92
Corporate earnings, translated into cope.
A batch-published archive of official call transcripts, company filings, earnings releases and consultancy reports scored for AI displacement language. No public waiting room. No live token burn. Just the cope, already processed.
25tracked sources
79.8average cope score
5call transcripts
2026-06-08last batch run
Microsoft: earnings call transcript
Microsoft Fiscal Year 2026 Second Quarter Earnings Conference Call
Microsoft just stood on an investor call and calmly described the factory farm for white-collar labour.
Microsoft scores 94/100 on the FY26 Q2 call because the call is not about software adoption. It is a map of the white-collar replacement stack: compute, models, agents, governance, identity, audit, security and paid seats. The old software model sold tools to workers. The new model sells synthetic workers to management.
Watchlist
Public companies use official call transcripts where available, then the latest regular SEC filings or official earnings releases. Private consultancies are treated as report-scanner sources.
96
Microsoft
Copilot, Azure, OpenAI exposure, agentic computing language and workforce efficiency pressure.
94
Amazon
AWS agent infrastructure, warehouse automation, retail labour scale and capex-heavy AI buildout.
94
Meta
AI ad automation, infrastructure spending and the familiar efficiency narrative after prior workforce cuts.
94
NVIDIA
The upstream factory for AI displacement, with direct cost-per-token and agentic AI framing.
92
Palantir
Sells AI operating layers into institutions that want fewer people touching more decisions.
91
IBM
Consulting plus automation is the old labour-arbitrage model meeting agentic software.
91
Block
Public AI efficiency and job-cut signals make it useful for measuring fintech displacement language.
91
Salesforce
Agentforce is a direct white-collar workflow substitution narrative wrapped as CRM productivity.
90
Alphabet
Search is exposed to AI answer engines while Google Cloud and Gemini push the replacement frontier.
90
Adobe
Firefly and creative automation sit directly on top of paid creative labour.
89
ServiceNow
Enterprise workflow automation with agentic AI is practically a cope-signal generator.
88
Uber
Driverless mobility, delivery automation, pricing algorithms and platform labour exposure.
88
Workday
HR and finance automation means the software that manages jobs is also eating job tasks.
88
Tesla
Robotaxi, Optimus, manufacturing automation and humanoid labour-substitution claims.
76
BCG
Publishes unusually direct productivity experiments, then sells the result as reshaping rather than replacing work.
74
McKinsey Global Institute
Quantifies automation and productivity pressure, then routes the conclusion through reskilling and advisory transformation.
74
PwC
Tracks AI-exposed job ads and wages, then sells workforce strategy rather than naming the discontinuity plainly.
69
Shopify
Merchant tooling, support automation and AI-first internal operating expectations.
68
Deloitte
Uses human-centred transformation language while documenting task automation and enterprise AI adoption.
66
Goldman Sachs
Publishes labour-displacement research while selling advisory, trading and automation into the transition.
62
Oracle
AI infrastructure deals, database automation and enterprise back-office replacement pressure.
60
Netflix
Content production, recommendation, advertising and generative creative tooling make it a useful culture-industry signal.
59
Accenture
The consultancy cope archetype: quantify working-hour exposure, sell reinvention, dodge displacement.
52
Spotify
AI audio, creator tools, ad automation and headcount discipline around a global media platform.
39
Apple
Massive device installed base, services margin expansion, AI feature pressure and supply-chain automation risk.
Latest Reports
The archive is designed for batch refreshes around earnings season. Rerun the scanner, review the top rows, publish.
87
91
90
85
68
91
53
60
90
93
88
96
96
95
48
79
86
Highest Cope
These are not claims that the companies are wrong. They are high-signal sources where the labour question is buried hardest under productivity, AI infrastructure and transformation language.
96
Microsoft Fiscal Year 2026 Third Quarter Earnings Conference Call
Microsoft is not being eaten by AI. It is building the machinery that eats everyone else.
95
Meta Q1 2026 Earnings Call Transcript
Meta is not being eaten by AI. It is building the machinery that eats everyone else.
94
IBM 1Q 2026 Earnings Webcast Transcript
IBM is trying to become indispensable to the Sovereigns before the agent stack turns it into a feature.
94
Microsoft Fiscal Year 2026 Second Quarter Earnings Conference Call
Microsoft just stood on an investor call and calmly described the factory farm for white-collar labour.
96
NVIDIA 10-K (period ended 2026-01-25, filed 2026-02-25)
NVIDIA is not being eaten by AI. It is building the machinery that eats everyone else.
96
Amazon 10-K (period ended 2025-12-31, filed 2026-02-06)
Amazon is not being eaten by AI. It is building the machinery that eats everyone else.
96
Tesla 10-K (period ended 2025-12-31, filed 2026-01-29)
Tesla is not being eaten by AI. It is building the machinery that eats everyone else.
96
Meta 10-K (period ended 2025-12-31, filed 2026-01-29)
Meta is not being eaten by AI. It is building the machinery that eats everyone else.
93
Alphabet 10-K (period ended 2025-12-31, filed 2026-02-05)
Alphabet is not being eaten by AI. It is building the machinery that eats everyone else.
88
Salesforce Q1 FY26 Earnings Conference Call Transcript
Salesforce is trying to become indispensable to the Sovereigns before the agent stack turns it into a feature.
92
Salesforce 10-K (period ended 2026-01-31, filed 2026-03-02)
Salesforce is trying to become indispensable to the Sovereigns before the agent stack turns it into a feature.
92
Block 10-K (period ended 2025-12-31, filed 2026-02-26)
Block is trying to become indispensable to the Sovereigns before the agent stack turns it into a feature.
Method
Cheap first-pass scoring for batch work. Local models can be added later as a second-pass editor, but the site does not need live interaction.
Pull
Fetch official call transcripts first, then the last two regular public filings or official earnings releases for each tracked company.
Score
Measure AI, labour, capex, euphemism and displacement signals with source weighting so transcripts do not get buried under long filings.
Publish
Render static public pages, JSON, sitemap and llms.txt. Visitors see finished verdicts, not a spinner.