Rostelecom plans to replace part of the missing IT specialists with artificial intelligence
[ad_1]
Due to a shortage of IT personnel, Rostelecom intends to automate software development, data analytics and testing using artificial intelligence (AI) based on its own platform within two to three years. Taking into account the fact that the authorities are actively investing in the training of IT personnel, and large companies are introducing AI, there may be a surplus of personnel in the market, which will lead to a decrease in wages, its participants fear. Foreign companies are already developing the practice of replacing part of their employees with neural networks, reducing personnel. Rostelecom promises to share it with other market participants upon completion of the project development.
Senior Vice President for IT at Rostelecom, Kirill Menshov, on September 25, at the planning session “Challenges of Technological Sovereignty,” which was held as part of the Moscow Evenings congress, said that due to personnel shortages in the industry, the company plans to compensate for some of the missing IT specialists technologies: “We need to automate the work of developers using artificial intelligence, which will reduce the need for personnel, we are actively pursuing this. We need to do this now in order to get results in two or three years.”
Rostelecom clarified to Kommersant that the idea relates to the Lukomorye project (a full-cycle software development platform): the solution will cover development, including in the areas of business analysis, design, testing and documentation creation. The operator announced Lukomorye in mid-July, but then it was only about simplifying the work of IT specialists. The head of the Ministry of Digital Development, Maksut Shadayev, estimated the shortage of IT personnel in the Russian Federation at approximately 500–700 thousand people. (see “Kommersant” dated July 18). The ministry did not answer Kommersant.
Rostelecom promises not to fire or replace existing AI employees: “At first, Lukomorye will be used in the operator’s own projects, then the platform is planned to be sold to other companies that are faced with a personnel shortage.”
The company did not disclose investments in the project. According to the director of the Association of Computer and Information Technology Enterprises (APKIT) Nikolai Komlev, such a development “at the start” could cost Rostelecom in the amount of 1.5 billion rubles.
The CEO of custom software developer Only, Kirill Vladimirov, believes that given the complexity of the task, the platform can cost “billions of rubles” just at the stage of training the neural network, and to implement “even the basic version (MVP) of the project with all sorts of scenarios can take more than three years.”
Dmitry ChernyshenkoDeputy Prime Minister, September 26, quote from TASS:
“Russian GDP will be able to receive an additional 11 trillion rubles. by 2030 thanks to the massive introduction of artificial intelligence technologies.”
Other Russian companies are also developing developments in the field of generative AI. Thus, Sberbank, which presented the generative neural network GigaChat, created the GigaCode code development and auto-completion tool. Yandex is also considering a service for AI-based developers. Market participants clarify that so far such tools can only add small fragments of code (see “Kommersant” dated September 1).
Most of the same type of mechanical work can already be entrusted to specially trained neural network algorithms based on Open Ai or ChatGPT, says Irina Ostroukhova, director of work with partners at RecFaces, but “one should not expect a complete exclusion of humans from IT developments in the coming years.” The development of an AI model will also require the emergence of new professions related to neural network management, adds Svetlana Vronskaya, an expert at Corus Consulting Group.
But there may be too many people.
Thus, Mr. Vladimirov fears that “taking into account the large investments in universities,” over the next decade there may be an oversupply of IT specialists in the Russian Federation, and the active introduction of AI will lead to a surplus of personnel and a decrease in salaries.
In March, the American bank Goldman Sachs published a report in which it predicted that generative AI could automate about 300 million jobs worldwide. According to McKinsey, up to 30% of working time in the United States could soon be automated, including through generative AI.
There are specific examples. In April, the cloud service DropBox announced the reduction of about 500 people – about 15% of the staff, which will be replaced by AI. In May, IBM CEO Arvand Krishna said the company intended to stop hiring for positions that could be replaced by AI. There are approximately 26 thousand such people at IBM, according to Mr. Krishna: “30% of them can be easily replaced by AI and automation in a five-year period.” Thus, about 7.8 thousand workers may be laid off at IBM, which currently employs about 260 thousand people. However, according to the head of IBM, “the total number of company employees is still growing.” In June, the American Chegg, which develops technologies in the field of education, announced the reduction of 80 employees in the near future, whose functions will be replaced by AI—that’s 4% of the company’s staff.
[ad_2]
Source link