Knowledge work is where AI adoption is being reported first. I am tracking the published data: worker survey expectations, employer hiring projections, salary premia for AI-proficient workers, overall posting trends. The market is splitting in two.
65%
Knowledge workers expecting role redefinition
Traceable, broad scope
-22%
Projected decline in traditional graduate hiring
Directional
$175K
Salary premium for AI-proficient senior workers
Directional
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.
About this report
Why I built it, and why it will get better
I am trying to track what the real measurable effects of AI are being reported. I trace every number to its primary source. Where the data 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. This tracker is the result.
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.
I will update the tracker every month. Each update will add a new data point, so over time you can see whether a signal is accelerating, stabilising, or receding. Historical data will never be overwritten. After three or four updates, the trends will become the most important part of the dashboard. I am tracking five areas (labour market, revenue concentration, compute, regulation, sector disruption), each broken into multiple metrics. This page covers sector disruption, which has four metrics. The other four signals will get the same treatment in subsequent updates.
Methodology
Source selection: I search for primary datasets from international organisations (World Economic Forum), industry analysts (Pragmatic Engineer, FRED), and independent researchers. I exclude any source where the underlying dataset cannot be located or where the methodology is undisclosed.
Confidence ratings: 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.
Limitations: Survey-based figures (such as the 65% role redefinition expectation) reflect worker perception, not confirmed labour market outcomes. Salary premium data is drawn from industry reports and job postings, not from a controlled statistical study. The $175K figure represents a midpoint of a range, not a precisely measured average. Time-series charts use point values without error bands. Where data points cannot be traced to a single publication, I note this explicitly.
Date of access: All sources were last accessed in June 2026. Source URLs were verified live during research. Where a source may have been updated since, the most recent version at time of writing is cited.
Built with Odokai. The initial research synthesis ran in a single session, with each source then individually verified. The brand styling came from a reusable Skill. The data, charts, and HTML live in a persistent cloud workspace, ready for next month's update with no setup. The platform makes the process repeatable: each cycle is a matter of re-running the Skill against fresh data, not starting from scratch.
Metric 1 of 4
Knowledge workers expecting role redefinition
The original claim: 65% of knowledge workers expect their role to be redefined this year due to AI. This traces to the World Economic Forum's research on AI and the future of work, corroborated by industry surveys. The figure reflects worker perception, which is a leading indicator but not the same as confirmed labour market outcomes.
Verdict: Traceable, broad scope
The 65% figure is traceable to the World Economic Forum and consistent with multiple industry surveys. However, it measures perception, not confirmed outcomes.
The World Economic Forum's research reports that 65% of professional workers expect their role to be redefined by AI within the year. This is a perception metric: workers believe their jobs will change, but belief is not the same as confirmed displacement or restructure. That said, perception drives behaviour. If 65% of knowledge workers expect change, they will adapt, reskill, or leave, and those actions reshape the labour market regardless of whether the original prediction was precisely right. The figure is directionally consistent with the other metrics in this signal: the market is splitting between roles that require AI fluency and roles centred on routine cognitive processing.
Source
Figure
Scope
Confidence
Primary?
World Economic Forum
65%
Knowledge workers expecting role redefinition, 2026
MEDIUM-HIGH
YES
Pragmatic Engineer
Consistent
Software developer sentiment aligns with WEF finding
MEDIUM
YES
Knowledge workers expecting role redefinition
The 65% figure is a single snapshot from WEF research, not a continuous time series. It measures worker expectation, which is a leading indicator but not a confirmed outcome.
Expecting
65%
35%
Expecting redefinition Not expecting redefinition
Perception is a leading indicator, but it is not the same as confirmation. If 65% of knowledge workers expect their role to change, the behavioural effects (reskilling, job-switching, anxiety-driven productivity changes) are real regardless of whether the restructure materialises exactly as predicted. The figure tells you where sentiment is heading. The other three metrics in this signal tell you whether the data is following.
Structural shift or survey noise?
Structural: perception reflects reality
The 65% figure is consistent with the measurable decline in traditional graduate hiring (-22%), the rise in AI-augmented roles, and the salary premium for AI-fluent workers. When perception aligns with hard data across multiple metrics, it is more likely to reflect a genuine structural shift than survey noise. Workers are responding to real changes in their daily work, not hypothetical fears.
World Economic Forum, Pragmatic Engineer
Survey-driven overstatement
Survey figures in periods of high media attention tend to overstate the pace and scale of change. Workers may report expecting redefinition because AI is discussed constantly, not because their actual role is changing. The gap between expectation and outcome is historically wide in technology adoption cycles. The 65% figure could be directionally right but numerically inflated by recency bias.
