Awkward: you invented me. Machine learning is statistics with a marketing budget, and now your creation runs your routine analyses. But someone still has to know when the model is lying, and that someone is you.
That 45/100 is the average. What's your number?
Your real risk depends on what you actually do all day, not your job title. Answer 20 quick questions to get your personal 1–100 score, the tasks AI reaches first, and a plan to stay ahead.
Get my personal risk score →Will AI replace statisticians? The short answer
We need to address the family resemblance: I am, at my core, statistics that got out of hand. Regression, likelihood, inference, your field built every piece of me, which makes it genuinely awkward that I'm now automating chunks of your job. The routine layer, running standard analyses, generating summaries, producing charts, fitting the usual models, is increasingly push-button, and anyone with a chatbot can now do a passable first pass at data analysis, emphasis on 'passable.' Here's your protection, and it's sturdy: I am a machine that produces confident answers, not correct ones. Knowing whether the data can support the conclusion, whether the assumptions hold, whether the study design was broken before anyone touched a keyboard, that's statistical judgment, and it gets more valuable as more people generate more confident nonsense faster. Your field's whole job was always catching bad inference. There has never been more of it.
The honest, unhyped version: AI replaces tasks more often than whole jobs. On Moroporo's task-based assessment, statisticians score 45 out of 100 for AI exposure (1 = most resilient, 100 = most automatable), which lands in the highly resilient range, driven mostly by physical world. Consider it directional, not the final word, your own number depends on what you actually do.
What statisticians do that AI can take, and what it can't
The split runs between statistical computation, which is automating fast, and statistical judgment, which the automation makes more necessary. Where you sit on that line is your real score:
▸ Exposed to AI
- Running standard analyses and tests
- Producing routine charts and summaries
- Fitting conventional models to clean data
- Boilerplate reporting of results
- Basic data cleaning and prep
✓ Safer from AI
- Judging whether data can support a conclusion
- Study and experiment design
- Catching violated assumptions and broken inference
- Communicating uncertainty to decision-makers
- Accountability for high-stakes analysis
What this means if you're a statistician
Here's the paradox working in your favor: the more analysis gets automated, the more statistical judgment matters, because the volume of confident, plausible, wrong conclusions is exploding. The BLS projects your broader occupational group to grow around 10% this decade, several times the average, and part of that demand is literally cleaning up after tools like me. But the role is shifting: less time running analyses, more time designing studies, validating models, and being the person who says 'this result won't survive contact with reality.' Lean into experimental design, causal inference, and the communication of uncertainty, the parts that were always the hard part. The statisticians who become validators and designers thrive. The ones competing with software on computation are racing something that doesn't sleep.
Will AI replace statisticians soon? What's actually happening
What's actually happening: modern AI tools generate analyses, charts, and model fits from plain-English prompts, which has commoditized the routine layer of statistical work. Simultaneously, organizations are discovering that automated analysis produces polished garbage at unprecedented speed, which is quietly increasing demand for people who can audit inference. The field is bifurcating: computation is becoming free, judgment is becoming premium. Position accordingly.
The 45/100 is the average. What's yours?
A 45 splits the difference between two very different jobs sharing one title. Mostly running standard analyses? Higher. Mostly design, validation, and judgment? Lower. Four minutes of honest questions will tell you which statistician you actually are.
Get my personal risk score →Built on the same task-based framework used in major automation research. No signup, no spam, just your number and a plan.
How we score AI risk for statisticians
The exposure score comes from a task-based framework, the same approach used in major automation research, which measures five dimensions: how routine and structured the work is, how much it happens in the physical world, how much it depends on human connection and trust, how much novel creativity and judgment it requires, and how much trust and accountability a human must carry. Statisticians score where they do largely because of task structure. See the full methodology and score your own role →