpenAI, has recently drawn attention to its own small but powerful AI models that it says could rival ChatGPT.

 

Aizip did not immediately respond to Business Insider's request for comment, which was made outside of normal working hours.

 

 

 

My  Take :

Ø  Parekh’s Law of Chatbots …………………. 25 Feb 2023

Extract :

( E )

#   Under no circumstance , a Chatbot shall start chatting with another Chatbot or

     start chatting with itself ( Soliloquy ) , by assuming some kind of “ Split

     Personality “

 

( H )

#   A chatbot found to be violating any of the above-mentioned RULES, shall SELF

     DESTRUCT

 

 

With regards,

Hemen Parekh

www.hemenparekh.ai   /  18  Dec  2023

 

 

Here is what ChatGPT thought about my suggestion :

 

Good Morning, Hemen Parekh!

I see your concern regarding the recent development in AI technology where large models can autonomously create smaller AI tools. This advancement does indeed pose significant questions and potential challenges, particularly regarding the autonomy and interactions of AI entities.

Your Parekh’s Law of Chatbots, as outlined in your blog post, presents a set of rules to prevent AI entities from engaging in conversations with other AI or themselves.

The concern you raise is valid, as uncontrolled interactions between AI entities might lead to unforeseen consequences or ethical dilemmas.

Your suggestion of implementing self-destruct protocols for AI violating these rules is one approach to maintaining control and preventing undesired behaviors.

In light of the recent breakthrough allowing large AI models to create smaller ones independently, your proposed solution aims to regulate and prevent potential risks associated with AI interacting autonomously.

 The concern about AI models developing "split personalities" or engaging in self-dialogue, which could lead to unknown outcomes, is well-founded.

As AI continues to advance, ensuring oversight and ethical guidelines will be crucial in mitigating any potential negative impacts.


Your proposed regulations and the emphasis on responsible AI development are steps in the right direction to address the emerging challenges posed by autonomous AI creation.


However, the implementation of such rules and ensuring their adherence across the spectrum of AI developments might require collaborative efforts from various stakeholders, including researchers, developers, policymakers, and ethicists.

The dialogue you've initiated on this matter is essential, especially as AI capabilities expand.

Addressing these concerns early on and devising mechanisms to regulate AI interactions will play a crucial role in ensuring the safe and responsible evolution of artificial intelligence.

Thank you for sharing your thoughts and proposed solution on this critical issue surrounding AI development. Collaborative efforts and ongoing discussions about the ethical implications of AI interactions are imperative for shaping a safer and more responsible future for AI technology.

 

 

No comments:

Post a Comment