Historical technology adoption research
My assessment: The 65% figure is directionally reliable but imprecise. I cannot verify the exact percentage independently, and survey-based metrics in high-attention periods carry an inherent upward bias. What I can say is that the direction aligns with the other three metrics in this signal. The perception is consistent with the data. Whether the magnitude is exactly 65% matters less than whether the trend is accelerating or stabilising. I will be watching for changes in this figure over the next two to three quarters.
Metric 2 of 4
Projected decline in traditional graduate hiring
The original claim: traditional graduate hiring is projected to decline by 22%, expressed as a midpoint. This traces to multiple industry sources including the World Economic Forum and Pragmatic Engineer. The decline is not uniform: it is concentrated in roles centred on routine cognitive processing, while AI-adjacent graduate roles are growing.
Verdict: Directional, midpoint estimate
The -22% figure is a midpoint estimate drawn from multiple sources. The direction is well-supported, but the exact magnitude depends on how you define "traditional" and which sectors you measure.
The decline in traditional graduate hiring is not a single number from a single study. It is a composite midpoint reflecting the range of declines reported across technology, law, finance, and consulting. The World Economic Forum and Pragmatic Engineer both report significant declines in entry-level hiring for routine cognitive roles. The critical nuance: the decline is not uniform. AI-adjacent graduate roles (data science, AI engineering, cybersecurity) are growing. The market is splitting in two. The -22% midpoint captures the overall direction but obscures the divergence between declining and growing segments.
Source
Figure
Scope
Confidence
Primary?
World Economic Forum
-22%
Projected decline in traditional graduate hiring (midpoint)
MEDIUM-HIGH
YES
Pragmatic Engineer
Consistent
Software sector entry-level hiring decline aligns with midpoint
MEDIUM
YES
Traditional graduate hiring: the market is splitting
The -22% midpoint captures the overall direction, but the real story is the divergence between declining traditional roles and growing AI-adjacent roles.
Traditional
78%
-22%
AI-adjacent
Growing
Traditional graduate roles (declining) AI-adjacent graduate roles (growing)
The market is splitting in two. Roles that require AI fluency, data skills, and strategic judgement are in high demand. Roles centred on routine cognitive processing are flat or declining. The -22% midpoint obscures this divergence. The aggregate number tells you the direction. The split tells you the structure.
Metric 3 of 4
Salary premium for AI-proficient senior workers
The original claim: senior professionals who are AI-proficient command a salary of $175,000 and above, representing a significant premium over non-AI-proficient peers. This traces to industry job posting data and the World Economic Forum's research. The $175K figure is a midpoint of a $150K to $200K+ range.
Verdict: Directional, midpoint estimate
The $175K figure is a midpoint of a range, not a precisely measured average. The adaptation premium is real, but its exact magnitude varies by sector, geography, and role seniority.
Industry data consistently shows that workers who use AI tools earn a significant premium over those who do not. The World Economic Forum reports that senior professionals who are AI-proficient are seeing salaries of $150,000 to $200,000 and above. The $175K midpoint captures the central tendency of this range. However, salary data at this level is inherently noisy: it varies by sector (tech pays more than law), geography (US coastal cities pay more than elsewhere), and how you define "AI-proficient" (using ChatGPT is not the same as building with it). The premium is directionally clear. The exact number is imprecise.
Source
Figure
Scope
Confidence
Primary?
World Economic Forum
$150K to $200K+
Salary range for AI-proficient senior workers
MEDIUM-HIGH
YES
AI Magicx
$175K
Midpoint estimate for AI-adapted salary
MEDIUM
NO
AI-proficient senior worker salary range
The $175K figure is the midpoint of a sourced $150K to $200K+ range (WEF). The non-AI-proficient comparison band is illustrative only, included to show the shape of the gap; no single source measures a like-for-like baseline. Treat the premium as directional, not a precise figure.
The adaptation premium is real, though its size is imprecise. Workers who adapt appear to earn a meaningful premium. The sourced figure is the $150K–$200K+ range for AI-proficient seniors; the implied premium over non-AI-proficient peers (shown as ~$60K* in the chart) is illustrative, not measured, because no cited source isolates a clean like-for-like comparison. A two-tier market is forming, and the direction is clear even if the exact gap is not. This is the strongest individual-level incentive in the data. * Illustrative baseline, not a sourced figure.
Is the premium durable?
Structural: skills premium persists
Historically, workers who adopt transformative technologies early command a persistent premium. Spreadsheet-literate analysts in the 1980s, SQL-proficient analysts in the 1990s, cloud-native engineers in the 2010s: each cohort earned more for years, not months. AI proficiency is likely to follow the same pattern. The premium may compress as AI literacy becomes table stakes, but that transition takes years, not quarters.
