Craig Brown’s Post

Diffusion Models: Midjourney, Dall-E Reverse Time to Generate Images from Prompts: We have AI, physics, thermodynamics, probability, beautiful images AND time travel? What are you waiting for. Continue reading on Towards Data Science » #MachineLearning #ArtificialIntelligence #DataScience

The value of your intelligence is in its predictive capacity. The mind predicts to prevent or perform. If machine intelligence and learning is designed as artificial human intelligence embodied as humanoid AI robots, then…some existential consequences follow. One of them is about a high risk of all human job displacement. There are no emotions here, but just non-formal real logic, inference, reasoning and conclusion. The cold logic conclusion, we must discard the human-like, human-compete AI and develop machine intelligence as a non-human, human-complete, or real and true AI. Simple as that. https://lnkd.in/dJ2MbcDy

It is both, dense and ignorant, showing collective mindless behavior. It could be justified by that machine intelligence and learning is hard to understand because of its complexity of ideas. Many believe in AI as artificial human intelligence or humanoid AI, like as many believe in some supernatural entities, as gods or demons, or as we in the Soviet Union believed in communism as a classless, stateless, and moneyless society of full abundance, to be replaced by AI-enabled communism. In reality, the AI capitalism of more efficient, productive and cost-economical humanoid robots is coming with total unemployment and job loss as the least risk. It is hard to count how many AI communities exist as the networks of people with a shared interest in artificial intelligence, machine learning, deep learning and related fields. They exist online and offline in the form of social media network groups, forums, conferences, institutions, and more. Just Quora, Facebook and LinkedIn, where i participate, might have dozens of millions, with groups like “Artificial Intelligence and Machine Learning”, “Deep Learning Neural Networks”, etc. If we count all the users and subscribers of LLMs and chatbots, the number could be the hundreds of millions. Not all of them is dense and ignorant. Some of the global AI community are intelligent and knowledgeable enough to demand to “Pause Giant AI Experiments: An Open Letter" (33707 signatures). And one of the major reasons is the fact that "AI systems with human-competitive intelligence can pose profound risks to society and humanity, as shown by extensive research and acknowledged by top AI labs".... https://lnkd.in/dvRxUP3y

Nope. Today’s AI follows a simple stimulus-response regression model, with receiving data input as predictor variables/regressor/explanatory variables statistically related to the responding data output as a "response variable", "criterion", "predicted variable", "measured variable", "explained variable", "experimental variable", "responding variable", "outcome variable", "output variable", "target" or "label". Being software applications of standard statistical techniques, it is sold as if mimicking or replicating or simulating human intelligence or cognitive functions or human behavior. You need to design a real or true AI, which is capable of interacting with the world or any its environments, physical or digital, social or virtual. Such a real AI follows a reality-stimulus-AI [intelligence/brain/mind]-response-reality model. It has the power to detect and interpret all sorts of stimuli, external or internal, as the detectable information or measurable changes in the environments to cause behavioral changes: physical stimuli chemical stimuli biological stimuli mental stimuli social stimuli information stimuli… The AI’s general intelligence (or brain or mind) should be capable of registering any stimuli and analyzing receiving information from the environments to infer or generate intelligent outputs, discoveries, problem-solving solutions, recommendations, content, predictions, goals, decisions, and actions. Confer it with OECD’s revised definition: An AI system is a machine-based system that, for explicit or implicit objectives, infers, from the input it receives, how to generate outputs such as predictions, content, recommendations, or decisions that can influence physical or virtual environments. Different AI systems vary in their levels of autonomy and adaptiveness after deployment. https://lnkd.in/dBP_U23x

#AI #ML #Tech Diffusion Models: Midjourney, Dall-E Reverse Time to Generate Images from Prompts: We have AI, physics, thermodynamics, probability, beautiful images AND time travel? What are you waiting for. Continue reading on Towards Data Science » #MachineLearning #ArtificialIntelligence #DataScience

