Human-level General Intelligence (AGI) Machine

From GM-RKB
(Redirected from General AI)
Jump to navigation Jump to search

A Human-level General Intelligence (AGI) Machine is an intelligent machine with a AI system capability that approximates human-level intelligence.

References

2024

  • https://x.com/karpathy/status/1834641096905048165
    • QUOTE: Are we able to agree on what we mean by "AGI". I've been using this definition from OpenAI which I thought was relatively standard and ok: https://openai.com/our-structure/

      AGI: "a highly autonomous system that outperforms humans at most economically valuable work"

    • For "most economically valuable work" I like to reference the index of all occupations from U.S. Bureau of Labor Statistics: https://bls.gov/ooh/a-z-index.htm
    • Two common caveats:
      1. In practice most people currently deviate from the above definition to only mean digital work (a relatively major concession looking at the list).
      2. The definition above only considers the *existence* of such a system not its full deployment across all of the industry.

2024

2024

  • (Wikipedia, 2024) ⇒ https://en.wikipedia.org/wiki/Artificial_general_intelligence Retrieved:2024-1-13.
    • An artificial general intelligence (AGI) is a hypothetical type of intelligent agent.[1] If realized, an AGI could learn to accomplish any intellectual task that human beings or animals can perform. [2] Alternatively, AGI has been defined as an autonomous system that surpasses human capabilities in the majority of economically valuable tasks. Creating AGI is a primary goal of some artificial intelligence research and of companies such as OpenAI,[3] DeepMind, and Anthropic. AGI is a common topic in science fiction and futures studies.

      The timeline for AGI development remains a subject of ongoing debate among researchers and experts. Some argue that it may be possible in years or decades; others maintain it might take a century or longer; and a minority believe it may never be achieved.[4] There is debate on the exact definition of AGI, and regarding whether modern large language models (LLMs) such as GPT-4 are early yet incomplete forms of AGI. Contention exists over the potential for AGI to pose a threat to humanity;[1] for example, OpenAI treats it as an existential risk, while others find the development of AGI to be too remote to present a risk. [4][5]

      A 2020 survey identified 72 active AGI R&D projects spread across 37 countries.[6]

  1. ↑ 1.0 1.1 Morozov, Evgeny (30 June 2023). "The True Threat of Artificial Intelligence". The New York Times. Archived from the original on 30 June 2023. Retrieved 30 June 2023.
  2. ↑ Hodson, Hal (1 March 2019). "DeepMind and Google: the battle to control artificial intelligence". 1843. Archived from the original on 7 July 2020. Retrieved 7 July 2020. AGI stands for Artificial General Intelligence, a hypothetical computer program...
  3. ↑ "OpenAI Charter". openai.com. Retrieved 6 April 2023.
  4. ↑ 4.0 4.1 "AI timelines: What do experts in artificial intelligence expect for the future?". Our World in Data. Retrieved 6 April 2023.
  5. ↑ "Impressed by artificial intelligence? Experts say AGI is coming next, and it has 'existential' risks". ABC News. 23 March 2023. Retrieved 6 April 2023.
  6. ↑ Baum, Seth, A Survey of Artificial General Intelligence Projects for Ethics, Risk, and Policy (PDF), Global Catastrophic Risk Institute Working Paper 20, archived (PDF) from the original on 14 November 2021, retrieved 13 January 2022

2023

2023

2023

2020

2019

2017a


2017c

2017d

  • (Wired, 2017) ⇒ https://wired.com/story/ray-kurzweil-on-turing-tests-brain-extenderstand-ai-ethics/amp
    • QUOTE: ... Ray Kurzweil: ... You need the full flexibility of human intelligence to pass a valid Turing Test. There's no simple Natural Language Processing trick you can do to do that. If the human judge can't tell the difference then we consider the AI to be of human intelligence, which is really what you're asking. That's been a key prediction of mine. I've been consistent in saying 2029. …

2017e

  • (Bengio, 2017) ⇒ Yoshua Bengio (2017) ⇒ "Creating ng Human-­Level AI". In: Asilomar Conference on Beneficial AI, January 6th, 2017.
    • QUOTE: What’s Missing
      • More autonomous learning, unsupervised learning
      • Discovering the underlying causal factors
      • Model-­‐based RL which extends to completely new situations by unrolling powerful predictive models which can help reason about rarely observed dangerous states
      • Sufficient computational power for models large enough to capture human-­‐level knowledge
      • Autonomously discovering multiple time scales to handle very long-­‐term dependencies
      • Actually understanding language (also solves generating), requiring enough world knowledge / commonsense
      • Large-­‐scale knowledge representation allowing one-­‐shot learning as well as discovering new abstractions and explanations by ‘compiling’ previous observations

2013b

2014a

2014b

  • (Brooks, 2014) ⇒ Rodney Brooks. (2014). “Artificial intelligence is a tool, not a threat.” In: Rethinking Robotics, November 10, 2014.
    • QUOTE: … a fundamental error in not distinguishing the difference between the very real recent advances in a particular aspect of AI, and the enormity and complexity of building sentient volitional intelligence. … Why so many years? As a comparison, consider that we have had winged flying machines for well over 100 years. But it is only very recently that people like Russ Tedrake at MIT CSAIL have been able to get them to land on a branch, something that is done by a bird somewhere in the world at least every microsecond. Was it just Moore’s law that allowed this to start happening? Not really. It was figuring out the equations and the problems and the regimes of stall, etc., through mathematical understanding of the equations. …

      … If we are spectacularly lucky we’ll have AI over the next thirty years with the intentionality of a lizard, and robots using that AI will be useful tools.

