At a very high level, artificial intelligence can be divided into two broad types:
Narrow AI
Narrow artificial intelligence (AI) is what we see around us in computers today.Intelligent systems that are taught how to perform certain tasks without being explicitly programmed.Examples include speech and language recognition of the Siri virtual assistant on the Apple iPhone, image recognition systems on self-driving cars, or recommendation engines that suggest what you might like. Unlike humans, these systems can only learn or be taught how to do defined tasks, so these are called narrow AI.
General Artificial Intelligence
General artificial intelligence is very different and is the type of adaptive intelligence found in humans, a flexible form of intelligence that can learn to perform a wide range of tasks, from haircuts to creating spreadsheets or reasoning on a wide range of topics based on their accumulated experience.
This kind of artificial intelligence, more commonly seen in movies, such as Hal in 2001 or Skynet in the Terminator, but not available today, is being researched by AI experts as to how soon it will happen.
What can narrow AI do?
There are numerous applications emerging for Dar AI:
* Interpret video broadcasts from drones performing visual inspections of infrastructure such as oil pipelines.
* Organization of personal and business calendars.
* Responding to simple customer service queries.
* Coordination with other intelligent systems to perform tasks such as booking hotels at a convenient time and place.
* Help radiologists detect potential tumors in X-rays.
* Flag inappropriate content online, detect wear and tear in elevators from data collected by IoT devices.
* Create a 3-D model of the world from satellite images and so the list goes on.
New applications of these learning systems are emerging all the time. Graphics card designer Nvidia recently unveiled Maxine, an artificial intelligence-based system that allows people to make quality video calls regardless of the speed of their internet connections. Full system to transmit a static image and the video stream over the internet, not Few the caller, the caller's facial expressions and movements in a way that is designed to reproduce liven up instead of such a call 10 times and reduces the bandwidth required for real-time video are indistinguishable from.
What can general AI do?
A survey conducted by artificial intelligence researchers Vincent C Müller and philosopher Nick Bostrom in 2012/13 among four expert groups reported that Artificial General Intelligence (YGZ) would improve by 50% between 2040 and 2050, Dec Dec to 90% by 2075.The group went further and estimated that the so-called "Superintelligence," which Bostrom described as "any intelligence that greatly exceeds the cognitive performance of people in almost all fields of interest," was expected about 30 years after ygz was acquired.
However, recent assessments by artificial intelligence experts are more cautious. Pioneers in the field of modern artificial intelligence research, such as Geoffrey Hinton, Demis Hassabis and Yann LeCun, say that society is not even close to developing YGZ.
Given the skepticism of pioneers in the field of modern artificial intelligence and the very different nature of modern narrow AI systems from AGI, there is perhaps little basis for fears that a general AI will disrupt society in the near future.
However, some artificial intelligence experts believe that such projections are wildly optimistic, given our limited understanding of the human brain, and believe that ygz is still centuries away.
Comments