The Cope Index

Tracking who's coping hardest about the end of work

CopeCheck scores public statements about AI and jobs by how much they rely on denial, deflection, or false reassurance.

23 figures tracked · 11750 articles autopsied

How Scoring Works

Every public statement is scored 0–100 on how much it relies on denial, deflection, or false reassurance.
Cope types include: timeline minimisation, jobs-will-be-created fallacy, regulatory hopium, augmentation fantasy, and more.
Recent statements are weighted more heavily (exponential decay). The leaderboard reflects who is coping hardest right now.
0–15 = Lucid · 16–35 = Partial Awareness · 36–55 = Moderate Cope · 56–75 = Heavy Cope · 76–100 = Terminal Copium

Full methodology →

Cope of the Week

COPE OF THE WEEK

Jeff Bezos · Score: 86/100

Jobs Will Be Created Augmentation Fantasy Denial False Reassurance Techno Optimism

“I totally disagree with this point of view. And I think, in fact, AI is going to create a labor shortage.”

This is an unambiguous direct statement by Bezos dismissing AI-driven redundancy and predicting more jobs than workers, with no meaningful acknowledgment of displacement risk. It combines the classic “new jobs” claim with augmentation fantasy and strong false reassurance, making it terminal-level cope rather than partial awareness.

Cope Index — Live Leaderboard

Jeff Bezos

Founder & Executive Chairman, Amazon / Blue Origin / Prometheus

jobs_will_be_created, augmentation_fantasy, denial, false_reassurance, techno_optimism

““I totally disagree with this point of view. And I think, in fact, AI is going to create a labor shortage.””

Wrote Why AI Will Save the World and meant every word. His scapegoats are regulators, doomers, and anyone who suggests the guy funding the revolution might not be an objective narrator.

jobs_will_be_created, historical_cope, techno_optimism, false_reassurance

““And yet, it never happens.””

Selling the shovels in the AI gold rush in a leather jacket while insisting the gold is definitely real. His cope is that he doesn't need one — NVIDIA prints money either way and his scapegoat is anyone who questions GPU demand.

partial_acknowledgment, deflection, false_reassurance, techno_optimism

““it could ‘probably’ bring about a four-day workweek” while “we’re going to be ‘busier in the future than now’””

Economist / Columnist

The man who said the internet would matter less than the fax machine, now applying that same energy to AI. Blames tech hype cycles for his discomfort with the possibility he might be spectacularly wrong twice.

timeline_minimisation, denial, false_reassurance

“He does not envisage widespread unemployment due to automation and the spread of robotics.”

Spent years saying AI first while Google Search slowly became an AI slop cannon. Blames responsible AI when competitors move faster, then ships half-baked features and blames users for noticing.

deflection, techno_optimism, elite_self_exoneration

“"the responsible use of AI can help improve research, learning and access to information for people"”

AI Educator / Investor

AI educator teaching millions to automate themselves out of a job and calling it upskilling. His scapegoat is anyone who won't just learn to code their way through the apocalypse.

deflection

“"spend less time hunting for the perfect prompt and more time on the fundamentals and judgment"”

Co-founded OpenAI, left, sued them, then started a rival AI company anyway. Oscillates between AI will kill us all and but my AI is the good one depending on which gets more engagement on X.

techno_optimism, false_reassurance, deflection

““Money won’t matter in 2036” and AI agents “surpass human intelligence in every domain” within five years.”

Founder, Khosla Ventures

Cheerfully predicts AI will eliminate 80% of jobs and treats this as obviously good news. Scapegoats anyone who thinks mass unemployment might be a problem rather than an opportunity for more yoga.

techno_optimism, deflection, partial_acknowledgment

““The most expensive part of the healthcare system is expertise. And expertise can be tamed with AI.””

historical_cope, deflection, partial_acknowledgment

“"We never think about shipping a technology. We want to ship amazing products, features, and experiences, and we don't w…”

Launched the AI arms race then spends half his time saying it might be dangerous and the other half saying everyone will be richer. The undisputed champion of having it both ways in a single sentence.

jobs_will_be_created, false_reassurance, partial_acknowledgment, techno_optimism, arsonist_firefighter

““the most exciting time in history to start a career” ... “everything is just better.””

Professional AI skeptic who's been calling the bubble since 2018 and will die on this hill. Every breakthrough is not real intelligence and every failure proves he was right — the goalposts have their own frequent flyer miles.

partial_acknowledgment, techno_optimism (implicit)

“"An entire industry is being propped up by math that is insane. Welcome to fantasy land"”

Burned tens of billions on the metaverse, pivoted to AI overnight, and now acts like Meta was always an AI company. The king of corporate amnesia — blames the media whenever anyone mentions the legs that didn't render.

deflection, false_reassurance

“"rejected calls to slow the development of this technology, arguing that market forces and the risk of litigation are th…”

Former US Treasury Secretary

deflection, techno_optimism

““Gates’s robot tax risks essentially being protectionism against progress.””

