Five signals, monthly updates
AI Inflection Point
Signal Tracker
I am trying to track what the real measurable effects of AI are being reported. This page summarises five signals where primary data is being published. Every number is traced to its source. Where the data is confirmed, I say so. Where it is imprecise, I say that too. The five signals are detailed below.
Disclaimer: This is personal research, not professional advice. I am a technologist, not an analyst, economist, or forecaster. Nothing here constitutes financial, investment, career, or legal advice. The data comes from third-party sources I believe to be reliable, but I make no representations or warranties as to its accuracy or completeness. My views may be wrong. Past trends do not guarantee future outcomes. If you make decisions based on this, that is your call, not mine. Consult a qualified professional before acting on any of it.
Current state
Where the five signals stand
Two signals are at Critical severity, one at High, and two at Elevated. The data is grounded in primary sources where it is grounded, and flagged where it is not. Every card below links to the full signal page with source verification, charts, and watch items.
Signal 01
Labour Market
Critical
Entry-level graduate roles are down 20% to 45% depending on metric, with a unique junior-senior divergence in tech. The 28% headline figure has unknown methodology and cannot be directly compared to the verified range.
- -36% US tech job postings vs pre-pandemic (Confirmed)
- 148K Tech redundancies, Jan-May 2026 (Traceable)
- -28% Entry-level roles vs 2022 peak (Cannot compare)
Read full signal →
Signal 02
Revenue Concentration
Elevated
Anthropic's $30B ARR is roughly 75% business/API and consumption-based, with a single product (Claude Code) estimated at about 8%. Consumption-based revenue can fall as fast as it rose if usage shifts.
- $30B Anthropic ARR, April 2026 (Confirmed)
- 75% Business/API revenue share (Estimated)
- 500+ Enterprise customers over $1M/year (Confirmed)
Read full signal →
Signal 03
Compute Bottleneck
High
The physical constraint on AI growth has shifted from chips to electricity. The US grid interconnection queue holds 2,600 GW of pending requests. GPUs are "sitting in inventory" because there is not enough power.
- $349B Big tech AI capex, 2025 (Confirmed)
- 2,600 GW Grid interconnection queue (Confirmed)
- ELECTRICITY Current primary constraint (Confirmed)
Read full signal →
Signal 04
Regulatory Response
Elevated
The EU AI Act comes into force August 2026. The US has no federal AI law. 20+ states have created a patchwork. Regulation is the wildcard: it is low now but could escalate rapidly.
- Aug 2026 EU AI Act high-risk deadline
- None US federal AI law
- 20+ US state AI laws enacted/proposed
Read full signal →
Signal 05
Sector Disruption
Critical
Knowledge work is the canary. 65% of professional workers expect role redefinition. AI-proficient workers earn 15-20% premiums. The second-order cascade risk (client budget contraction) could amplify the first-order effects.
- 65% Workers expecting role redefinition
- -22% Projected decline in traditional graduate hiring
- $175K Salary premium for AI-proficient workers
Read full signal →
Three of five signals are at Critical or High severity. The junior hiring decline and sector disruption are already visible in the data. The compute bottleneck is a physical constraint that will tighten further. Revenue concentration is elevated but not yet dangerous. Regulation is the wildcard: it is low now but could spike rapidly if labour displacement becomes politically salient.
Trend tracking
How the signals are moving
I update the tracker every month. Each update adds a new data point. Historical data is never overwritten. After three or four updates, the trends will become the most important part of the dashboard. The direction matters more than the absolute level.
Data points:
2 Jun 2026 Baseline established
Severity score by signal
| Signal | Current severity | Score (0-100) | Direction since baseline |
|---|
| 01 Labour Market | Critical | 72 | Baseline |
| 02 Revenue Concentration | Elevated | 55 | Baseline |
| 03 Compute Bottleneck | High | 65 | Baseline |
| 04 Regulatory Response | Elevated | 45 | Baseline |
| 05 Sector Disruption | Critical | 70 | Baseline |
Methodology
How the tracker works
I am not an analyst or forecaster. I trace every number to its primary source. Where a number is confirmed, I say so. Where it is directionally accurate but imprecisely sourced, I say that too. Where it cannot be verified, I say that as well. Where credible sources disagree, I present both positions. This means my dashboard will sometimes show fewer data points than a typical industry report. I think that is a feature, not a bug.
Confidence ratings
| Rating | Definition |
|---|
| VERY HIGH | Published government dataset or peer-reviewed study with large sample |
| HIGH | Primary dataset with transparent methodology |
| MEDIUM-HIGH | Primary aggregator with broad scope |
| MEDIUM | Secondary source or different methodology |
| MODERATE | Contextual only, not directly comparable |
| LOW | Qualitative or untraceable |
What I do not do
I do not draw conclusions the data does not support. I do not use causal language without regression evidence. I do not fabricate chart data points. I do not present unverifiable figures in the hero. I do not promise certainty where the evidence is mixed.
Limitations
Margins of error and confidence intervals are not included in the figures shown. Time-series charts use point values without error bands. Where data points cannot be traced to a single publication, I note this explicitly. The report's purpose is to present what the data shows, not to settle the question of whether AI is reshaping the world.