Kazakh companies and government agencies are already handing over a significant part of HR work to artificial intelligence — from resume analysis to matching vacancies for the unemployed. Businesses save recruiters' time, and job seekers find work faster, but without quality data and human oversight, automation carries legal and staffing risks. Thousands of citizens have already found employment thanks to the technology.
For employees, this means the algorithm can flag a person at increased risk of dismissal in advance. However, the reasons and the final decision remain with the HR specialist. Most companies, judging by an industry study, are not yet ready for full-scale adoption of the technology.
What AI already does in HR work
The state is applying AI in the labor market: a pilot of the Digital Employment Service launched in July 2026 in Pavlodar and Mangystau regions. In the first two months, the AI assistant sent job seekers 59,000 notifications about vacancies, and about 3,000 people found work.
The system generates a digital resume on its own, matches jobs taking into account education and experience, and offers training when qualifications are lacking. Elena Poleshchuk, director of the HR company LavoroSolutions, notes that AI helps HR departments process resumes, match skills to job requirements, and prepare interview questions and reports.
Another area is assessing a candidate by specific skills rather than by job title and previous employer. With AI, it is possible to determine how well a resume matches a vacancy, but the candidate's motivation and team fit are still assessed by a human.
Why companies are not yet ready to entrust HR to algorithms
According to the industry study "Kazakhstan HR Digital Maturity Index — 2026," the average score of 44 participating companies was 44.4 out of 100. Almost half of them (45.5%) operate in fragmented systems without integration, and only one company, or 2.3%, has reached AI readiness. The sample is small, and the results cannot be extrapolated to the entire market. Nevertheless, they point to the main risk: if data is incomplete and processes are not standardized, AI will only accelerate work with poor-quality information.
Analysts advise starting implementation not with choosing a service, but with a goal: reducing time to fill vacancies, cutting manual work, or planning training more accurately. Then it is worth checking the quality of the source information, testing the tool on a small scale, and ensuring the legality of transferring personal data to external AI services.
Poleshchuk believes that HR decisions — the final choice of candidate, dismissal, changes to working conditions — should remain with humans. Regulation also requires this: the law "On Artificial Intelligence" in some cases obliges informing citizens about automated processing, and certain provisions govern work with personal data. If the algorithm decides and the HR specialist merely formally approves the outcome, the company assumes managerial and legal risks.
Where else AI is being implemented in Kazakhstan
The state applies AI not only in the labor market but also within its own apparatus: the authorities intend to use AI to oversee budget planning, execution, accounting, and auditing. Here the same question arises as in recruitment: who ultimately bears responsibility.
The same approach is penetrating lending, where AI is used for ESG screening of borrowers, that is, assessing the environmental and social risks of their business. For companies, this means access to financing will increasingly depend on the completeness and transparency of data about themselves. A similar balance of benefits and risks is being discussed in Kazakhstan's agricultural sector, where AI is viewed as a tool for improving farm efficiency. In all these industries, as in HR, the outcome depends not so much on algorithms as on data quality and people's willingness to verify the conclusions.
Cryptalist analysis: The Kazakhstan case shows that the state is acting as a driver of AI adoption in HR faster than private business: the pilot employed 3,000 people, while only 2.3% of surveyed companies have reached readiness for the technology. For investors, this is a signal of emerging demand for HR-tech and data management solutions in the region, but scaling will be constrained by low digital maturity and regulatory requirements for personal data processing.