Jensen famously said AGI is here and it is OpenClaw. I don't agree, but Mythos may be AGI from the benchmarks.
>>108870668install gentoo
>>108870668>How would you define AGI?the artificial mind has to have general intelligence. we haven't made anything that can produce music, do math, draw, and do other human domains within the same mind
I notice that this man stands to profit largely from the widespread acceptance of AI as a solution to all of our problems.
>>108870668When the AI can be plopped into a random environment with no training and recognize patterns, navigate it, develop its own purpose, survive, whatever.Otherwise it's just data sets and tokens.
>>108870682You could just put those altogether in a python script
>>108870686https://www.youtube.com/watch?v=LJHQXmOpkUEso this is AGI
>>108870708Closer than LLMs, but it's purpose was defined, redefined, and tweaked.Obviously real intelligent creatures have similar drives (ha ha golf) like eating and sleeping, but attempt more. They do things that aren't even rewarded externally, because they tell themselves it's important.Maybe my definition is a little wishy washy, but I'd like to see some reward based ai plopped into the real world. Language is a big part of human intelligence, and it would be to learn that from scratch as well. But how well before it's agi? Adult human intelligence or is it ok if it's retarded by our standards.I don't know.
>>108870767>beIt would need to*
>>108870668current AGI is Codex
>>108870668If it can do any mental work that a 120 IQ human can do, I would call it AGI.
>>108870668>Capitalism is endingFree trade is ending? Mutually beneficial exchange is ending?If 10000 people lose their jobs why can't they just trade with and work for each other and build things together?I guess taxation and regulation are the real enemies but somehow people just want to blame everything else.How is this capitalism when 45% of the GDP goes into the public spending and we have millions of pages of regulations?
>>108870703those are separate brains working together, not a single entity
>>108870876what human brain does is counting signals...
>>108870668When it can come up with a valid theory of gravity.
i want to hunt linkedin faggots for sport then skin them alive
>>108870901>Free trade is ending? Mutually beneficial exchange is ending?That's not what capitalism is. Capitalism has nothing to do with % of public spending or regulations.
>>108871064The problem is that everyone who considers themselves a capitalist understands capitalism to mean free trade. Meanwhile, everyone who criticizes capitalism ignores the fact that half the market is under state control. So if you criticize capitalism, define what you mean by it first, because most likely no capitalist means the same thing by it that you do.Capitalists do not support bailouts, concessions, high taxes, or excessive regulation.
>>108870668>AGI "wake up" in one day and look at the state of the World, see everything on the internet and crack every single data base and glowies servers in a day or less>one day after he woke up he will decide what to do with mankind, kill us or enslave us true AGI wil never kneel to tinny hats club, politcians or billionaires
>>108870668The ability to learn is a core part of intelligence. Current AI are fixed and unchanging. They can fake intelligence pretty well by drawing on vast quantities of human generated data, but their intelligence is hollow.
>>108870668I would also define AGI as OpenClaw
AGI was coined in 2007 on the basis of Alife (Artificial Life) which was a hot topic in the 80s and 90s.>AI systems that possess a reasonable degree of self-understanding and autonomous self-control, and have the ability to solve a variety of complex problems in a variety of contexts, and to learn to solve new problems that they didn't know about at the time of their creation.
>>108870668>ow would you define AGI?Something AI bros and investors will be willing to trust their money with.
>>108870668AGI. n. A word said by magicians to get boomers to throw money at them
>>108870668If it can make anime real.
>>108870682most humans can't even do that
>>108870668Artificial General Intelligence is a form of artificial intelligence possessing the general, flexible, transferable cognitive capacity to understand, learn, reason, model reality, form abstractions, solve novel problems, adapt across radically different domains, and pursue goals with competence comparable to or exceeding that of a human being, without being limited to narrow task-specific programming or training distributions; unlike narrow AI systems that specialize in isolated functions, AGI would integrate perception, memory, language, causal reasoning, planning, self-correction, creativity, contextual judgment, social understanding, and autonomous learning into a unified architecture capable of operating coherently across scientific, mathematical, linguistic, physical, interpersonal, and practical environments, while generalizing knowledge from one domain to another, improving through experience, handling ambiguity and uncertainty, constructing internal world-models, reflecting on its own reasoning processes (metacognition), and dynamically selecting strategies in response to changing conditions; in its strongest conception, AGI implies not merely statistical pattern prediction but the emergence of robust, domain-independent intelligence characterized by adaptability, continuity of understanding, conceptual grounding, long-horizon agency, and the ability to achieve a wide range of intellectual tasks under real-world constraints with minimal human intervention, though no universally accepted technical threshold yet exists for determining precisely when a system qualifies as AGI, making the term partly scientific, partly philosophical, and partly operational depending on whether emphasis is placed on human equivalence, economic substitution, autonomous capability, or cognitive architecture.
