AI Skills: Companies want them, and places try to offer them, but what exactly?

Poster-sized bingo cards for AI buzzwords, including but not limited to Agent AI, democratic AI, augmentation, automation, and super intelligence, have AI skills, the highest concern for both businesses and governments around the world.

According to our IT services company UST, more than three-quarters of companies face AI skills shortages. According to a PWC study, skill requirements for sectors exposed to AI change 66% faster than other sectors.

To close this gap, countries have begun launching national luxury programs to better adapt their respective labor forces to the changing needs of the labour market. These programs may be increasingly relevant in the foreign direct investment environment as local workforces begin to consider the ability of the local workforce to use AI tools to drive business outcomes. Over the next few years, the definition of a skilled workforce will likely evolve, incorporating the ability to leverage AI.

But as everyone wants to learn AI skills, important questions are often left out of discussion: what are they exactly? Are there any benchmarks that can measure people's skills? When can a country, or workers, claim to be AI literacy rate?

Answering these questions is important for the designers and beneficiaries of AI Upskills programs. If specific outcomes are not outlined, governments are wasting money on programs with undefined objectives, missing out on important opportunities to educate the workforce and attract investment.

What is AI Upskill?

Last summer, the UK government announced partnerships with leading technology companies such as Amazon, BT, Google, IBM, Microsoft and Sage. This was followed by the government's AI Opportunity Action Plan launched in January, and presented as a roadmap for the UK to become a global leader in AI.

The Ministry of Business and Trade claims that each participating company “convene different areas of expertise in its work,” and Microsoft has already shattered 1 million workers by the end of 2025. Another initiative, the £187 million ($253 million) program called TechFirst, aims to train AI skills training to “classrooms and future AGEs and future AGE people.”

Glynn Townsend is SAS's Senior Director of Education Services and is one of the providers that partner with the government to provide this training to 7.5 million workers. He emphasizes that AI literacy is a difficult term to define, but given that most people encounter AI using large-scale models (LLMS), the ability of a person to understand the potential benefits and limitations of chatbots can be one central aspect of the term.

“It understands the bias of your model. Here [that the model has been trained on] He outlines. “It can be interrogated and confirmed that there is a line of accuracy through it.

However, this is primarily at the consumer level. “AI literacy from consumers is very different from those who are trying to build a model compared to what they do on a daily basis,” Townsend adds.

Rob Woodstock, managing director of Technology Consultancy Slarom, says understanding the role of workers within a company's broader AI recruitment strategy is important for the type of training they should receive. The three broad categories he outlines are “C-Suite executives who need to understand the possibilities of AI to enhance). [their] Business”; average corporate workers who can use AI when “driving results or providing services” and engineers and AI makers who need to “have a very high level of understanding how LLM works and how people can build infrastructure that can use AI to deliver change.”

At the same time, both Woodstock and Townsend emphasize that learning AI skills must be an ongoing process given the rapid pace of AI evolution and integration into businesses. For example, the rise in agent AI could overturn what AI literacy means in the coming years.

“Major cultural change needs to be around continuous learning throughout life,” Townsend emphasizes. “This is not a one-off activity, because the speed at which this technology is moving is always continuous learning every year.”

Work Questions

One obstacle to deploying skill training is the level of people's trust in AI systems. Woodstock and Townsend say they can mix reactions. Some people object to using AI on ethical grounds (Townsend said it expressed doubt about using AI because its customers have an environmental impact). Additionally, workers may be reluctant to adopt these tools, especially given the explicit warnings from high-tech CEOs that it threatens employment. According to a Trade Union Parliament poll, 51% of UK adults are worried about the impact AI will have on their jobs.

These fears are not unfounded, as recent research on the impact of AI on the workforce suggests that younger job seekers are already facing a more demanding labor market for AI. Stanford economists have found that entry-level employment in the most AI-exposed sectors had experienced a 16% reduction in the US between late 2022 and mid-2025, when ChatGPT first launched. Meanwhile, more experienced workers in these same industries are experiencing more opportunities.

