AI vs human intelligence: who's ahead?
Status as of 2026-09-05
- METR Time Horizon 1.1 (2026-01-29): the task length AI agents complete at 50% success has been doubling roughly every 4.3 months since 2023 (about every 3 months since 2024) — the autonomy axis is the one moving fastest.
- Microsoft Research, “Working with AI” (Jul 2025, CC-BY-4.0): even the most AI-applicable of 785 occupations — Interpreters and Translators — scores 0.49, i.e. about half its work activities were observed being done with AI. Observed applicability, not job loss.
Start here if the question is about your own job: Should you be worried about AI taking your job? →
It's not one axis — it's a jagged frontier
"Is AI smarter than humans?" assumes a single scale. Real intelligence is jagged: AI is already superhuman at some things and stubbornly subhuman at others, often at the same time.
| Dimension | Where it stands (mid-2026) |
|---|---|
| Knowledge recall & breadth | Superhuman |
| Speed & scale | Superhuman |
| Scoped knowledge work / coding | Near-to-above skilled human |
| Reliable long-horizon autonomy | Below human |
| Accountability for real-world outcomes | Below human |
So AI has surpassed humans on the axes benchmarks measure well, while humans keep the lead on the axis that's hardest to measure and matters most for real work: reliably owning a task end-to-end, unsupervised.
Why the gap that remains is the whole game
That last gap isn't a detail — it's the definition of AGI. A system that matched humans on reliable autonomous work would be the "drop-in remote worker," which is exactly the bar the AGI-2027 prediction is graded against. Until AI closes it, "smarter than humans" is true on the benchmarks and false where it counts. This is the same reason ChatGPT isn't AGI despite its scores.
How to watch the comparison shift
The frontier moves fastest on the autonomy axis, and that's the one that changes the answer. The AGI-2027 Thesis Tracker distills whether AI is closing the human-advantage gap into one auditable score. It resolves the headline question by January 1, 2028. Where do you land on that bet? The Future Bet maps your own predictions onto the real forecaster spectrum in 60 seconds.
AI leads on the benchmarks, humans on autonomy — so when does that flip?
One tap — see which real forecaster you side with, instantly. No sign-up.
Frequently asked questions
On some axes, yes — AI is superhuman at knowledge recall, speed, and scale, and near-to-above skilled humans on many scoped tasks (~83% GDPval). But it's below human on reliable, autonomous, long-horizon judgment — the axis that matters most for real work and that defines AGI.
Intelligence is jagged, not a single scale. AI already exceeds humans on breadth, recall, speed, and many scoped tasks, while humans keep the lead on reliable end-to-end autonomy and accountability for real-world outcomes. The comparison depends entirely on which axis you measure.
Reliably own an open-ended job end-to-end without supervision, take accountability for consequential real-world outcomes, and act with physical dexterity and deep interpersonal trust. These are the axes where human intelligence remains ahead.
'Overall' hinges on the autonomy axis, which is what AGI measures. Forecasts for AGI span 2026–2047, clustering around 2030–2033; the headline AGI-by-2027 verdict resolves by January 2028.
The live scorecard updates as models ship and verdicts change.
View the live scorecard →Be first to know when a verdict flips
The human-vs-AI gap narrows on one axis — subscribe to hear when it closes. Free, no hype.
Subscribe free →