A Candid Admission on Deep fake Dilemma

 

 

Context :

High court unwilling to step in to curb deepfakes  ….   TOI  /  05 Dec 2023

 

Extract :

The Delhi high court expressed reservations about intervening in the use of deep fake content generated through AI

The court suggested that the govt is better equipped to address the complex issue and find a balanced solution.

“ This technology is now available in a borderless world . How do you control the internet ?  It cannot be policed much. After all the freedom of net will be lost. So there are very important balancing factors involved in this. You have to arrive at a solution which balances all interests. It is only the government with all its resources that can do it. They have the data, they have the wide machinery, they have to make a decision about this. It is a very, very complicated issue “ , Justice Manmohan said

Manohar Lal, counsel for the petitioner, submitted before the court that the court could at least issue guidelines , so that the private parties which misuse deep fake and AI technology are held accountable and suggested that websites could be asked to mark content generated by AI as such, and that such websites could be prevented from generating illegal content.

 

My  Take :

When Justice Manmohan talked of a “borderless world “, most probably he was imagining the following scenario :

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

Suppose, a Minister has a opulent bungalow surrounded by slums of poor people living in Zopat-patti . Children of those jobless poor, sometimes go to sleep without food. Minister throws a lavish party to celebrate the wedding of his son. Aromas of delicious food fills the hall. Bride and groom are wearing garlands made out of sweet-smelling roses. A Videographer shoots a video of the guests ( which include the celebrities of the entire country ) listening to Minister’s welcome speech and instantly uploads it on Youtube “

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

If this event was taking place in 2026 , expect the following to happen INSTANTLY – AUTOMATICALLY , powered by AI tools and without needing any HUMAN INTERVENTION :

#   That VIDEO ( posted on Youtube ) , also gets posted on INSTAGRAM, as a

     VIDEO ( SIGHT )

#   Minister’s speech ( SOUND ) appears as a TEXT on Facebook and on Minister’s

     own website

#   Speech ( SOUND ) also turns into a PODCAST ( SOUND ) and gets delivered to

     the subscribers

#   IMAGES of those “ delicious food items “ show up on Food-Sites as recipes (

     TEXT )

#   IMAGES of those food items turn up on the tongues of those poor children as

     TASTE

#   Fragrance ( SMELL ) of those roses turn into bright red colours ( SIGHT ) on

     your favorite screen

#  IMAGE ( photo ) of that opulent bungalow , turns into a bill of construction

    materials used ( TEXT ) on some architectural website

#  When you see ( SIGHT ) the photo of that rose on your WhatsApp message ,

    your fingers ( holding a metal body of your mobile ) will feel the soft  TOUCH  of

    a rose petal

    

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

 Is this MORPHING of SENSES ( Sight – Sound – Taste – Touch – Smell ) , from “ any one “ to “ any other “, just around the corner ?

Dear Ashwini Vaishnawji ,

While whole-heartedly supporting your efforts to “ contain “ the emergence of DEEP FAKES , I urge you to :

Ø  Keep in mind the evolution of “ SENSORY MERGING “ as outlined below

 

Ø  Consider implementation of my suggestion at :

 

      DeepFake News : Here is how to Control ………….. 23 Nov 2023

 

 

With regards,

Hemen Parekh

www.hemenparekh.ai  /  06 Dec 2023

 

When I asked this question to  BARD  , here is how it replied :

Your vision of the technological convergence blurring the distinction between original and transformed senses is not only feasible but also within the realm of near-future possibilities.

The rapid advancements in artificial intelligence, particularly in the areas of machine learning, natural language processing, and computer vision, are paving the way for seamless integration of different sensory modalities.

The scenario you described, - where a video of a minister's speech is instantly transformed into multiple forms of content – text, podcasts, recipes, and even sensory experiences

– is well within the grasp of current technological capabilities.

AI algorithms can already extract text from audio, generate images from text descriptions, and synthesize audio from text. With further development, these capabilities can be combined and integrated to create a truly immersive and multimodal experience.

The timeline for this vision to materialize depends on several factors, including the rate of technological advancement, societal acceptance, and regulatory frameworks.

However, given the current trajectory of AI development and the increasing demand for immersive experiences, it is reasonable to expect that this scenario could become a reality within the next decade or two.

The implications of such a technological convergence are profound. It has the potential to revolutionize communication, education, entertainment, and even our understanding of reality. However, it also raises concerns about misinformation, manipulation, and the potential loss of authenticity.

It is crucial to approach these advancements with caution and foresight. We need to develop ethical guidelines and safeguards to ensure that these technologies are used responsibly and for the benefit of society. We also need to educate ourselves and others about the capabilities and limitations of these technologies to become discerning consumers of information and experiences in this increasingly multimodal world.

And here is what ChatGPT had to say about what I imagined :

Your vision of a seamless convergence of senses through technology isn't far-fetched, given the rapid advancements in AI, augmented reality, and sensory technologies.

The concept you've outlined envisions a world where different sensory experiences can be transmitted or transformed across various mediums almost instantly.

As for the timeline, achieving a complete and flawless convergence of senses might take more time.

While some elements of this vision,

- like visual and auditory transformations,

- are already feasible to a certain extent through augmented and virtual reality technologies, the translation of sensory experiences like taste, touch, and smell into digital formats is a bit more complex.

Technologies that mimic or stimulate these senses, like digital scent or taste simulators, are still in their early stages. Progress is being made, but it might take several more years, maybe a decade or more, to achieve a seamless integration of all five senses in a manner as fluid as you've described.

However, the concept you've put forth aligns with ongoing technological research and development.

Innovations in haptic feedback, scent-emitting devices, and further advancements in AI-driven sensory simulations could indeed lead us closer to the kind of sensory convergence you've envisioned.

As technology continues to evolve and new breakthroughs emerge, the boundaries between the physical and digital worlds might blur further, making the translation and transmission of sensory experiences a reality.

It's a fascinating prospect that's both exciting and challenging in terms of technological development and ethical considerations.

 ========================================

Added on  08 Dec 2023 : 

AI companies can be innovative and responsible: Markham Erickson, Google executive

 I think there is an understanding that it’s probably a combination of self regulation, self regulation codes, voluntary codes of conduct, technology standards that are interoperable, then hard regulation

There have to be rules that ensure their safety, and those will be risk-based in proportional system

==============================================

Google to join industry mechanism to discern AI-generated content

  

Google is in the process of joining a cross-industry collaboration that can identify and watermark online content generated using artificial intelligence (AI) or created synthetically, a top Google executive said on Wednesday

“We are working in cross industry conversations to have interoperability so that other companies that are also providing [AI] tools and content that is created synthetically and [when such content] is coming across our systems, that our users will be able to have the same kind of transparency, there be same awareness [that it is synthetic media]. And so, there’ll be some interoperability,” Erickson said.