It starts happening in next year. As of today’s AI, it belongs to big tech oligopolies, being really giant quackery and mega-fraud. If your AI systems are not grounded in reality, real or simulated, with embedded/programmed/encoded rational/scientific world models, it is a fiction or simply fraud, as today's large language models (LLMs) like Microsoft/OpenAI's GPT-4, Google's Gemini or Meta's LLaMA models. AI must learn all major abstract representations of the world, enabling machine rationality, its intelligence, learning, inference, planning, self-knowledge, understanding, and interaction. Truly intelligent systems (Real AI) are deeply grounded in the world, physical and virtual. Otherwise, they are fake and false or irrational as the human brain could be. The existential concerns expressed by a few rational minds have all the chances to come out from the human-mimicking, fake AI technology. “Mitigating the risk of extinction from A.I. should be a global priority alongside other societal-scale risks, such as pandemics and nuclear war,” [Statement on AI Risk]. "AI systems with human-competitive intelligence can pose profound risks to society and humanity, as shown by extensive research and acknowledged by top AI labs" [Pause Giant AI Experiments: An Open Letter]. The big tech AI is the biggest fraud of the modern industry, with its blatant false commercial advertising and exploiting the fact of irrationality, together with daydreaming or mind-wondering, to be the human brain's default state. So why do big tech brands make fraudulent “AI” promises at the scale of Microsoft, Apple, Nvidia, Amazon, Alphabet, Meta, or Tesla? Because it’s HUGELY profitable to mislead consumers and shareholders.https://https://lnkd.in/dKfh68yQ

Can AI Learn Physics from Sensor Data? Newton AI is learning Physics from scratch!

"Doing, not listening, was the starting point for how he [Einstein] learned physics,” wrote Scott Young about how Einstein learned physics. In the same way, the best way to learn about Generative AI is by building projects. Here is a guide to create a Retrieval Augmented Generation (RAG) application that provides a detailed project description, a step-by-step outline of the RAG process, which offers both theoretical insights and practical example code. Start your RAG project: https://lnkd.in/gEi4JF_w #GenerativeAI #MachineLearning #AIProjects #RAG #AICommunity #TechInnovation #AIExplained #DataScience

Nope. As it is, AI tools, like LLMs and ChatGPT, have the following intellectual disabilities: cognize nothing but nothing perceive nothing but nothing conceive nothing but nothing know nothing but nothing recognize nothing but nothing learn nothing but nothing understand nothing but nothing plan nothing but nothing reason nothing but nothing decide nothing but nothing perform nothing but nothing AI tools are dumb as any complex graph calculators or statistical software algorithms. or numerical datasets processing complex mathematical functions, sold as ANNs, deep learning, DNNs, etc. But they are great to create lies, fake news, deepfakes, disinformation and all sorts of biased hallucinations To understand something, machine intelligence and learning (MIL) encodes and makes sense of reality and its contents, like the human mind/brains maps, conceives, categorizes the world of things. MIL has the world modeling and simulating, learning and inference, understanding and interacting engine to differ facts from fake, truths from lies. MIL could run Cerebras’s Massive Wafer-Scale CS-3 Chip Compute Engine leapfrogging Nvidia GPU-powered services by more than 10x.

AI is about abstract and practical intelligence in order to interact with the environment, physical or digital. A tripling of knowing, inference, and acting comprises an agent. An agent acts in an environment, which includes other agents. An agent together with its environment is called a world, of physical or virtual or mixed realities. An agent could be, a coupling of a computational engine with physical sensors and actuators, called a robot, where the environment is a physical setting. An autonomous agent is one that acts in the world without human intervention. A semi-autonomous agent acts with a human-in-the-loop who may provide perceptual information and carry out the task. An agent could be a program that acts in a purely computational environment, a software agent, often called a bot. So, Sims, NPCs, Replikas or LLMs, such as GPT or Claude, are not AI agents with all identities, traits and individualities, knowingly interacting with each other, They are either a bot programs or language interface or life simulation video game with controlled virtual characters or NPCs controlled by gamemasters.. To be real intelligent agents, they need to have some necessary features, as pictured and defined in DAVID L. POOLE & ALAN K. MACKWORTH AI: FOUNDATIONS OF COMPUTATIONAL AGENTS prior knowledge about the agent and the environment stimuli received from the environment, which can include observations about the environment (e.g., light, sound, keyboard commands, web requests) as well as actions that the environment imposes on the agent past experiences, including history of interaction with the environment (its previous actions and stimuli) and other data, from which it can learn goals that it must try to achieve or preferences over states of the world abilities, the primitive actions the agent is capable of carrying out. https://lnkd.in/dDSTtznw