Will AI ever reach human intelligence?

2023 © Wikiask
Main topic: Tech
Other topics: Artificial intelligence (AI)
Short answer:
  • Recent advances in AI bring it closer to human intelligence, yet it remains limited in capability compared to the human brain.
  • Unlike AI, the human brain picks up knowledge through independent interactions with the environment.
  • The human brain has complex emotions like conscience, empathy, and self-awareness, which seem impossible to integrate within an AI, given our limited understanding of human emotions and technological limitations.

The word "AI" was first used in 1956 at Dartmouth College by Marvin Minsky, a cognitive scientist. AI uses machine learning for teaching a computer, a robot operated by a computer, or software to process a vast amount of data to build data sets. The AI can then process inputs to provide the outcome based on its algorithms and data sets.

Human-level intelligence is a complex general capability based on cognition, conscious state, conscience, self-awareness, logic, and reasoning.[edit]

Human intelligence is generalized and includes the capacity to pick up new skills through independent experiences, environmental interactions, and adaptation to novel circumstances. It also comprehends and manages abstract ideas and applies knowledge to influence one's surroundings.

Human intelligence has several unique attributes:

  • Cognition: This ability comprises perception and judgment, the states and activities involved in knowing.[1]
  • Conscious state: It denotes the awareness of physical and emotional being.[2]
  • Conscience: It is an ethical anchor that continuously evaluates the morality of your deeds. It brings a feeling of guilt in a person for actions that do not align with their moral compass.[3]
  • Self-awareness: Understanding your personality, behavior, habits, patterns, or character traits is self-awareness. [4]
  • Logic -Logic is the study of sound reasoning, particularly when it includes making judgments.[5]
  • Reasoning -Without using new information, the reason is the process by which the human intellect moves from what it already knows to what it does not.[6]

These attributes and their infinite interactions lend undeterminable complexity to the range of human intelligence. Human intelligence can specialize in a specific task through constant repetition and deliberate practice. However, the accuracy of the outcome has a varying degree of success compared to AI, which remains consistent.

AI is a specialized capability based on logic, reasoning, and a defined algorithm for what it may do[edit]

Artificial intelligence focuses on creating software that enables computers to exhibit intelligent behavior.

Most AI researches focus on specific applications like speech-to-speech translation or planning in small, well-defined job domains. However, the long-term objective of creating broadly intelligent, autonomous intelligence remains far in the future.

Artificial Intelligence refers to the wide range of specialized skills computers possess to complete jobs. Machine Learning is the subset of artificial intelligence that enables computers to learn by sorting and analyzing data.[7]

Machines are continuously expanding the range of their intelligence through Neural Networks, Deep Learning, Computer Vision, and Natural Language Processing algorithms. The strides in AI are now helping real-world issues across several industrial and scientific applications. Currently, AI machines and applications are in use for:

  • Optimizing retail business services and customer experience
  • Preventing financial fraud
  • Providing better, faster, and more accurate healthcare
  • Enhancing government systems
  • Facilitating transportation and logistics
  • Automation in the oil and gas industry[8]

Machine learning is not the same as human learning[edit]

Humans learn by independent experiences, whether on their own or that of others. While machines learn via curated experience in the form of selected historical data.[9]

Neither the human brain nor a computer is comparable to the human brain. While computers can conduct "neural network" functions, they are inspired by the neurons of the brain but are not self-organizing or adaptable. Currently, machine learning cannot replace human intelligence because of technical limitations.

Machine learning and processing considerably surpass the human brain's ability to remember, process, and comprehend information without switching off. Humans need sleep to recalibrate brain and body functions. On the other hand, AI can be constantly "online" and consistently process a colossal amount of data with precision without needing rest.

The combination of human and machine learning knowledge is cutting edge in various business sectors as soon as more complex skills such as knowledge application, abstraction, and issue analysis is required.[10]

AI doesn't communicate in human terms[edit]

Humans communicate through language, intonations, pronunciation, and intent. Words frequently have meanings based on context and complex articulations, which is why understanding language is so challenging for computers and AI systems. For example, AI cannot differentiate sarcasm from its literal meaning.

