Wednesday, April 10, 2024
No menu items!
HomeFuturologyContinuous learning and flexible careers: how neural networks will change the labor...

Continuous learning and flexible careers: how neural networks will change the labor market

become a familiar tool for any person, says Alexander Krainov, director for the development of artificial intelligence technologies at Yandex.

Generative neural networks have been developing rapidly over the past few years and are already acting as a full-fledged human assistant in everyday and professional tasks. This process can be compared to how the Internet was born and developed. In those years, it was impossible to imagine that social networks, search engines, marketplaces, video streaming and personal assistants would appear.

Now in the field of text generative neural networks, humanity is at the same point. Therefore, it is impossible to predict what will happen next. Nevertheless, several trends are already visible today.

Neural networks are more than IT

Even 20-30 years ago, people added a line to their resumes – “confident PC skills”. It was a talent that not everybody possessed.. Today, this skill is basic and obvious; no one lists it as a competitive advantage. It seems that in the very near future we will see how people will start adding the phrase “confident use of neural networks” to their profiles. Interestingly, the skill of prompt engineering (the ability to write the correct neural network query) can really be useful for people right now. But, most likely, this stage will pass quickly and almost all people will learn how to work with neural networks.

Until recently, neural networks were trained exclusively by data scientists, ML developers, and analysts. They select the correct architecture, or, more simply, the type of formula for the model, collect data, train the neural network, that is, using mathematical methods, they obtain formulas that are as close as possible to the optimal parameters, and measure the quality of the resulting model. But now we see that people of different professions are needed to train neural networks: those who know natural language well, who can carefully check the accuracy of facts, quickly dive into any topic, or are a professional in some subject area.

Today, professionals, having reached a high level of qualification and recognition, are committed to passing on their experience to students and colleagues. In this case, they are less busy with practice and devote more time to enlightenment. It is likely that another important type of their activity will be the training of neural networks: specialists will transfer their knowledge and train the neural network to answer reliably and completely the most complex questions from their subject area. Thanks to artificial intelligence, their knowledge can be applied an unlimited number of times.

AI and new career opportunities

Society has gone through similar changes in the past. When computers appeared, many feared that they would leave millions of people without work. But time has shown that the development of technology, on the contrary, has created many new professions and jobs. A similar situation is developing with neural networks: they will become not only a tool that will help reduce the routine part of the work of many specialists, but also the cause of the emergence of new professions and jobs.

This process is already underway: a new specialty has emerged – an AI trainer. AI trainers write examples of texts on which the neural network learns to give complete, reliable and accurate answers. They must be able to quickly dive into any complex topic, be it astronomy or phonetics, in order to be able to write a reference example for a neural network. This is a new career opportunity for editors, art historians, linguists, teachers, translators and many other professionals who are good at language, can check facts and love working with texts. Yandex is already hiring hundreds of such employees.

Most likely, in the future, neural network trainers will have a specialization and professionals in certain areas will be responsible for training various models. For example, doctors, engineers, and lawyers will appear among AI trainers. And when a specialist reaches high qualifications, he will teach neural networks what he himself can do. First of all, this will affect people who solve the most complex and atypical tasks in their fields.

Neural networks and expansion of professional competencies

Technology development has accelerated dramatically over the past ten years. Previously, a person could get one profession and work in it all his life, his activity could change very slightly. Now, thanks also to AI, we can expect accelerating progress. Transformation and the constant emergence of new professions provide an opportunity to repeatedly change their field of activity, learn how to use new tools and expand their competencies.

Even 30 years ago, for example, there were no professions of a web developer, blogger or SMM manager. At the early stage of the Internet’s birth, it was impossible to imagine that such specialties would appear. It’s the same with neural networks. This is a young and very fast-growing technology, so even forecasts for five years ahead look naive, just as forecasts for the development of the Internet looked 30 years ago.

How to interact with neural networks now
Today we are at a very important point in the development of neural networks. Until now, AI technologies have worked inside products, and the average person could not “feel them with their hands”. Artificial intelligence algorithms have long been built into search and recommendation systems, machine translation, weather forecasts, and voice assistants.

Now users have the opportunity to interact with artificial intelligence directly. For example, a new generation generative neural network YandexGPT is built into Alice and is available to millions of users right in a smartphone, search engine or smart speaker. Thanks to this, people will begin to better understand the capabilities and limitations of neural networks, many fears, prejudices and false expectations will go away, and there will be much more ideas where and how neural networks can be used.

Over time, artificial intelligence will become an indispensable tool that helps reduce the time spent on routine tasks. A well-trained neural network will be able to provide a person with access to a huge amount of information in various fields of knowledge and structure it.

The most important skill is the ability not only to receive information from a huge number of different sources, but also to evaluate its reliability, think critically and systematize your knowledge. This skill is especially relevant when training and using neural networks.

And, perhaps, most importantly, people need not lose the ability to interact with each other, think, communicate and negotiate, because the neural network will not replace us in this.



Please enter your comment!
Please enter your name here

Most Popular

Recent Comments