Everyone seems better at AI than me - am I too far behind?

Almost certainly not, and here’s the honest measurement: the "AI fluency" you’re envying in colleagues mostly decomposes into weeks of deliberate practice, not years of expertise - because the tools themselves are only a few years old and change so fast that nobody’s head start is deep. What your fluent colleague actually has: daily habit (they use the tools on real work), calibrated judgement (they know when output is wrong), and a few workflow tricks - all acquirable in a focused month. The apparent gap gets inflated by the same filter that inflates everything: you see their confident output, never their fumbling first fortnight. Behind, yes, perhaps. Far behind - measured in catch-up time - no.

My friend got promoted for AI skills - what did they actually do?

Strip the envy and study the mechanism, because it’s replicable. In almost every "promoted for AI" story we hear from hiring partners, the promoted person did three unglamorous things: used AI daily on their actual job until fluency was real (not performative), automated something the team visibly hated - the report that assembled itself, the process that stopped eating Fridays - and became the person others asked for help, which management reads as leadership. None of that is genius; all of it is available to you starting today. Envy is just ambition without a plan attached - the plan is the part you can add this week.

How long does it take to catch up on AI skills?

For workplace fluency - the level most colleagues you envy actually hold: about 30 days of deliberate daily use. Week one: one frontier tool on your real tasks daily, noting where it fails. Week two: push into your job’s core workflows - drafting, analysis, summaries. Week three: automate one team-visible annoyance. Week four: help someone else, which cements both the skill and the reputation. For the builder tier - the one carrying the RM 6,000-9,000/month starting premium in Malaysia - the honest budget is months of part-time hours, not weeks. But the feeling of being left behind dissolves at the first tier; the second is ambition, not catch-up.

Is it too late to start learning AI in 2026?

Run the numbers against the panic: the tools are a few years old, most companies are still early in real adoption, the fluency premium is still being paid (measurably - thousands of ringgit monthly at junior level), and the deepest skills in the field are measured in months of practice, not decades. You’re not boarding a departed train; you’re early-middle in a transition where the spreadsheet-era precedent says fluency will take a decade to become fully assumed. What IS true: the differentiation window narrows every year. Late enough to stop procrastinating, early enough that starting now still puts you ahead of most - that’s the honest position.

Should I feel bad about being behind on AI?

Feel it briefly, use it precisely, then drop it - because the feeling’s only value is as fuel. You’re "behind" for ordinary, forgivable reasons: the tools arrived mid-career while you were busy doing your actual job, and nobody scheduled adaptation time. The colleagues ahead of you simply started their fumbling phase earlier. What deserves genuine concern isn’t the current gap (weeks) but the compounding one: every month of feeling-bad-without-starting widens what a month of starting would close. The emotionally honest move is the practically optimal one - convert the discomfort into the first 30 minutes tonight, and watch how fast "left behind" becomes "catching up," which is a completely different feeling.

All posts

AI & your future · Careers · Anxiety to action

Feeling left behind by AI? Measure the gap first.

The colleague whose reports write themselves. The friend promoted for “AI initiatives.” The meeting where everyone nodded at a tool you’ve never opened. The feeling is real - but before it calcifies into “too late,” measure it: the fluency you’re envying decomposes into weeks of deliberate practice, not years of expertise - and the 30-day catch-up below closes most of it.

Deric Yee Updated 25 August 2026 9 min read

Deconstruct the person you’re envying

Take the most AI-fluent colleague you know and audit what they actually possess. Not a computer-science degree acquired in secret. Not years of machine-learning study - the tools they’re fluent in barely existed a few years ago, which caps everyone’s head start. What they hold is: a daily habit (they reach for the tool on real work, so fluency compounded), calibrated judgement (a few hundred interactions taught them what AI nails and where it confidently fails), and a handful of workflow moves(the prompt patterns and integrations that fit your specific workplace). Total acquisition cost: weeks of deliberate use. That’s the entire moat - and the reason it looks like more is the usual filter: you see their polished output and never saw their fumbling first fortnight, the same rigged inside-versus-outside comparison we dissect in am I smart enough for tech.

The promoted friend? Same audit, one addition: visibility. In nearly every promoted-for-AI story that reaches us through hiring partners, the mechanism was unglamorous - daily use until fluency was real, then one team-visible automation (the report that assembles itself, the Friday-eating process that stopped), then becoming the person others ask. Management read it as leadership; it was mostly earliness. Envy, examined closely, is just ambition without a plan attached. The plan is the part this page adds.

The 30-day catch-up (to parity), then the overtake

Week one - habit:one frontier tool (Claude or ChatGPT), used on your actual work every day - drafts, summaries, analysis - with one rule: note every failure. The failures are the curriculum; noticing them is the judgement your colleague has and you’re building. Week two - depth:push into your role’s core workflows, not toy tasks. This is where fluency stops being performative. Week three - visibility: automate one thing your team visibly hates. This single move is what converted your friend’s fluency into a promotion; it works because everyone can see it. Week four - teach:help one colleague who’s where you were a month ago. Teaching cements the skill and quietly repositions you from behind to ahead - the full working-adult ladder, with what each level is worth, is in how to upskill for AI.

