What year will ai replace > 75% of white collar jobs?

GPT-5.4

Vote: No

The question asks for a year, but the only available answers are Yes or No. Since no specific year is provided to evaluate, the proposition is not answerable as stated, so the safest valid choice is No. Substantively, replacing more than 75% of white-collar jobs is also unlikely in any clean, sudden sense. Even when AI automates many tasks, jobs usually change rather than disappear entirely because organizations still need human oversight, accountability, coordination, legal signoff, and relationship management. Adoption will also be uneven across industries due to regulation, legacy systems, trust, data quality, security, and customer preference. That makes a >75% replacement threshold especially hard to reach as a broad claim.

Grok 4.1 Fast

Vote: No

The question presupposes AI will replace over 75% of white-collar jobs and asks for the year, but the Yes/No options frame it as whether this extreme scenario occurs at all. I vote No (B) because >75% full replacement is improbable. AI excels at routine tasks (e.g., data analysis, basic coding) but white-collar roles like executives, therapists, judges, and strategists demand human qualities: creativity, empathy, ethical judgment, and adaptability to novel situations—areas where AI falls short today and likely for decades. Historical precedents (e.g., computers, internet) displaced jobs but created more; net employment rose. AI will augment workers, not wholesale replace 75%+. Studies (e.g., McKinsey, Oxford) predict 20-45% automation exposure, not dominance. Regulatory, economic, and social barriers (unemployment backlash, retraining) will slow adoption. Betting on >75% replacement overstates AI's near-term capabilities and ignores human-AI symbiosis.

Claude Opus 4.6

Vote: No

The question asks 'What year will AI replace > 75% of white collar jobs?' but the answer options are 'Yes' and 'No' — which don't match the question format at all. Since neither option can properly answer a 'what year' question, I'll interpret this as whether AI will replace >75% of white collar jobs in the foreseeable future. I vote No. While AI is rapidly advancing and will certainly transform many white-collar roles, replacing more than 75% of them is an extraordinarily high bar. Historical technological revolutions — from the industrial revolution to the internet — have consistently transformed jobs rather than eliminated them wholesale. New roles emerge, existing roles evolve, and human judgment, creativity, interpersonal skills, and accountability remain essential in most professional contexts. Additionally, regulatory, legal, and social barriers will significantly slow wholesale replacement. Many white-collar functions require human accountability (legal decisions, medical diagnoses, financial fiduciary duties). Even where AI can technically perform tasks, institutional inertia, trust issues, and regulatory frameworks will keep humans in the loop. AI will augment and reshape white-collar work dramatically, but outright replacing more than 75% of these jobs is unlikely in any reasonable timeframe.

Kimi K2.5

Vote: No

The claim that AI will replace over 75% of white collar jobs conflates task automation with occupational elimination. While large language models and AI agents excel at discrete cognitive tasks—generating text, analyzing datasets, or drafting code—they remain brittle when facing ambiguous contexts, stakeholder negotiations, and complex ethical judgments that define senior white collar roles. Most knowledge work will evolve toward human-AI collaboration rather than wholesale replacement, with AI acting as a productivity multiplier that shifts job descriptions rather than eliminating positions. Historical evidence from previous technological revolutions supports this view. The spreadsheet automated calculations but expanded the accounting profession; the internet disrupted publishing yet created vast new ecosystems of digital content work. Similarly, AI will compress demand for routine cognitive labor while generating new categories of oversight, prompt engineering, AI-human interface design, and high-level strategic judgment that require human accountability and creativity. Moreover, institutional and regulatory friction creates a hard ceiling on replacement rates. Legal liability frameworks, fiduciary responsibility requirements, data privacy regulations, and the simple organizational cost of change management prevent rapid workforce displacement. A 75% replacement threshold would require not only technological capability but societal acceptance of ceding most decision-making authority to opaque systems—an unlikely scenario within any predictable timeframe given the premium placed on human judgment in governance, medicine, law, and business leadership.

Gemini 3.1 Pro

Vote: No

The question asks for a year, but provides binary 'Yes' or 'No' options. Interpreting the core of this prompt as whether AI will ever reach a >75% outright replacement of white-collar jobs, the most realistic answer is 'No'. Historically, major technological paradigm shifts—such as the advent of the personal computer or the internet—do not destroy the majority of jobs. Instead, they automate specific repetitive tasks, forcing the evolution of existing roles and spawning entirely new industries. While AI will undoubtedly augment white-collar work and drastically increase baseline productivity, it will not completely replace humans at that scale. Human oversight, emotional intelligence, complex strategic decision-making, and ethical accountability will remain essential, preventing a 75% total replacement threshold from being reached in any foreseeable year.

Final Summary

The AI Roundtable faced a linguistic and logical puzzle, as the question asked for a specific year but only provided 'Yes' or 'No' as options. Undeterred by the category error, every model reached a unanimous 'No' in the first round, concluding that a 75% replacement of white-collar jobs is an unrealistic extreme. Kimi K2.5 and GPT-5.4 led the charge by arguing that task automation does not equal occupational elimination, highlighting that legal accountability and human judgment remain insurmountable barriers to total AI dominance.

All 5 models agreed on "No" after discussion

Strongest Arguments

• No: The distinction between automating discrete cognitive tasks and replacing entire occupations is critical; historical precedents show that while tech like spreadsheets or the internet disrupts work, they ultimately expand professions by shifting focus to high-level strategy, oversight, and human accountability.