摘要
arXiv:2605.29367v1 Announce Type: new Abstract: When workers lose jobs to AI-driven restructuring, two very different conversations happen on X (formerly Twitter) at the same time. Tech executives and AI researchers talk about productivity, transformation, and opportunity. Laid-off workers and labour critics talk about job loss, uncertainty, and fear. This paper asks a simple question: which conversation gets more reach? We report three studies using two collection methods and 763 tweets from 20 named public accounts. Study 1 used keyword-based collection (n=392) and found no significant difference between corpora (p=0.891), revealing that keyword search is too noisy for this task. Study 2 used account-based collection (n=96) and found a 3.12x mean amplification advantage for capital discourse over labour discourse (p=0.000003, Cohen's d=0.555). Study 3 combined both methods (n=763) and confirmed the finding at 4.18x mean and 10.77x median amplification ratio (p<0.000001).