Artificial intelligence is the ability of machines to perform certain tasks, which need the intelligence showcased by humans and animals in some of their own behaviour. This definition is often ascribed to Marvin Minsky and John McCarthy from the 1950s, who were also known as the fathers of the field.
Artificial intelligence allows machines to understand and achieve specific goals. AI includes machine learning via deep learning. The former refers to machines automatically learning from existing data without being assisted by human beings. Deep learning allows the machine to absorb huge amounts of unstructured data such as text, images, and audio.
Any AI system must be able to have some of the following characteristics: Observation, analytical ability, problem solving, learning, etc.
AI was a term first coined at Dartmouth College in 1956. Cognitive scientist Marvin Minsky was optimistic about the technology’s future. The 1974-1980 saw government funding in the field drop, a period known as “AI winter”, when several criticised progress in the field.
However, the fervour was revived afterwards in the 1980s when the British government started funding the technology again, especially because they were worried about competition with the Japanese.
Some technology companies have recently managed to double their market value and get a lot of profits. This is due to its investments in developing its own AI departments and adapting the artificial intelligence techniques to its innovative applications and devices.
Apple is integrating as much AI into its products and operations as possible. In 2019, Apple, the world’s largest technology company made several AI-based technologies, which served the following purposes:
Apple is infusing as much AI into their products and operations as possible. In 2019, the world’s largest technology firm made several AI-based technologies, which served the following purposes:
1) Bring greater personalization to web search and Siri results.
2) Shore up their competitive disadvantage in the self-driving car space.
3) Make images “shoppable” by allowing users to search using photos, rather than keywords.
4) Improve iPhone photography with AI-powered photo enhancement.
Siri, of course, is another example of how Apple runs on AI. The voice-powered assistant is designed for continual, at-the-edge improvement, which means it uses customer communications to further train itself without having to transmit those private communications to Apple servers.
Ford is at the forefront of the transportation transformation. It was one of the first companies on earth to deploy a neural net at scale and has since brought artificial intelligence to both their assembly lines and in the operation of the vehicles they sell.
Ford Edge’s all-wheel-drive system, for example, uses artificial intelligence to automatically determine if all-wheel drive is needed — more quickly and accurately than a human driver. In Ford factory, AI can detect wrinkles on seat fabric.
In 2019, to better compete in the race for full vehicle autonomy, the organization made a $1billion investment into Argo.ai. Ford expects to roll out a Geo-fenced self-driving car fleet within 3 years.
Last year, Google released Tensorflow, its open-sourced platform for machine learning, giving everyone access to one of the most advanced machine learning platforms ever created. More than 50 Google products have adopted the platform to put deep learning to work.
Internally, Google has hundreds of employees who are working on AI field. Their ultimate goal is to transform their panoply of AI-related services into a cohesive digital assistant that can proactively manage and automate your entire life.
By releasing Tensorflow to the Open Source community, Google is sending a clear message that artificial intelligence is for everyone. The platform makes available all sorts of pre-trained models and machine learning algorithms.
Together, they represent millions of hours of computer training, meaning everyone has access to the most powerful AI tools in existence.
Today, machine learning is driven, in part, by major technology companies who train models using their massive data sets. These out-of-the-box tools are very powerful, but we can make them even more powerful by layering additional functionality, customized to your individual needs.
By applying pre-trained machine learning models to new datasets or the new information, we are able to efficiently apply complex rules and learning to a new problem, without having to reinvent the wheel.
Bank of America is turning to artificial intelligence to help reduce its labor force and drive more of its customers to receive help via automated systems and chatbots.
In 2018, the company rolled out Erica, an in-app customer service agent. By October of 2019, the digital assistant had handled about 75 million in-app customer service interactions.
Unsurprisingly, 2019 also saw a steep decline in new hires for customer service positions. One study on Bank of America’s hiring practices found that the conglomerate had reduced branch-related job openings by half. At that same time, the number of Artificial Intelligence and Machine Learning (AI-and-ML) related Bank of America’s job openings had doubled.
As artificial intelligence becomes increasingly adept at automating repetitive tasks and interacting with humans, we can expect customer service as we know it to increasingly be handled by machine.
AI integrated chips:
Tesla aims to create AI integrated chips that will enable cars to navigate through freeways and even traffic. Approximately 6 billion transistors constitute the circuit of each Tesla chip.
These Tesla chips are 21 times faster than the original Nvidia chips and 20% cheaper too. They have 32 megabytes of high-speed SRAM memory on the chip because of which fetching data is faster and easier compared to DRAM.
For better performance, Tesla automobile systems have two AI chips. Both the chips make separate assessments of the traffic and danger situation around the cars.
The assessments are then matched and the car is guided accordingly if the outputs are the same. If there is ambiguity in the outputs obtained from the chips, then revaluation is done until a safe and suitable decision is taken. Thus, dual chips will enable better control over the navigation in self-driving Tesla cars.
In the modern world, we are surrounded by artificial intelligence. From assistants such as Amazon’s Alexa to the internet predicting what we may like to buy next. Artificial intelligence (AI) is everywhere. Self-driving cars are also an example of the application of artificial intelligence (AI).
Broadly, AI can be divided into two categories: Narrow AI and General AI.
Narrow AI is the kind we use everywhere -- from flagging content online, detecting faces in pictures, to simple customer care inquiries.
General artificial intelligence (AI) till date remains just a concept. The idea behind General AI is to make it as adaptable and flexible as human intelligence. When scientists will be able to develop general AI remains a hotly contested debate, with some saying it’ll arrive by as soon as 2040 to others saying its centuries away, given the lack of understanding of the human brain.