Future Index

AI

The same four readings, over the topics this product tracks under ai. Every panel below is ranked within that selection rather than filtered out of a world-wide ranking.

World conditions

A fairly ordinary day

Attention37below usual−21.1 ptsdown over 24 hours

Where the Attention figure comes from, what it reads and what it does not claim. Method version 5.0.0.

How Attention is calculated

06:0009:0012:0015:0018:0021:0000:0003:00

10-2610-2710-2810-2910-3010-3111-01

10-0810-1510-2210-29

High confidence0.77
5 sourcesPublic discussion3Search & activity2
19 items
Mood+10more positive than usual+11.8 ptsup over 24 hours

Where the Mood figure comes from, what it reads and what it does not claim. Method version 5.0.0.

How Mood is calculated

Right now 31% positive · 53% neutral · 15% negative — the figure is measured against its usual mix

06:0009:0012:0015:0018:0021:0000:0003:00

10-2610-2710-2810-2910-3010-3111-01

10-0810-1510-2210-29

High confidence0.91
3 sourcesPublic discussion3
1,987 items
Turbulence56active+11.9 ptsup over 24 hours

Where the Turbulence figure comes from, what it reads and what it does not claim. Method version 5.0.0.

How Turbulence is calculated

06:0009:0012:0015:0018:0021:0000:0003:00

10-2610-2710-2810-2910-3010-3111-01

10-0810-1510-2210-29

High confidence0.92
8 sourcesForecast markets3Public discussion3Search & activity2
117 items
Disagreement40some disagreement−4.3 ptsdown over 24 hours

Where the Disagreement figure comes from, what it reads and what it does not claim. Method version 5.0.0.

How Disagreement is calculated

06:0009:0012:0015:0018:0021:0000:0003:00

10-2610-2710-2810-2910-3010-3111-01

10-0810-1510-2210-29

Medium confidence0.49
3 sourcesPublic discussion3
1,903 items

Forecast conditions

Forecast Movement54moving−15.1 ptsdown over 24 hours

Where the Forecast Movement figure comes from, what it reads and what it does not claim. Method version 5.0.0.

How Forecast Movement is calculated

06:0009:0012:0015:0018:0021:0000:0003:00

10-2610-2710-2810-2910-3010-3111-01

10-0810-1510-2210-29

High confidence0.89
3 sourcesForecast markets3
31 items

31 questions · 3 sources

Forecast Movement measures how much expectations moved, not whether they moved towards any particular outcome. Why there is no single score, why what we can see is reported separately.

Today in brief — the last 24 hours on one page

What changed today

The largest movements across forecasts, attention, discussion sentiment and reporting in the last 24 hours, each compared with how much that topic usually moves.

  1. AttentionPublic discussionHacker News

    AI and work

    Attention rose sharply

    5.9× its usual variation — far larger than usual

    +8 mentionsup over the last 24 hours

    High confidence0.97Up to date

  2. AttentionPublic discussionMastodon

    OpenAI

    Attention rose

    2.9× its usual variation — larger than usual

    28mentions

    +12 mentionsup over the last 24 hours

    Medium confidence0.57Up to date

153 changes were compared.

All changes today

Signals moving together

  1. EARLY ATTENTION

    Public discussion rose before the forecast contracts moved.

    Anthropic

    Forecast
    25.0 pts over 24 hours, across 1 contract, 2.2× its usual daily move
    Will Anthropic’s IPO Form S-1 be public by 31 October 2026? opens in a new tab from 57.2% to 82.2%
    Public attention
    6.3σ from usual, 5 sources
    Observed order
    2 hours apart

    Observed Oct 26, 15:00 UTC High confidence

    How this was detected

    Strength within early attention: 79%

    • Public discussionHacker News
    • Search & activityFuture Index internal search and pageviews
    • Public discussionLemmy
    • Forecast marketsManifold Markets
    • Public discussionMastodon
    • Search & activityWikimedia Pageviews (Analytics API)
    • One forecast contract moved.
    • Some of the underlying activity was discounted as concentrated.

  2. EARLY ATTENTION

    Public discussion rose before the forecast contracts moved.