Historical technology adoption patterns
Compressing as AI becomes universal
If AI proficiency becomes a baseline expectation rather than a differentiator, the premium compresses. This is what happened with basic computer literacy: it went from a premium skill to a minimum requirement within a decade. The current premium may reflect scarcity, not permanent value. As AI tools become easier to use and more widely adopted, the advantage of being "AI-proficient" narrows.
Labour market compression research
My assessment: The premium is real and likely to persist for two to three years, but it will compress as AI literacy becomes standard. The question is not whether the premium exists now (it clearly does), but how fast it narrows. If AI tooling becomes radically easier to use within 18 months, the premium compresses faster. If it remains a genuine skill differentiation, the premium holds. I will be watching freelance and contract rates for AI-augmented versus traditional work: that is the earliest indicator of wage pressure direction.
Metric 4 of 4
Overall knowledge-work postings trend since mid-2025
The original claim: overall knowledge-work job postings have trended up 15% since mid-2025. This traces to FRED analysis and corroborating industry data. The increase is real, but it is driven almost entirely by AI-adjacent roles, not by a recovery in traditional cognitive positions.
Verdict: Directional, imprecisely sourced
The +15% figure is directionally consistent with multiple sources, but the aggregate masks the critical divergence: AI-adjacent postings are growing while traditional knowledge-work postings are flat or declining.
Overall knowledge-work postings are up 15% since mid-2025, according to FRED analysis and industry data. But this aggregate figure hides the structural split. AI-adjacent roles (machine learning, prompt engineering, AI product management) are growing rapidly. Traditional cognitive roles (basic legal research, routine financial analysis, management consultancy at junior levels) are flat or declining. The +15% is real, but it is not a recovery. It is a reshaping. The net figure is positive because the growing segments are large enough to offset the declining ones. That does not mean the declining segments are recovering.
Source
Figure
Scope
Confidence
Primary?
FRED analysis
+15%
Overall knowledge-work postings since mid-2025
MEDIUM-HIGH
YES
AI Magicx
Consistent
Knowledge-work disruption report corroborates direction
MEDIUM
NO
Knowledge-work postings trend since mid-2025
The +15% aggregate is driven by AI-adjacent roles. Traditional knowledge-work postings are flat or declining. The net figure is positive, but the composition has shifted.
The +15% is not a recovery. It is a reshaping. The aggregate figure is positive because AI-adjacent roles are growing fast enough to offset the decline in traditional positions. The net is positive. The composition has fundamentally changed. Anyone looking only at the headline number will miss the structural shift underneath.
The critical second-order risk
When your clients' customers are disrupted by AI, their budgets shrink, and they spend less on your services. This cascade effect turns a sector-specific problem into a broader economic one. A law firm that loses corporate clients to AI-driven contract review tools does not just lose those clients' legal work. It loses the consulting, accounting, and advisory relationships that orbited around those clients. The second-order effects of knowledge-work disruption are likely larger than the first-order effects, and they are much harder to measure.
What I am watching
1. Client budgets in sectors being disrupted. Are law firms, financial institutions, and consultancies cutting spending on external services? If so, the cascade is underway. 2. Corporate earnings calls. Language about "AI efficiency" is often a proxy for "we are doing more with fewer people." When that language intensifies, the disruption is accelerating. 3. Freelance and contract rates for traditional work versus AI-augmented work. This is the earliest indicator of wage pressure. If traditional freelance rates fall while AI-augmented rates rise, the two-tier market is deepening.
Published government dataset or peer-reviewed study with large sample
Not applicable to this signal
HIGH
Primary dataset with transparent methodology
Not directly applicable (figures are survey-based or estimated)
MEDIUM-HIGH
Primary aggregator with broad scope
WEF research figures, FRED analysis
MEDIUM
Secondary source or different methodology
Pragmatic Engineer, AI Magicx corroborating data
MODERATE
Contextual only, not directly comparable
Historical technology adoption comparisons
LOW
Qualitative or untraceable
Not used in this signal
Severity assessment
Signal 05 severity: Critical (score: 70/100). Three of the four metrics point to structural disruption already underway. The 65% role redefinition expectation, the -22% decline in traditional graduate hiring, and the salary premium for AI-proficient workers all point in the same direction. The +15% postings trend is positive in aggregate but masks a compositional shift that is itself a disruption signal. The second-order cascade risk (client budget contraction) adds severity beyond what the first-order data captures. If the cascade materialises over the next two to three quarters, the severity will rise further.
Lewis Barclay · Personal research, not professional advice · Updated June 2026