Thirty days closes the gap you’re feeling. Then there’s the overtake, for those whose envy turns out to be ambition: the builder tier- actually creating tools, automations, and products with AI - is the level most of your fluent colleagues will never climb to, and it carries the market’s documented premium: RM 6,000-9,000/month starting for AI-capable juniors in Malaysia (the data), reachable part-time in months via the AI-native path. Here’s the quietly satisfying arithmetic: start the builder tier now, and in six months the colleague you’re currently envying is asking how you got ahead - because almost nobody who reaches tool-fluency pushes further, which makes the next tier the least crowded competitive move available. The free trial - one signup, no card, real projects with a live instructor session - is its zero-cost first week, runnable alongside the catch-up above.

One last honesty about the feeling itself: “left behind” is only ever a snapshot, and snapshots of fast-moving transitions expire in weeks. The colleagues ahead of you today were behind someone else six months ago; the transition’s real sorting - documented across every automation wave since the spreadsheet (the full pattern) - isn’t between early and late starters. It’s between starters and spectators. Tonight’s thirty minutes moves you permanently out of the second group, and that group is the only one that actually gets left behind.

Why the feeling lies: the mechanics of AI FOMO

The left-behind feeling has a specific anatomy, and seeing it drawn reduces its power. Your feed shows you a highlight reel of thousands of people’s single best moments- someone’s polished agent demo, someone else’s AI-built product launch - compressed into one scroll, which your brain reads as “everyone is ahead of me”. The base rates say otherwise: across most Malaysian workplaces, genuine AI fluency - using it daily on real work, building small automations - is still rare. The median colleague has typed a few prompts into a chatbot. The gap between you and the demo-reel people is real but shallow: these tools are new for everyone, the interfaces are plain language, and the compounding advantage everyone fears has barely begun to compound. What feels like a widening canyon is, this year, still a few weeks of deliberate practice wide.

The correct response to shallow gaps is speed, not despair - and the arithmetic is on your side. Thirty focused days of daily real-work AI use plus a first build project puts you ahead of the majority of your own workplace; the premium end of the market (the RM 6,000–9,000/month AI-capable band in the hiring data) is months away, not years. The one genuinely dangerous move is the one the feeling pushes you toward: freezing, doomscrolling more demos, letting the shallow gap actually deepen. Close the feed, open one project - the free trial’s six days of real building is a purpose-built first week - and the feeling converts into its only useful form: fuel.

FAQ

Everyone seems better at AI than me - am I too far behind?
Almost certainly not, and here’s the honest measurement: the "AI fluency" you’re envying in colleagues mostly decomposes into weeks of deliberate practice, not years of expertise - because the tools themselves are only a few years old and change so fast that nobody’s head start is deep. What your fluent colleague actually has: daily habit (they use the tools on real work), calibrated judgement (they know when output is wrong), and a few workflow tricks - all acquirable in a focused month. The apparent gap gets inflated by the same filter that inflates everything: you see their confident output, never their fumbling first fortnight. Behind, yes, perhaps. Far behind - measured in catch-up time - no.
My friend got promoted for AI skills - what did they actually do?
Strip the envy and study the mechanism, because it’s replicable. In almost every "promoted for AI" story we hear from hiring partners, the promoted person did three unglamorous things: used AI daily on their actual job until fluency was real (not performative), automated something the team visibly hated - the report that assembled itself, the process that stopped eating Fridays - and became the person others asked for help, which management reads as leadership. None of that is genius; all of it is available to you starting today. Envy is just ambition without a plan attached - the plan is the part you can add this week.
How long does it take to catch up on AI skills?
For workplace fluency - the level most colleagues you envy actually hold: about 30 days of deliberate daily use. Week one: one frontier tool on your real tasks daily, noting where it fails. Week two: push into your job’s core workflows - drafting, analysis, summaries. Week three: automate one team-visible annoyance. Week four: help someone else, which cements both the skill and the reputation. For the builder tier - the one carrying the RM 6,000-9,000/month starting premium in Malaysia - the honest budget is months of part-time hours, not weeks. But the feeling of being left behind dissolves at the first tier; the second is ambition, not catch-up.
Is it too late to start learning AI in 2026?
Run the numbers against the panic: the tools are a few years old, most companies are still early in real adoption, the fluency premium is still being paid (measurably - thousands of ringgit monthly at junior level), and the deepest skills in the field are measured in months of practice, not decades. You’re not boarding a departed train; you’re early-middle in a transition where the spreadsheet-era precedent says fluency will take a decade to become fully assumed. What IS true: the differentiation window narrows every year. Late enough to stop procrastinating, early enough that starting now still puts you ahead of most - that’s the honest position.
Should I feel bad about being behind on AI?
Feel it briefly, use it precisely, then drop it - because the feeling’s only value is as fuel. You’re "behind" for ordinary, forgivable reasons: the tools arrived mid-career while you were busy doing your actual job, and nobody scheduled adaptation time. The colleagues ahead of you simply started their fumbling phase earlier. What deserves genuine concern isn’t the current gap (weeks) but the compounding one: every month of feeling-bad-without-starting widens what a month of starting would close. The emotionally honest move is the practically optimal one - convert the discomfort into the first 30 minutes tonight, and watch how fast "left behind" becomes "catching up," which is a completely different feeling.

Back to all posts Start the overtake free →

The gap is weeks. The overtake is months.
Both start with tonight's thirty minutes.

Thirty days of deliberate use reaches parity with the colleagues you're envying. The builder tier - the one they'll never climb to - starts free: one signup, no card, real projects with a live instructor session. Starters versus spectators. Pick tonight.

Start the free trial The full upskilling ladder