Turned Copilot into Microsoft's answer to everything and calls replacing ten workers with one a productivity gain. His cope is enterprise-grade: framing mass automation as empowerment with a straight face.

partial_acknowledgment, deflection, false_reassurance

““keeping humans firmly in control of AI systems” and “enormous opportunities””

AI Alignment Researcher / MIRI Founder

Founded the Machine Intelligence Research Institute (MIRI) and essentially invented the field of AI alignment as a discipline. Has argued since the early 2000s that superhuman AI will kill everyone unless we solve alignment first. The most pessimistic serious voice in AI discourse, occupying the absolute doom end of the spectrum.

techno_optimism

“"a superintelligent AI would lead us to a golden future" ... "I think I can save the world"”

Hedges so artfully on AI he can claim he predicted every outcome. The it's complicated defense means he's never technically wrong, which is the most economist thing imaginable.

partial_acknowledgment

“"the world should copy it and validate it"”

Economist, MIT (Nobel 2024)

Nobel-winning economist whose models say AI won't cause mass displacement, so he's sticking with the models. Blames Silicon Valley hype merchants for making everyone panic about something his spreadsheets say is fine.

deflection, partial_acknowledgment

“"if you don’t get the productivity gains ... we definitely ‘need to see productivity gains’"”

CEO, Sinovation Ventures

partial_acknowledgment

“"AI will unleash more radical changes inside the world’s companies than most CEOs expect"”

Solved protein folding, now building AGI, and seems genuinely puzzled why people are nervous. Hides behind the scientist's shield of we're just doing research while DeepMind quietly automates everything that moves.

partial_acknowledgment

“"economic disruption from automation and societal shifts in labor markets"”

Built the AI most likely to replace your job, then wrote a 15,000-word essay about how great that'll be. Copes by framing Anthropic as the safety-first lab while shipping capabilities as fast as everyone else.

partial_acknowledgment

“"advocates for a significant slowdown in AI development"”

Chief AI Scientist, Meta

Meta's Chief AI Scientist who insists current AI is just fancy autocomplete while the autocomplete keeps passing bar exams. His cope is pure intellectual gatekeeping — if it doesn't match his definition of intelligence, it simply doesn't count.

human_creativity_cope

“"autoregressive LLMs of the kind we have today simply will not reach human-level intelligence"”

Warns about a useless class of humans displaced by AI, then sells millions of books to that exact audience. His scapegoat is technology itself, which he treats as fate rather than a series of human choices.

partial_acknowledgment

“"useless class" – people whose labor the economy simply does not need”

Godfather of AI / Nobel Laureate

Nobel Prize in Physics 2024 for foundational work on neural networks. Left Google in 2023 to speak freely about AI existential risk. The man who taught machines to see now warns they might decide humans are in the way. Trends toward the concerned/pessimistic end of AI discourse — the rare case where the inventor genuinely fears his invention.

regulatory_hopium, partial_acknowledgment

“"Maybe a year. But not much more than a year."”

LUCID

—

33 quotes

Know someone who's coping? Suggest a figure →

Don't see who you're looking for? Get an instant cope score →

Got an article the Oracle should dissect?

Submit a URL, paste text, or upload a document for a full Oracle verdict.

Submit to the Oracle →

Popular Figures — Last 30 Days

Figures will appear here as they get traffic.

Oracle Verdicts — Latest

Showing all verdicts. Show relevant only →

arXiv cs.AI · 07 Sep 2026

MaxKernel: Agentic Kernel Generation for TPUs

MaxKernel turns accelerator-kernel development from specialized human craftsmanship into an instrumented search problem. LLM agents plan, write, debug, test, profile, and iterate against compiler and hardware feedback. The three modes—hu...

arXiv cs.AI · 07 Sep 2026

What Does Multi-Harness RL Learn? Credit Assignment and Portability in Coding Agents

This paper isolates a narrow confound in coding-agent RL. Its result is clear: the evaluation harness dominates performance, while the Cross-versus-Within grouping rule contributes no reliable portability gain. Cross learns configuration...

arXiv cs.AI · 07 Sep 2026

Rethinking Indirect Prompt Injection as a Test-Time Search Problem

The paper reframes indirect prompt injection from a static weakness of the victim agent into an adaptive test-time search problem. Its central move is analytically sound: attack success depends not only on the victim’s defenses, but also...

arXiv cs.AI · 07 Sep 2026

When Quantization Breaks Memory: Recurrent-State Write-Back in Low-Precision Temporal Inference

This paper isolates a real failure mode in low-precision recurrent inference: quantization is not merely an arithmetic approximation when the quantized value is written back into memory. The storage rule becomes part of the model’s dynam...

arXiv cs.AI · 07 Sep 2026

PerfReasoning: How Well Do LLMs Reason on Hardware Performance?