Multiple experts interviewed Investment Monitor It claims that the economic and political upheavals caused by AI are comparable to the chaos caused by technology in the second half of the 20th century. Computer introduction, Townsend highlighted, office typing pools became obsolete, but in the end people found new jobs and the market was adjusted.

In fact, the World Economic Forum estimated in January that AI would transform the global workforce and create 170 million new jobs by 2030. At the same time, it means that 92 million jobs have been destroyed and 78 million jobs worldwide have increased by 7%.

“I think there's a very short-term confusion because we don't know what the impact will be, but I think it'll be resolved very quickly,” Townsend says.

Fabien Braeseman, AI & Work researcher at Oxford Internet Institute, Investment Monitor He also interprets this study as the short-term impact of new technologies, and interprets the job market as being ultimately adjusted.

“I think this is a short-term observation and people who are still training to go to these [AI-exposed] Jobs is currently seeing a shift in technical demand and requirements. “AI will become as standard tools as modern smartphones.”

For Braeseman, another consideration for how AI affects the workforce is about demographic changes over the years. Birth rates are the lowest ever in England and Wales. This means that in the long term, fewer people are working ages and the majority of the population will retire.

“Modification of demographics is a slow process where we have already reached a state of importance in certain occupations in certain places, as there are not so many people in all of these jobs,” he outlines. “Maybe everything will not change more or less so that AI can help us become more effective.”

But as humanity CEO Dario Amodei warns them to do so, if companies really automate most entry-level jobs in the coming years, this could create a vacuum for younger workers as well. If new opportunities and proper training don't appear quickly enough, a greater unemployment crisis could be looming.

Mark Graham, professor of internet geography at the Oxford Internet Institute, Investment Monitor: “Young workers risk losing an entry-level “stepping stone” task that helps them build their careers. ”

AI enthusiasts acknowledge that some jobs are eliminated or automated, but in many cases they argue that workers create more jobs than they do simple jobs. With more time, people can focus on dealing with higher levels of issues in their industry. Graham emphasized that this is not necessarily the case.

“For example, Amazon warehouses use AI systems to track productivity on a minute basis and automatically flag tasks that raise concerns about job enhancement and work quality. Rather than freeing people to think at a higher level, the technology there often narrows down discretion and increases surveillance. Investment Monitor.

In a world where Stanford University research reflects the onset of structural problems in youth unemployment, the role of AI luxury programs may become even more important, as well as adjustments to the short-term labor market for new technologies.

However, Graham argues that in this scenario, the UK should also expand its redistributive policy. “The catch is that most of the large winners of AI companies are based outside the UK, limiting what the UK can raise through corporate taxes. “It's important to lift, but it can't solve the structural reality of less work.”

What does it look like to employ a successful AI upskills program?

In August, MIT's network agents and distributed AI projects discovered that 95% of generative AI pilot programs were unable to drive revenue and changed little or no profit and loss changes. Research shows that the main factor in this was the “learning gap” between the tools and the organization. Companies that have most successful in using AI to drive revenue tend to buy external AI tools rather than building internal tools, focusing on automating backend processes.

So, what will create a good AI upskill initiative?

“The biggest success I've seen is where we start with business results and work on the other side,” Woodstock points out. “If it's a common 'let' try to get everyone to use AI More', that was a very complicated outcome. And when there was a level of uplifting or personalization of the type of service we want to offer, it was transformative. ”

Townsend also emphasizes the importance of focusing on outcomes as a way to demonstrate the benefits AI can bring to workers, particularly to the workforce.

“I think focusing on specific outcomes and how it increases productivity is the biggest change you need to make when you're communicating this to people,” he outlines.

Therefore, national high-class initiatives should be supported by one question. What are they trying to achieve? Otherwise, Woodstock is warned that they are at risk of being at risk of “generic” adoption, and workers are told to use more AI. When AI hype falls, it fails to impress foreign investors if the initiative is not outcome-driven.

Given the rapid pace of change, the biggest lesson for governments and businesses is that training programs could remain here.

As Townsend says, “We can't rely on the skills we learned at age 21 and then we'll do it. That's not how it works now.”