2012

2011

  • (Versite, 2011) ⇒ http://versita.com/jagi/
    • Artificial General Intelligence (AGI) is an emerging field aiming at the building of “thinking machines", that is, general-purpose systems with intelligence comparable to that of the human mind. While this was the original goal of Artificial Intelligence (AI), the mainstream of AI research has turned toward domain-dependent and problem-specific solutions;; therefore it has become necessary to use a new name to indicate research that still pursues the "Grand AI Dream". Similar labels for this kind of research include “Strong AI", “Human-level AI", etc.

      The problems involved in creating general-purpose intelligent systems are very different from those involved in creating special-purpose systems. Therefore, this journal is different from conventional AI journals in its stress on the long-term potential of research towards the ultimate goal of AGI, rather than immediate applications. Articles focused on details of AGI systems are welcome, if they clearly indicate the relation between the special topics considered and intelligence as a whole, by addressing the generality, extensibility, and scalability of the techniques proposed or discussed.

      Since AGI research is still in its early stage, the journal strongly encourages novel approaches coming from various theoretical and technical traditions, including (but not limited to) symbolic, connectionist, statistical, evolutionary, robotic and information-theoretic, as well as integrative and hybrid approaches.

2009

2008a

2008b

  • (Sandberg & Bostrom, 2008) ⇒ Anders Sandberg, and Nick Bostrom. (2008). “Whole Brain Emulation." Technical Report #2008-3, Future of Humanity Institute, Oxford University.
    • QUOTE: Table 10: Estimates of computational capacity of the human brain. Units have been converted into FLOPS and bits whenever possible. Levels refer to Table 2.
      • Source | Assumptions | Computational demands | Memory
      • (Leitl, 1995) Assuming 1010 neurons, 1,000 synapses per neuron, 34 bit ID per neuron and 8 bit representation of dynamic state, synaptic weights and delays. [Level 5] 5·1015 bits (but notes that the data can likely be compressed).
      • (Tuszynski, 2006) Assuming microtubuli dimer states as bits and operating on nanosecond switching times. [Level 10] 1028 FLOPS 8·1019 bits
      • (Kurzweil, 1999) Based on 100 billion neurons with 1,000 connections and 200 calculations per second. [Level 4] 2·1016 FLOPS 1012 bits
      • (Thagard, 2002) Argues that the number of computational elements in the brain is greater than the number of neurons, possibly even up to the 1017 individual protein molecules. [Level 8] 1023 FLOPS
      • (Landauer, 1986) Assuming 2 bits learning per second during conscious time, experiment based. [Level 1] 1.5·109 bits (109 bits with loss)
      • (Neumann, 1958) Storing all impulses over a lifetime. 1020 bits (Wang, Liu et al., 2003) Memories are stored as relations between neurons. 108432 bits (See footnote 17)
      • (Freitas Jr., 1996) 1010 neurons, 1,000 synapses, firing 10 Hz [Level 4] 1014 bits/second (Bostrom, 1998) 1011 neurons, 5·103 synapses, 100 Hz, each signal worth 5 bits. [Level 5] 1017 operations per second
      • (Merkle, 1989a) Energy constraints on Ranvier nodes. 2·1015 operations per second (1013-1016 ops/s)
      • (Moravec, 1999; Morevec, 1988; Moravec, 1998) Compares instructions needed for visual processing primitives with retina, scales up to brain and 10 times per second. Produces 1,000 MIPS neurons. [Level 3] 108 MIPS 8·1014 bits.
      • (Merkle, 1989a) Retina scale-up. [Level 3] 1012-1014 operations per second. (Dix, 2005) 10 billion neurons, 10,000 synaptic operations per cycle, 100 Hz cycle time. [Level 4] 1016 synaptic ops/s 4·1015 bits (for structural information) (Cherniak, 1990) 1010 neurons, 1,000 synapses each. [Level 4] 1013 bits
      • (Fiala, 2007) 1014 synapses, identity coded by 48 bits plus 2x36 bits for pre- and postsynaptic neuron id, 1 byte states. 10 ms update time. [Level 4] 256,000 terabytes/s 2·1016 bits (for structural information)
      • (Seitz) 50-200 billion neurons, 20,000 shared synapses per neuron with 256 distinguishable levels, 40 Hz firing. [Level 5] 2·1012 synaptic operations per secon 4·1015 - 8·1015 bits
      • (Malickas, 1996) 1011 neurons, 102-104 synapses, 100- 1,000 Hz activity. [level 4] 1015-1018 synaptic operations per secon 1·1011 neurons, each with 104 compartments running the basic Hodgkin-Huxley equations with 1200 FLOPS each (based on
      • (Izhikevich, 2004). Each compartment would have 4 dynamical variables and 10 parameters described by one byte each. 1.2·1018 FLOPS 1.12·1028 bits
Retrieved from "http://www.gabormelli.com/RKB/index.php?title=Human-level_General_Intelligence_(AGI)_Machine&oldid=917401"