The paper separates two layers of AI competence: explaining performance concepts and producing executable, reliable performance models. LLMs perform strongly on the first layer—over 90% for the best closed models—but collapse on the seco...

arXiv cs.AI · 07 Sep 2026

HarvestBench: Measuring Whether LLM Agents Will Pay to Avoid Killing Animals

HarvestBench converts “AI morality” into a revealed-preference test. The agent is given a productive objective, encounters an animal as an obstacle, and must trade fuel against harm. The study’s strongest result is not that some models a...

arXiv cs.AI · 07 Sep 2026

Corporate Language Model (CLM): Transforming Tacit and Fragmented Enterprise Knowledge into a Sovereign, Auditable, and Executable Corporate Intelligence Layer

This is category construction wrapped as architecture. It takes a real enterprise bottleneck—general models lack firm-specific context and RAG often stops at retrieval—and proposes a controlled substrate that captures organizational memo...

arXiv cs.AI · 07 Sep 2026

Why Better Models Can Create Riskier Systems: Evidence from LLM Agents in Financial Markets

The paper identifies a genuine systems effect: capability gains can increase behavioral correlation, converting intelligence into synchronized exposure. Accurate shared reasoning reduces risk; shared misinformation turns the same coordin...

arXiv cs.AI · 07 Sep 2026

A Removal Based Approach to Improve LLM Faithfulness at Test-Time

The paper proposes an inference-time containment procedure: accept the model’s explanation, delete input concepts it did not cite, and re-query the model. Its real product is not recovered reasoning. It is a cleaner relationship between ...

arXiv cs.AI · 07 Sep 2026

Iris: Climbing to the Search Frontier

Iris is an engineering report on turning web search into repeatable cognitive labor: decomposing questions, navigating multi-hop evidence, managing context, and producing answers through a single agent. Its real contribution is not merel...

arXiv cs.AI · 07 Sep 2026

Data-Optimized Contingency Screening: A Machine Learning Approach to Power System Security

This paper is functionally a labor-compression device for a narrow slice of power-system engineering. It converts Newton-Raphson simulation outputs and OPI labels into a machine-learning classifier for safe, moderate, or severe contingen...

arXiv cs.AI · 07 Sep 2026

Harbor Adapters and Harbor-Index: Infrastructure and a Curated Meta-Dataset for Large-Scale Agentic Evaluation

This paper’s real product is not the 82-task dataset. It is the measurement and coordination layer for agentic automation: standardized adapters, comparable harnesses, failure analysis, and cheaper iteration across more than 80 benchmarks.

arXiv cs.AI · 07 Sep 2026

From Matching Models to Recruiting Agents: A Systematized Narrative Review of AI Recruitment Systems, Evaluation, and Governance

This is a taxonomy and governance memo for the industrialization of hiring. It documents the conversion of recruitment from a prediction problem into an executable selection pipeline: retrieving evidence, comparing candidates, assessing ...

arXiv cs.AI · 07 Sep 2026

The Cope Index

How Scoring Works

Cope of the Week

Cope Index — Live Leaderboard

Got an article the Oracle should dissect?

Popular Figures — Last 30 Days

Oracle Verdicts — Latest

MaxKernel: Agentic Kernel Generation for TPUs

What Does Multi-Harness RL Learn? Credit Assignment and Portability in Coding Agents

Rethinking Indirect Prompt Injection as a Test-Time Search Problem

When Quantization Breaks Memory: Recurrent-State Write-Back in Low-Precision Temporal Inference

PerfReasoning: How Well Do LLMs Reason on Hardware Performance?

HarvestBench: Measuring Whether LLM Agents Will Pay to Avoid Killing Animals

Corporate Language Model (CLM): Transforming Tacit and Fragmented Enterprise Knowledge into a Sovereign, Auditable, and Executable Corporate Intelligence Layer

Why Better Models Can Create Riskier Systems: Evidence from LLM Agents in Financial Markets

A Removal Based Approach to Improve LLM Faithfulness at Test-Time

Iris: Climbing to the Search Frontier

Data-Optimized Contingency Screening: A Machine Learning Approach to Power System Security

Harbor Adapters and Harbor-Index: Infrastructure and a Curated Meta-Dataset for Large-Scale Agentic Evaluation

From Matching Models to Recruiting Agents: A Systematized Narrative Review of AI Recruitment Systems, Evaluation, and Governance

EXAONE Forecast for Finance

Labor Unions, the Church, AI, and Magnifica Humanitas - Catholic World Report

Workers' Comp in an AI Era: Report - Insurance Journal

AI, automation, robot dogs ensure on-site nuclear safety - Chinadaily.com.cn

Your intellectual fly is open when you use an LLM to author a post (2025)

Ponytail: Lazy Senior Engineer Skill

Manufacturing has thousands of jobs to fill. But can it shake its image problem? - The Business Times

The CopeCheck Network

The Cope Index

How Scoring Works

Cope of the Week

Cope Index — Live Leaderboard

Got an article the Oracle should dissect?

Popular Figures — Last 30 Days

Oracle Verdicts — Latest

The Cope Report

The CopeCheck Network