Experts estimate that this year will be the breakthrough of artificial intelligence: computer systems will hire and train employees, Deep Pike will make us doubt what our eyes see, and stores will sell us more products by collecting and analyzing information about us.
By Tal Shcaf
Once, when we were told “artificial intelligence”, we would think of an intelligent computer, one that could be consulted on financial matters and have a mental conversation before bed. In the last year, the romantic image of artificial intelligence (AI) has faded, and it has been presented under the cold light of criticism: a means of making profits for companies, a tool for collecting data while infringing on privacy and demonstrating a biased and racist attitude, and a strategic arena.
Right now, two opposing processes are taking place: on the one hand, the capabilities of artificial intelligence are intensifying. On the other hand, the awareness of the ethics of artificial intelligence is growing. What will be the result of this steadfast struggle? Maybe fair artificial intelligence. Experts estimate that 2021 will be the year of AI’s breakthrough, not least under the influence of the Corona, which has created new human needs, such as avoiding human contact. Social distance is the beautiful hour of artificial intelligence.
A report by research firm IDC published this week found that revenues from the global artificial intelligence industry – software, hardware and services – will continue to grow in 2021 to a total of $ 328 billion, and next year are expected to cross the half-trillion-dollar line. Another IDC report He has stated that global spending on AI will double in the next four years and reach $ 110 billion in 2024. Companies use AI to streamline and increase sales. In the coming years, we will meet fewer and fewer customer service people and more and more artificial intelligence systems. This is the beginning of the AI revolution.
The hot trends
One can see several areas of artificial intelligence, which are leading the revolution right before our eyes:
Beneath this polished expression lurks a strategy of using artificial intelligence to give the user exactly what he wants, or thinks he wants, and communicate with him through chatbots, digital assistants and smart agents. According to IDC, the two industries that will invest the most in AI this year will be commerce and banking. Trading companies, from Amazon to the neighbourhood shoe store, will assimilate more and more artificial intelligence, which is proving to increase sales. Banks are investing capital in keeping customers away from branches by providing automated advice and services, forcibly collecting data on customers and, also, smart fraud prevention systems, another sick evil of the time.
Employers, especially the largest ones, have realized this year that an AI system can manage and care for employees much better than human resources people. It starts with artificial intelligence systems for hiring employees who know so much about job applicants that they eliminate the need for a job interview. It continues with tools for training employees, increasing employee productivity and automating their tasks, while on the other hand, they intensely monitor and monitor employee work.
Rise of the Deep Pike
Deep Pike technology, synthetic video in its professional nickname, was born a total of two years ago as a dubious tool for making porn videos. Now it’s becoming a legitimate technology, one of 30 technologies that are about to shake the world, according to Gartner’s hype cycle report.
Over the past year, Deep Pike has been used in a documentary about the LGBT community in Russia. There was a movie that documented a non-existent event: a speech by President Nixon following the failure to land on the moon of Apollo 11. Hulu produced videos with the stars of the sports leagues, who did not come to the studio at all but were transplanted using Deep Pike. And there were also political videos ahead of the US election in which Vladimir Putin and Kim Jong-un were seen speaking out in favour of democracy.
But Deep Pike’s most significant leap was made in the Sassy Justice series by Trey Parker and Matt Stone, South Park creators. This is a series of sharp slaps, done entirely using Deep Pike. The star is Detective Sassi, whose face is the face of Donald Trump, and he interviews synthetic figures like Jared Kushner (who is assimilated into a child), Mark Zuckerberg, who markets dialysis treatments, Al Gore, Julie Andrews, Ivanka Trump and even Donald Trump himself. The quality of the Deep Pike is so good that without the ripping texts, it would have been possible to make a mistake and think that these are the people themselves. And the future is just beginning.
In the last year, the verbal communication ability of artificial intelligence (Conversational AI) has reached new heights. Natural Language Recognition (NLP) technologies make it possible to read and understand texts at a better level than humans, which is nothing compared to OpenAI’s GPT-3 text writing capability. Last August, an American student named Liam Four, who used GPT-3 to produce an entire blog, was written entirely by a computer. The blog was so interesting that many asked to join it as subscribers.
Artificial intelligence that produces texts? The possibilities are endless. Already today, some systems automatically summarize long texts. The business world will use these capabilities to create summaries, customer information summaries, and sales call scripts. We are already at a stage where it is not possible to tell whether the text we are reading is authentic or synthetic. Combined with the ability to understand the texts of artificial intelligence, we are close to a day when we will not know if we are getting service from a person or a machine.
Pharmaceutical companies are expanding the use of artificial intelligence. For example, companies such as Pfizer and Moderna have developed vaccines for Corona in a short time. The other side of medical research is examining the effects of drugs using a vast amount of data on patients, from the moment they are born to the end of their treatments, which is no longer healthy for privacy. Take, for example, Machine Intelligence, which scanned all existing drugs to find diseases that they could cure in addition to the initial indication. The company used deep learning and natural language analysis tools to scan 1.2 million heart patients’ claims against insurance companies, a massive amount of text detailing all stages of treatment and the disease’s characteristics. So while this is an invasion of patient privacy, the company has identified six new drugs to treat heart patients. So maybe that’s a fair price.