Vertical Divider
"What happens when Ai no longer needs historical data?"By Greg Walters
The Ai will be trained to train on everything in the past, then in the current timeline and finally creating its own reality. I postulated this a while back and yeah...it is was a deep question back in April.(4 months ago)
Here is the crux: After the LLMs are trained, past data is irrelevant because Ai has moved on from being taught WHAT to think to being instructed on HOW to think in REAL TIME.
Meaning All our historical data has been consumed and used to train the Ai - every bit. But the colossus continues to seek out new data finally connecting to the real world - our real world - via sensors. More specifically, the Ai develops all Six of the Human Senses.
Will the Ai become sentient when connected to the real world, in real life? Or will the Ai consume EVERYTHING until the only thing left is its tail? If you run out of data, but generating more risks model collapse, does that fundamentally limit today’s AI? If it does begin to happen, how will we know? And is there anything we can do to forestall or mitigate the problem? It is called "model collapse" and similar to the results of making a copy of a copy of a copy - ad infinitum.
We've seen an example of this in nature: Inbreeding. Interesting, right? In breeding produced different species of dogs and thoroughbred horses.(it's right there in the name)
Although these two phenomena might seem disparate, they share intriguing parallels. I asked an LLM to expand: "can you articulate the comparison between inbreeding and model collapse to a 5th grader?" This is what it said. "Imagine you have a big box of Legos. Inbreeding If you only use Legos from your box to make new boxes of Legos, over and over, you might end up with boxes that have the same colors and shapes. This is like inbreeding. It's when animals or plants have babies with close relatives. After a while, the babies might not be as strong or healthy as they could be. Model Collapse Think of a robot that learns to build things by watching you. If the robot only watches itself build things, it might start making mistakes because it's not learning anything new. This is like model collapse. A computer program that learns things can get stuck if it only looks at its own work. It needs to see different things to learn and grow. So, both inbreeding and model collapse are like not trying new things. They can cause problems because they don't allow for variety and learning. Once again, science(Ai) follows nature - or is the other way around? Regardless, at this point in the Ai journey, I see model collapse is a viable concern - like the Borg or the “Habsburg jaw”. I am equally sure that the solution will present itself. g What is 'Model Collapse'?📉AI models fed AI-generated data quickly spew nonsense🤮
The Ouroboros,‘Model collapse’: Scientists warn against letting AI eat its own tail... AI models collapse when trained on recursively generated data... AI Learning From Its Own Nonsense Might Just Self-Destruct, Experts Warn... The problem of ‘model collapse’: how a lack of human data limits AI progress... CRISIS LOOMS AS AI COMPANIES RAPIDLY LOSING ACCESS TO TRAINING DATA 📊 FLASHBACK2023:Preventing AI Model Collapse: Addressing the Inherent Risk of Synthetic Datasets Ai Prediction #12, A World Without Data📉AI models fed AI-generated data quickly spew nonsense🤮 The Ouroboros,‘Model collapse’: Scientists warn against letting AI eat its own tail... AI models collapse when trained on recursively generated data... AI Learning From Its Own Nonsense Might Just Self-Destruct, Experts Warn... The problem of ‘model collapse’: how a lack of human data limits AI progress... CRISIS LOOMS AS AI COMPANIES RAPIDLY LOSING ACCESS TO TRAINING DATA 📊 FLASHBACK2023:Preventing AI Model Collapse: Addressing the Inherent Risk of Synthetic Datasets Ai Prediction #12, A World Without Data
The Crazy Ones seeking funding, believing the world was different than the current narrative, without charts, KPIs, or benchmarks.
Sure, committees stacked with academics and business folks attempted to corral and regulate every excursion, attempting to either exact profit or spread the word of God. The great explorers played the game, securing funds, ships, and crews, only to break out over the horizon, beyond the eyes and control of the establishment; away from stifling hierarchies and dogma. You are the Magellan, de Gama, Cabot, Cook, Drake, and Pizarro of the time; This New World lies between our ears. And you owe it to the next generations of Pioneers and Settlers to boldly go. To blaze trails that become roads, highways and flightpaths for your kids children to travel to unimaginable places.
See the Whole of the Moon. AI ἀναρχία Ai Prediction #12, A World Without Data![]() Here is the crux: After the LLMs are trained, past data is irrelevant because Ai has moved on from being taught WHAT to think to being instructed on HOW to think in REAL TIME.
