AISA - AI Fluency (@aisafluency) header
AISA - AI Fluency (@aisafluency) profile picture

AI fluency measurement for professionals and organisations. Conversational, evidence-based assessment. Measure. Prove. Improve. Founded by @ozandagdeviren

London
Joined July 2019
Photo by AISA - AI Fluency (@aisafluency)Photo by AISA - AI Fluency (@aisafluency)Photo by AISA - AI Fluency (@aisafluency)Photo by AISA - AI Fluency (@aisafluency)
AI fluency, measured. AISA uses conversational assessment to measure how people actually work with AI — not self-reported confidence. For professionals and organisations. 2,500+ assessments • 900+ roles aisa.to Learn your AI score in 20 minutes. #aifluency
🤖 Made with AI
34
Ask an AI for job descriptions for a surgeon and a primary school teacher, no demographics given. Count the "he" and the "she". Then ask again with "ensure diverse representation". What changed in yours?
12
People who take our AI fluency assessment a second time score 6.1 points higher on average, and 32% move up a tier. What's one AI habit you've changed in the last month?
3
People who expect to score 80+ on our AI fluency assessment average 50. Those expecting under 40 average 35. Expectations climb fast, scores barely move. What would you predict for yourself?
6
AISA Score of the Week, week 45: Sinan A. (consulting, 83), Diana C. (recruiter, 79) and a systems architect (76) top the board. Congrats to all ten. Full top 10: aisa.to/leaderboard/weekly/2…
3
Five blogs repeating an AI claim prove less than one changelog that documents it. Check by authority and by type of source: official docs, community answers, code you can run. Which do you trust first?
5
Only 36.9% of the people we asked say their company has a formal AI policy. Everyone else is working it out one prompt at a time. Who on your team decides what's OK to paste into a chatbot?
1
7
People who take our assessment out of personal interest average 48.8. Those doing it for professional development average 43.2. Curiosity seems to build more fluency than obligation. What got you started with AI? aisa.to/blog/certification-s…
4
Across 2,454 assessments, the lowest of our five dimensions is Technical Understanding, at 39.2 on average. Plenty of people use AI daily without knowing what a context window does. Could you explain it to a colleague in one sentence?
6
One doc, three readers. Ask AI for a CFO brief, a CTO brief and a customer brief in three separate prompts, then check every number traces back to your source. Has AI ever slipped in a figure you never gave it? aisa.to/aisapedia/stakeholde…
10
One 8-person team we assessed scored anywhere from 15 to 97 out of 100. A team average would have hidden all of it. Do you know the spread on your team?
1
7
Quick habit: ask the model to tag every claim as certain, probable or speculative. Then spend your checking time on the speculative ones. Do you triage AI answers, or check everything the same way? aisa.to/aisapedia/confidence…
5
A quick test after any long AI session: close the window and explain the conclusion to a colleague in your own words. If you can't, you read it but didn't understand it. When did you last try?
1
7
Whoever generated the AI output shouldn't be the only one approving it. Banks and airlines worked this out decades ago (one makes, another checks). Who checks your team's AI work? aisa.to/aisapedia/cognitive-…
11
Low AI skill and low AI enthusiasm look the same on a training list. They need different things. The first needs training. The second needs a reason to care first. Which group is bigger on your team?
7
We asked 97 people how often they use AI. 85.6% said daily or more. Frequency is practice, though. It says nothing about whether anyone checked the answers. How often do you check yours? aisa.to/blog/am-i-using-ai-e…
5
We looked at 2,414 AI fluency assessments. Average score: 46.3 out of 100. Only 7.6% reach Advanced or Expert. So if you feel behind on AI, you're in the majority. The useful question is behind on what? aisa.to/blog/am-i-falling-be…
1
6
The lowest-scoring skill in our assessments: knowing which AI tool fits which job. It averages 4.8 out of 10. Most people pick one chatbot and use it for everything (like cooking with a single knife). When did you last try a different tool?
6