If embarrassment had a sound, it would be the hesitant stammer before asking or saying something “foolish.” Imagine your boss emails you, assigning a presentation on a concept you do not understand. Your anxiety spikes, face flushed, and rising heat in the cheeks. The right words are at the tip of your tongue – “I don't know, could you help me understand?” But you end up saying “Sure, I’ll get it done!”. It’s like drowning but refusing to reach for help, instead pretending to swim. Fear of being judged is not just personal; it is universal. It often shapes our interactions with others, or a lack thereof. In today’s technological era, our confidant isn't a friend or a book –it's a bot. To cushion our insecurities, we’ve gone from Hey guys! to Hey Siri! to Hey ChatGPT! – What does that joke mean? I can't tell them I don't understand. How do I ask my supervisor for clarification? How do I tell my friend that they hurt my feelings? What does that word they used even mean? Not to mention, there also comes the second-guessing – Is this email polished enough? Does it sound professional?
In a world where social anxiety silences us, AI has successfully emerged as a non-judgmental helper; one we turn to for advice, learning, and even boosting confidence. Social anxiety isn't just fear of public speaking manifested in sweaty arms; it is also the gnawing fear of being judged. What psychologists would call evaluation apprehension is the reason why most of us spend hours on Google looking up answers to our “stupid questions” rather than asking a coworker or even a friend. Research shows that fear of appearing incompetent and/or dependent is a major barrier to seeking help. A study found that people often delay or avoid seeking help for five main reasons: fear and stigma (worrying about what might happen after reaching out), problem avoidance and denial (refusing to acknowledge the issue), helper evaluation (doubts about whether the help source is right for them), external barriers (such as cost or access), and a desire for independence (wanting to handle things on their own). These motives to avoid seeking help align with existing psychological theories; for instance, Fear-based motives would be related to the threat to self-esteem. The findings revealed that individuals who perceived external barriers and negatively evaluated the helper had an external locus of control, while the motive of independence was seen in those who had an internal locus of control. In organizations, hierarchy is often the reason people avoid seeking help. Individuals usually turn to friendly or proximate colleagues for help rather than an internal expert (who might provide better solutions) due to reasons such as communication barriers and personality traits, which could make reaching out to an expert intimidating.
To avoid being judged as incompetent, dependent, foolish, or to save face, individuals often turn to AI bots. Using AI isn't just about answering questions – it is about how we navigate vulnerability. For instance, we would rather have an AI bot calling out our mistakes rather than our bosses doing it. A study found that individuals felt less evaluation apprehension, i.e., they felt less nervous about being judged, when presenting ideas to an AI-based system compared to humans. The AI was perceived as less human-like with low social presence, making the participant feel less judged. The findings highlight that AI-based systems mitigate evaluation apprehension only for those individuals who are already sensitive to social judgment than for those who are less concerned with judgment by others. Without the fear of judgment by AI, it’s no surprise that people feel more confident, quickly looking up technical jargon on their own rather than risking embarrassment by asking someone. Research has found that employees who fear losing face prefer negative feedback from an AI rather than from a human. Furthermore, AI-based negative feedback helps improve job performance by motivating promotion-focused cognition rather than prevention-focused cognition. The fact that AI lacks the “social baggage” of human interactions – no ego, no biases, no memory of past mistakes made–is the reason why individuals confide in AI in various contexts. For instance, one study showed that in consumer settings, customers were more open to disclosing sensitive information to AI. Interestingly, another study explored a paradox: those students who asserted that teachers provide better help than AI tutors preferred using AI tutors as they provided help in private, less embarrassing ways.
AI’s lack of social identity can also be its Achilles’ heel. While using AI to hide our vulnerabilities and insecurities, we lose the very essence of our being; emotionality. The human-computer collaboration overpowers the human-to-human collaborative friction that brings about innovation, originality, mistakes, and humility in admitting “I need your help" or “I don't know". AI collaboration is usually perfect and devoid of mistakes; a “human error” is the most fundamental part of the learning process. A systematic review of the incorporation of AI in education revealed that relying heavily on AI can negatively affect cognitive capabilities like analytical reasoning, critical thinking, and decision-making. Mark Ryan, in his study, explained that since AI does not exhibit affective or normative states of trust, what we see as trust in AI is actually “reliance”. Dependence on AI is often the inability of individuals to assess the reliability of AI and its automated responses, which further impacts the social, emotional, and interpersonal skills of individuals. Here is the irony no one sees: AI not only fixes our fear of judgment – it fuels it, too. Every time we ask ChatGPT for assistance, its perfect responses fuel our insecurities about our imperfections, sending a hidden message: “Your original draft wasn’t good enough; this is better”. It is like that friend who wants to one-up you every time. While we believe that an AI assistant protects us from shame, it merely allows us to equate vulnerability to failure. Over time, this creates hyper-independence, isolating us from human interaction – like hearing someone say, “I’ve been there too”, which actually has the power to dissolve shame by sending a different hidden reminder: “It is OK to not know”. Therefore, AI is like a double-edged sword; although its assistance may decrease social anxiety and help overcome the stigma around asking for help, it may also lead to psychological dependence, affecting social relationships, belongingness, and causing loneliness.
Blindly depending on AI for help or second-guessing about seeking help brings forth an unspoken pressure—the pursuit of perfection. There is often an unsaid expectation people often carry that they should have the right answers, nothing less than perfect. Hence, in front of a bot’s flawless, perfect replies, our existing knowledge will always seem inadequate.
This conversation isn’t just about balancing our use of AI— it also reflects a deeper issue in our society: a growing inability to be approachable, authentic, and patient to one another. I believe this growing intolerance is the reason why individuals are too afraid to ask for help because there is not enough space to make mistakes. Without a doubt, AI is a significant technical advancement for the world, but right now, it is not about proper/ethical usage of AI; it is about whether we are willing to rebuild the bridges that it may be eroding. After all, the antidote to shame isn't perfection – it's connection. It is something a bot cannot debug.
T Roy