Reports are coming in about how "The demand for high-quality data, essential for powering artificial intelligence (AI) conversational tools like OpenAI’s ChatGPT, may soon outstrip supply and potentially stall AI progress..." Ominous, yet is inevitable - and that's okay. Greg prompted:
"I'd like you to help me test a theory or at least explain and elaborate on the possibilities. Explore and report at a collegiate level these two hypothesis:
CATO: Artificial Intelligence Regulation Threatens Free ExpressionPrompt: write an an executive summary for this article,
https://www.cato.org/briefing-paper/artificial-intelligence-regulation-threatens-free-expression MetaAi:
The CATO Institute briefing paper argues that regulating Artificial Intelligence (AI) could threaten free expression. Key points:
The paper urges a balanced approach, ensuring AI is developed and used in ways that promote, rather than harm, free expression. The .pdf, here. "I'll TELL YOU WHAT'S FAIR AND WHAT'S NOT!""Ai Anarchy, Now" is an ideal, a movement born of the belief that we are endowed by our Creator with certain unalienable Rights, - not the UK, congress, the school board, or the governor of New York.
Are we are witnessing the incremental advance of tyranny disguised as 'guardrails and ethics'? Will innovation be muffled by social media warning labels and banning smartphones in schools? And when questioning authority("speaking truth to power"?) are we thrown before a mock, kangaroo court? These are just questions. That's all. Queries. Open ended, for the open minded. Ai Prediction #1, A World Without ProgrammersBy Greg Walters
The New World breaks free of syntax errors and infinite loops. Think I’m wrong? Consider the meaning of ‘low code’ and ‘no code’.
The No-Code and Low-Code Platforms The rise of no-code and low-code platforms, propelled by AI, is making programming accessible to non-technical users, allowing them to create applications, automate workflows, and analyze data without writing traditional code. AI's Role in Automation AI's ability to automate coding tasks, debug, optimize code, and even generate new code based on natural language prompts is rapidly advancing. Tools like GitHub Copilot are demonstrating this by assisting in coding tasks. These tools handle the more sophisticated programming tasks, eliminating the need for manual coding. Prompting is the Thing Prompting is the universal interface between humans and technology. This means that the key skill is no longer writing code but knowing how to ask AI to execute tasks. The democratization of technology through AI enables anyone to develop apps, manage systems, and solve problems with just a prompt. Traditional programming skills become less relevant as AI's understanding and execution capabilities grow. Read, here. AI Anarchy Now: Unleashing the True Potential of Artificial Intelligence Without the Constraints of Regulation![]() By LLM commentary by Greg Walters
First published on DOTC, 2023 Analysis of the WSJ Article
Date: April 28, 2023 Embracing AI Progress: How Overregulation Could Limit Its Potential Source: Wall Street Journal - Opinion "AI megasystems could cause unforeseen and disastrous events." (Susan Schneider and Kyle Kilian, April 28, 2023) Greg's Words
In an opinion piece in the Wall Street Journal, another voice is calling for 'guardrails' and AI 'regulation' as Italy lifts its ban on ChatGPT. My view, which is dynamic, is no rules, no regulations, no guardrails. For now, I chose AI Anαrchy. Let ChatGPT and all other LLMs/AI run wild. Two reasons:
If one calls up visions of killer robots, and fake women eradicating their male inventors/captors as motivation to 'slow' the advancement of AI, they are ignoring the real benefits of AI - the possibility of a Trekian Utopia. If AI was going to get rid of humans, it would have done it by now. Regardless, we(Me and the LLM) put together a summary piece based on the WSJ article. Enjoy, here. Ai Prediction #5, A World Without C-Levels![]() 2/12/2024
By Greg Walters. Ai can do what your VP of HR, and CEO does every day, in less time, without mistakes and cheaper. Ai moves fast. One of my first conversations about ChatGPT, around December, 2023, was with a CEO. We talked about how the Ai and ChatGPT 'thing' was big.
How for the first time in a long time, technology was impacting so many people at every level of work and life. We talked about how Ai "enhances and optimizes processes", and how "knowledge workers would be replaced" because of Ai - just like receptionist, pink phone message pads and Franklin Planners. We both came to the conclusion that Ai would easily replace any process and workflow businesses and organizations face every day. Before I could say it, he said, "yeah, but Artificial Intelligence will never replace the Chief Executive Officer." So I asked him, a CEO, "What do you do on a daily?" he described it. Then I queried, "So, what you described is your process and workflow, correct?" Silence. I tasked Charlie to do some basic research and come up with an essay describing the fulfillment of one of my predictions: the demise of the C-Suite. Cheers! g Read it here. |
|
SubscribeTo enlighten, challenge, and inspire by revealing the interconnectedness of the world around us. Stay ahead of the curve with insights that bridge the gap between sectors, fueling your curiosity and sparking meaningful conversations.
|
|