we stopped asking "how good are you with AI" and started watching — the gap is wild
building aisa.to -- an AI that assesses peoples AI skills through conversation instead of quizzes.
the thing that keeps surprising us: when you ask someone to rate their AI skills, most say 7 or 8 out of 10. but when you actually watch how they work with AI in a conversation, the patterns tell a completely different story.
the biggest gap we see isnt prompting (everyone thinks thats the skill). its output verification. like 80% of people we talk to just accept whatever the AI gives them. no checking, no iteration, no "wait is this actually right." they copy paste and move on.
we score across 11 criteria and the one that consistently separates good from great is critical thinking -- not technical knowledge, not prompt sophistication. just the habit of questioning AI output before using it.
other thing we didnt expect: the people who are most skeptical of AI often score higher on critical thinking than the power users. being cautious is actually an advantage if you channel it right.
still early but this observation is shaping how we think about the whole product. curious if other builders have found similar gaps between what users say they do vs what they actually do?
This is a strong insight because “AI skill” is becoming too vague to measure by self-reporting. Watching how someone verifies, questions, and improves AI output feels much closer to the real capability companies will care about.
I’d probably lean harder into that positioning: not an AI quiz, but an AI judgment assessment. The valuable signal is whether someone can use AI without blindly trusting it.
One thing I’d pressure-test early is the name/domain layer. aisa.to is short, but for a product that may sell into teams, hiring, training, or internal enablement, the brand needs to feel credible before the assessment even starts. The current name may need extra explanation, and the .to extension can add a little trust friction for a serious skills product.
Beryxa .com would fit this direction better as a broader AI skills intelligence or workforce assessment brand. It gives the product room to expand beyond individual conversations into team benchmarking, training paths, hiring screens, or enterprise AI-readiness reports.
The product is measuring trust and judgment, so the brand itself should probably signal trust and seriousness from the first touch.