Human language is complex for machines to understand. At the same time, technology has developed ways for superior computers to comprehend other complex human problems but communication at a human level.[11]

The current capabilities of AI are limited and will expand as machine learning gets more refined[edit]

According to industry experts, AI and robotics will undertake some of the most challenging tasks by 2030. In the foreseeable future, AI will be used to prescribe highly personalized medical care, determine foreign policy, and tackle climate change issues.[12]

Research topics in Business-applied Artificial Intelligence
Research topics in Business-applied Artificial Intelligence CC BY: Wikimedia

Due to the current limitations of technology, and lack of understanding of human consciousness, kindness, and empathy, it seems improbable that AI will gain human-level intelligence in the foreseeable future.[13]

There is an unverified claim that Google chatbot AI is sentient[edit]

A senior software developer for Google, Blake Lemoine, was recently let go for making unverifiable claims that LaMDA, the company's AI chatbot, is self-aware. Google has refuted the claim.

LaMDA chatbot can have natural and easy conversations that mimic human comprehension. Blake was involved in evaluating the chatbot for any discriminatory or hate speech.[14]

References[edit]

  1. ↑ "cognition | Definition, Psychology, Examples, & Facts | Britannica". www.britannica.com. Retrieved 2022-10-19.
  2. ↑ "States of Consciousness". Noba. Retrieved 2022-10-19.
  3. ↑ Giubilini, Alberto (2022), Zalta, Edward N. (ed.), "Conscience", The Stanford Encyclopedia of Philosophy (Summer 2022 ed.), Metaphysics Research Lab, Stanford University, retrieved 2022-10-19
  4. ↑ "What Is Self-Awareness, and Why Is It Important?". www.betterup.com. Retrieved 2022-10-19.
  5. ↑ "What is Logic? – Introduction to Philosophy: Logic". press.rebus.community. Retrieved 2022-10-19.
  6. ↑ "Reasoning: Meaning, Definition and Types". Psychology Discussion - Discuss Anything About Psychology. 2016-08-17. Retrieved 2022-10-19.
  7. ↑ Zawacki-Richter, Olaf; Marín, Victoria I.; Bond, Melissa; Gouverneur, Franziska (2019-10-28). "Systematic review of research on artificial intelligence applications in higher education – where are the educators?". International Journal of Educational Technology in Higher Education. 16 (1): 39. doi:10.1186/s41239-019-0171-0. ISSN 2365-9440.
  8. ↑ "5 Capabilities of Artificial Intelligence and Machine Learning - DAPL". Digital Aptech. 2021-10-05. Retrieved 2022-10-19.
  9. ↑ JanardhananPS (2020-04-02). "Human Learning and Machine Learning - How they differ ? - DataScienceCentral.com". Data Science Central. Retrieved 2022-10-19.
  10. ↑ Henno, J.; Jaakkola, H.; Mäkelä, J. (2021-09). "Machine Learning vs Human Learning". 2021 44th International Convention on Information, Communication and Electronic Technology (MIPRO): 542–548. doi:10.23919/MIPRO52101.2021.9596771. Check date values in: |date= (help)
  11. ↑ "Why Computers Can't Master Human Language For Now". 2022-06-10. Retrieved 2022-10-19.
  12. ↑ Agrawal, Manas. "Council Post: The Possibilities Of AI In 2030: Transformation Across Dimensions". Forbes. Retrieved 2022-10-19.
  13. ↑ "How Will We Know When Artificial Intelligence Is Sentient?". Discover Magazine. Retrieved 2022-10-19.
  14. ↑ "Blake Lemoine: Google fires engineer who said AI tech has feelings". BBC News. 2022-07-23. Retrieved 2022-10-19.
Retrieved from "https://wikiask.org/w/index.php?title=Will_AI_ever_reach_human_intelligence%3F&oldid=7096"