Employers are looking for more than technical proficiency, placing greater emphasis on specialist knowledge, analytical thinking and decision-making. For fresh graduates, the challenge is how to gain experience as AI takes on more routine tasks.
Artificial intelligence is changing how people work across a wide range of professions, from researching information and writing content to preparing reports and conducting basic data analysis. Tasks that once took hours can now be completed much faster with AI tools. But for jobseekers, an important question remains: as AI becomes capable of doing more, what skills do employers still need from people?
An analysis by the International Monetary Fund (IMF), presented during the IMF-World Bank Annual Meetings in Bangkok from 12 to 18 October 2026, found that hiring demand in occupations highly exposed to AI fell by about 10% before beginning to stabilise, even as AI capabilities continued to advance. The analysis drew on millions of online job advertisements across Asia.
The trend suggests that AI is changing not only how work is done, but also the qualifications employers look for in candidates. Relevant knowledge remains essential, but the ability to use new tools and turn their outputs into practical results is becoming increasingly important.
Knowing how to use AI is not enough
Natthawat Thanatphongjira, Senior Account Manager at Workwise Consulting, said demand for AI-related skills was increasingly appearing in job requirements provided by clients seeking recruitment services.
These requirements were no longer limited to information technology positions. They were also appearing in marketing, sales, human resources and some office-based roles.
However, being able to use AI does not necessarily mean a candidate can meet all an employer’s expectations. AI tools can quickly retrieve information, suggest answers and conduct preliminary analysis, but people are still needed to check whether the results are accurate and appropriate for real-world situations.
“AI can analyse data or suggest answers very quickly, but ultimately someone still has to check whether those answers actually work,” Natthawat said.
For example, marketing staff may use AI to research information and draft content, but they still need to assess whether it suits their target audience and business objectives. Employees responsible for coordination may use tools to summarise sales or stock data, but they must still spot irregularities and know which department to contact to follow up on problems.
The skills employers seek therefore extend beyond familiarity with AI tools. They also include specialist knowledge, analytical thinking, data verification and the ability to decide how to use the results.
Fresh graduates may face a tougher challenge
Fresh graduates and people with just one to three years of work experience are among the groups worth watching closely. Tasks that once provided an entry point for learning in the workplace, such as research, report writing, document preparation and basic data analysis, are among those AI can increasingly help perform.
A study by the Stanford Digital Economy Lab, using US employment data through June 2026, found that employment among people aged 22 to 25 in occupations highly exposed to AI was about 19% below the expected level, compared with employment trends among people of the same age in less AI-exposed occupations. The difference was largely driven by a decline in new hires.
However, the findings are based on the US labour market and do not establish that AI alone caused the decline. They should not be taken as direct evidence of the situation facing fresh graduates in Thailand.
From a recruitment perspective, Natthawat said companies might not stop hiring fresh graduates, but their expectations of entry-level employees could change. As AI speeds up basic tasks, employers may expect new recruits to learn faster and take responsibility for more complex work sooner.
The challenge is that routine tasks do more than simply get work done. They also give new employees opportunities to learn workplace procedures, practise problem-solving, understand mistakes and receive guidance from supervisors before developing the ability to make decisions independently.
Natthawat noted that if companies reduced the tasks traditionally used to train new employees while continuing to expect stronger decision-making skills, they would need to find other ways to help young workers gain experience.
“If we keep taking away the tasks that new employees used to learn from, while still expecting them to make better decisions, companies also need to ask where the next generation will gain that experience,” he said.
Conversely, if organisations use AI to reduce time spent on repetitive work while giving new employees opportunities to learn from real problems, the technology could help them develop their skills faster. Experienced colleagues would still need to provide training and explain the reasoning behind important decisions.
Experience is about more than years on the job
As companies seek people who can solve problems and make decisions independently, work experience is no longer simply a matter of how many years someone has held a position. It also reflects what they have learnt and the responsibilities they have taken on.
Natthawat explained that when an employer asks for a candidate with three years of experience, it does not necessarily mean the company is looking only for someone who has worked for that length of time. Employers also want to know what problems candidates have encountered, the level of responsibility they have handled and what decisions they can make on their own.
“They want to know what problems you encountered during those three years, what you were responsible for, and what you can now decide for yourself,” he said.
For jobseekers, particularly those with limited experience, preparation should go beyond listing the software or tools they know how to use. They should be able to explain how they applied those skills, what problems they solved and what results they achieved, based on their actual experience.
Even without a full-time job, candidates can draw on internships, projects or relevant activities to demonstrate their thinking and problem-solving abilities, provided these experiences relate to the position they are applying for.
Knowing how to use AI may no longer make candidates stand out
As AI becomes more widespread, the ability to use these tools may increasingly become a basic skill employers expect, rather than a standalone advantage that sets one candidate apart from others.
What differentiates candidates may instead be how they build on AI-generated results, whether by verifying information, solving problems, understanding the business or taking responsibility for more complex tasks.
Jobseekers should therefore develop a combination of skills, including specialist knowledge, analytical thinking, communication, problem-solving and the ability to learn as tools and working methods change. Those already in employment should also use practical experience to take on greater responsibility and learn from their work, rather than relying on technology alone.
Ultimately, the key question in the age of AI is not simply whether technology will replace human workers, but how people can use these tools effectively and what additional capabilities they need to develop to handle increasingly complex work.
“Instead of asking only whether AI will take our jobs, I think we should ask another question: which parts of the job do companies still need us to do?” Natthawat said.
The post What do jobseekers need in the age of AI? Knowing how to use it may not be enough appeared first on Khaosod English.