No one is surprised by artificial intelligence anymore – it's hard to find someone who hasn't interacted with chat GPT. But is the professional use of AI capabilities in business processes hype or benefit?
First of all, AI in business processes now — is reality and a working tool that is rapidly developing, becoming smarter and taking on more and more functions. Those people and companies that will ignore this will be out of the game tomorrow. But it is very important to distinguish between using AI for entertainment and in professional activities. GPT and other common models are not suitable for working in such precise fields as law or medicine. Specialized models and systems are being developed for this.
That is, smart models from industry giants such as openAI or Google, which have invested hundreds of billions of dollars, cannot cope with a legal task that an application from aiXpert from a small Russian developer can solve?
There are many nuances, but one of the serious problems with generative models is hallucination. GPT very plausibly makes things up because it is trained on data obtained from open sources and containing errors, contradictions, and unreliable information. It can incredibly plausibly generate cases from judicial practice, with reference to case numbers. But such cases do not actually exist – it's plausible nonsense. Models themselves do not know how to check facts or evaluate the logic of their conclusions. For this, specialized companies create separate complex systems. For example, as part of the yurXpert legal complex, there is its own vectorized database of regulatory acts of the Russian Federation, with which the aiXpert ML complex interacts, and this together ensures the complete absence of hallucinations and high reliability of answers.
The legal sphere — one of the most conservative categories in any company. What does the introduction of AI-based legal services give to business?
Savings and efficiency. According to research, lawyers spend more than 25% of their working time searching for information. Translate these figures into hours, multiply by the cost of an hour of work, and it turns out that it costs Russian corporations tens of billions of rubles a year. It's worth thinking about, isn't it. Many ready-made AI-based solutions have been developed that form a legal position for lawyers, prepare document texts, and take on all kinds of routine tasks. AI is able to analyze incoming mail in a company, identify letters that knowingly carry legal risks, such as lawsuits, complaints, etc. – and send them to the relevant department. Such a tool will save any large company tens of millions of rubles a year.
So, it turns out that artificial intelligence can replace a person? Is this something to expect?
Of course not. But people who use artificial intelligence will replace those who don't. AI clearly expands capabilities, saves time and helps to make the right decisions. When at the next stage of development, humanity switched from horses to cars, people did not become superfluous. On the contrary, people have opened up new opportunities, increased labor productivity. But, of course, some professions had to retrain and master new competencies. The same thing is happening now with the introduction of AI into everyday life.
What advice can you give to people who are afraid of the introduction of artificial intelligence?
You shouldn't perceive artificial intelligence as a threat, it's more of a "digital upgrade." By automating routine processes, it will free up more time for creativity and strategic development. Therefore, it is now especially important to pay attention to improving your own education, develop, try, be in trend.
I think that the key skills of the future will be adaptability, emotional intelligence and working with AI as a tool.
aiXpert — a Russian IT company that develops solutions in the field of artificial intelligence. Specializes in creating professional AI solutions for business. Creator of a specialized legal AI system for lawyers - yurXpert. Laureate of the "Platform 2025 year" AI-Olympus award.
Diana Grabko