AI task length scaling
METR's scatter plot drew 146.4K views on X: 1.0x its typical post and #14 of 26 METR charts we track by views. Read Oct 10, 2026.
We're correcting a mistake in our modeling that inflated recent 50%-time horizons by 10-20% (and reduced 80%-horizons). We inappropriately penalized steepness in task-length→success curve fits. This most affects the oldest and newest models
Why it spread
This correction post from @METR_Evals spread because it explained itself clearly and landed at the right launch timing, drawing 906 likes, 69 reposts, 34 quotes, 186 saves and 146,415 views, well above the account's typical 523 likes.
What worked
- The post leads with a plain admission of a modeling mistake and states the size of the effect in words, which makes the chart easy to trust and share.
- The scatter chart shows task length growing exponentially with model release date, so the visual argument carries the message without extra explanation.
- It is branded, shows a source line, and comes from an account whose source reach sits in the top quarter, giving the correction extra credibility at launch.
What to borrow
We could publish our own chart alongside a short, honest note about what we changed and why, letting the chart show the trend while the text explains the correction.
Careful
Do not copy the mistake framing just to chase attention, and do not over-read the engagement as proof the underlying model is right, since the post itself says the oldest and newest fits are less data-constrained.
Written by our analysis model from the measured numbers above, not by METR.
Make it yours
Your numbers, in this line chart.
Starts from the real numbers of "Task length 50%-time-horizon measurements were inflated by over-regularization" by METR. Replace every number with yours before you post. Make one like this, step by step
Free pictures carry a small "Made with ViralChart" line. That line is how other people find us.
API · Max
Use this chart via API
On the Max plan you can pull this chart's numbers with the provenance of each one, as JSON or CSV, with its redraw spec, or call it from an agent over MCP. Chart id 7566cd16809b.
Approximate values. Our two readings of this chart agreed only in part (confidence 0.47 of 1), so check the original before you quote a number. The API marks it lowConfidence, and ?min_confidence=0.5 leaves it out.
| Path | Returns | Plan | |
|---|---|---|---|
| GET | /api/v1/charts/7566cd16809b | Title, publisher, date, form, point count; with a Max key, the data link | Open |
| GET | /api/v1/charts/7566cd16809b/data | Every point with its provenance and confidence, plus the redraw spec. Add ?format=csv for CSV | Max |
| GET | /api/v1/charts/7566cd16809b/render?view=1 | The draw-only view this page uses | Open |
| POST | /mcp | Tool get_chart_data with this id, for Claude, Cursor and other agents | Max |
curl -s https://viral-chart.com/api/v1/charts/7566cd16809b/data \
-H "Authorization: Bearer $VIRALCHART_KEY"claude mcp add --transport http viralchart https://viral-chart.com/mcp \
--header "Authorization: Bearer $VIRALCHART_KEY"60 requests a minute per key; over that you get 429 with Retry-After. Up to 5 keys per account, made on your account page and shown once. Without Max, the data route answers 402 with a link to the plans. The numbers describe a chart its publisher owns, and every file carries a license note.
More from METR
All 26
Other giants using a scatter plot
All 78
Questions people ask
How far did METR's "AI task length scaling" chart travel?
146,415 views, 906 likes, 69 reposts and 186 saves on X, read Oct 10, 2026. A normal post from @METR_Evals gets about 153K views.
What kind of chart is it?
A scatter plot. Task length grows exponentially with model release date.
Can I use METR's chart?
Use the format, not the picture. The chart, logo and design belong to METR. ViralChart redraws the chart type with your numbers and words, and credits the original publisher.