    AI investment

    Forecast
    3.3 pts over 24 hours, across 1 contract, 2.6× its usual daily move
    Will there be an AI bubble before 2030? opens in a new tab from 68.6% to 71.9%
    Public attention
    5.4σ from usual, 4 sources
    Observed order
    23 hours apart

    Observed Oct 28, 08:01 UTC Medium confidence

    How this was detected

    Strength within early attention: 17%

    • Public discussionHacker News
    • Search & activityFuture Index internal search and pageviews
    • Public discussionLemmy
    • Forecast marketsManifold Markets
    • Public discussionMastodon
    • One forecast contract moved.
    • Some of the underlying activity was discounted as concentrated.

  3. EARLY ATTENTION

    Public discussion rose before the forecast contracts moved.

    OpenAI

    Forecast
    3.0 pts over 24 hours, across 1 contract, 2.3× its usual daily move
    OpenAI announces it has achieved AGI before 2027? opens in a new tab from 10.0% to 13.0%
    Public attention
    12.0σ from usual, 5 sources
    Observed order
    21 hours apart

    Observed Oct 26, 18:00 UTC High confidence

    How this was detected

    Strength within early attention: 15%

    • Public discussionHacker News
    • Search & activityFuture Index internal search and pageviews
    • Public discussionLemmy
    • Public discussionMastodon
    • Forecast marketsPolymarket
    • Search & activityWikimedia Pageviews (Analytics API)
    • One forecast contract moved.
    • Some of the underlying activity was discounted as concentrated.
1 further detection on this page — show it

  1. CONFIRMED SHIFT

    Forecast probability and public attention moved in the same window.

    Anthropic

    Forecast
    25.0 pts over 24 hours, across 1 contract, 2.2× its usual daily move
    Will Anthropic’s IPO Form S-1 be public by 31 October 2026? opens in a new tab from 57.2% to 82.2%
    Public attention
    8.9σ from usual, 5 sources

    Observed Oct 27, 12:00 UTC High confidence

    How this was detected

    Strength within confirmed shift: 100%

    • Public discussionHacker News
    • Search & activityFuture Index internal search and pageviews
    • Public discussionLemmy
    • Forecast marketsManifold Markets
    • Public discussionMastodon
    • Search & activityWikimedia Pageviews (Analytics API)
    • One forecast contract moved.
    • Some of the underlying activity was discounted as concentrated.

Biggest movers

  1. AI and work61
    about usual
    +8.0 ptsup over 24 hours
    • Medium confidence
    • 5 sources
    • AI
    • Economy
  2. Deepfakes and AI misinformation43
    below usual
    +3.8 ptsup over 24 hours
    • Low confidence
    • 5 sources
    • AI
  3. AI in healthcare45
    about usual
    +2.3 ptsup over 24 hours
    • Low confidence
    • 5 sources
    • AI
    • Health

All measurements and ranges

Trending topics

  1. OpenAI32
    2 kinds of source are unusually busy with this
    • Discussion
    • Read here
    • Reported
  2. AI and work31
    One kind of source is unusually busy with this
    • Discussion
  3. AI capability milestones26
    2 kinds of source are unusually busy with this
    • Discussion
    • Read here
    • Reported
2 further topics on this list — show them
  1. Open-weight models18
    One kind of source is unusually busy with this
    • Discussion
  2. Anthropic18
    One kind of source is unusually busy with this
    • Discussion
    • Looked up
    • Read here
    • Reported

Where sources disagree

No topic currently shows a material difference between how much is being published about it and how much it is being discussed. A comparison needs both sides: a topic only one of them has anything to say about is not a disagreement.

Data conditions

How well this site can currently see. None of this is a reading of the world, and none of it changes the figures above.

Measurement quality
80%
Coverage
73% — 8 of 11 sources reporting
Freshness
Up to date
Representativeness
Moderate
  • 29 online communities
  • 3 kinds of evidence
  • one language scored; 10 more seen and not scored; 719 items carried no identification
  • 1 of 6 regions, over 1 located subject

How these are worked out, and why they are kept separate from the readings