Artificial intelligence (AI) is fundamentally changing the way we live and work, and employees are not as prepared as they think.

Udemy’s new research, conducted by YouGov, surveyed adults in four major economies to identify current trends in career skills readiness and the growing impact of AI. The findings reveal two major trends: First, workers are more worried about AI’s impact on the wider economy than about its impact on their own jobs. Second, a majority of workers overestimate their soft skills, such as communication and teamwork.

These gaps between perception and reality could indicate a psychological pattern called optimism bias, or our natural tendency to be more optimistic about our personal circumstances and abilities, even when we see challenges affecting others.

Some research suggests that optimism bias serves an important purpose: it can reduce anxiety, increase motivation and confidence, and it can help people adapt to uncertain environments. Optimism bias is pervasive and observable in a variety of contexts, and it likely provided our ancestors with evolutionary advantages. However, this tendency toward optimism can also have negative consequences, causing people to underestimate risks, and inflating feelings of competence, especially when we rely on self-assessment rather than feedback from others.

Understanding how optimism bias shapes workers’ perceptions and behaviors is crucial for preparing them for the inevitable workplace disruptions AI will bring. Here’s what you should understand:

Optimism Bias and Training Avoidance

Across all surveyed economies, workers experience a disconnect between awareness of AI’s broader impact and sense that it could impact them personally. Notable gaps are observable in the U.S. and UK. When asked about AI disruption at work, workers in these countries were 1.5 to nearly two times more likely to fear an economy-wide job impact than a personal career impact. And, while a majority of workers in these markets acknowledge they lack essential AI capabilities (only 14% of U.S. workers and 16% of UK workers report adequate AI skills), they’re also not taking advantage of available training. Over half of UK workers and nearly half of U.S. workers report that they’ve pursued no training in AI.

On the other hand, adults in India have the smallest gap between economy-wide and personal AI concerns, and they pursue AI training at significantly higher rates – only 14% of workers report no AI training. The difference between the U.S./UK and India suggests that when workers have a more balanced view of AI-related threats, they may be more interested in training. What’s more, finding ways to address optimism bias could help increase motivation to upskill in AI.

Soft Skills and AI Readiness

Udemy’s research also found that entry-level workers in all economies surveyed tend to overestimate their soft skills proficiency. For example, in the U.S., nearly half of hiring and training managers cite a lack of communication skills among entry-level workers, while 59% point to a deficit in critical thinking, and 41% report inadequate teamwork skills. Meanwhile, entry-level workers surveyed largely do not perceive these same gaps in themselves, with only a small fraction acknowledging a need for improvement: 14% for communication, 15% for critical thinking, and only 7% for teamwork.

This surfaces a takeaway that might sound counterintuitive: When developing AI training programs, smart organizations will include a focus on soft skills. As AI continues to automate more mundane tasks and humans engage in more cognitive “off-loading” to AI tools, we need to take care to reinforce critical thinking skills so they do not atrophy. AI adoption will require human oversight and collaboration to meet its full potential. From governance and ethics, to designing and building effective workflow integrations, workers with strong adaptability and soft-skills foundations will see the greatest success with AI training initiatives.

Strategizing for Effective AI Learning and Development

As Udemy’s research suggests, gaps in worker readiness stemming from potential optimism bias and inflated perceptions of soft skills could pose a hurdle to effective AI learning initiatives. By viewing these findings through the lens of psychology, training professionals can begin to bridge the gap between AI awareness and AI readiness.

To prepare workers for an AI-driven future, here are some strategies that address the root of these behavioral patterns:

• Begin with detailed, 360-degree skills assessments to expose blind spots. When optimism bias is at work, self-assessments won’t cut it. External feedback helps individuals recognize discrepancies between perceived and actual skill levels.
• Build AI training around existing workflows. Workers see that AI is impacting work, but they might not see how AI fits into their This will allow workers to actually visualize AI in the context of their day-to-day assignments, making it easier to see the value and where they might need to learn new skills.
• Prioritize soft skills training alongside AI training. This can help workers better integrate technical and human skills, enabling them to effectively collaborate with AI tools and teams.