Sabine Hossenfelder
The video discovers that both AI doomers and AI evangelists can have patrons, then treats the funding trail as the main plot twist. Useful media hygiene, certainly—but “check who paid” is not an analysis of whether AI is deleting labour while everyone audits the sponsorship label.
The speaker examines paid influence campaigns surrounding AI coverage. They describe being approached by an organisation warning that AI poses an existential threat and being given a detailed script, then discuss grants and sponsorships associated with the Center for AI Safety, Control AI, and the Future of Life Institute. They contrast these efforts with sponsorships and creator campaigns backed by major AI companies, including an offer from Build American AI to promote China’s AI progress as a threat to American jobs. The speaker argues that paying creators is not inherently wrong, but that undisclosed or ambiguous sponsorships can blur the line between lobbying, marketing, and independent reporting. Their conclusion is that viewers should investigate who funded anti-AI or pro-AI content. The video itself includes a disclosed sponsorship for Ground News, presented as a tool for comparing coverage across political and factuality perspectives.
The video neither clearly accepts nor rejects the Discontinuity Thesis. It identifies a genuine information ecosystem problem—organisations may shape public narratives through paid creators—but it does not examine whether AI reduces aggregate demand for human labour, how gains accrue to capital owners, or whether displaced workers can be reabsorbed into the employment-consumption circuit. Its central move is to substitute provenance analysis for substantive analysis. Funding can reveal incentives, but it does not establish whether a claim about AI’s labour effects is true or false. By presenting anti-AI and pro-AI campaigns as broadly symmetrical, the video risks turning a material economic question into a dispute over compromised messengers. It offers no comforting economic solution, so this is not solution cope or terminal denial; it is moderate, sophisticated omission and motive-based deflection.
The video is right that propaganda has a marketing budget. It is also using that fact as a very tidy escape hatch: if both camps are sponsored, the viewer can feel sceptical and sophisticated without deciding whether AI is making human labour economically redundant. “Who paid?” is a media-literacy question, not a theory of labour demand. The beneficiaries of this framing are not just influencers. Capital owners are perfectly happy for the conversation to become a morality play about disclosure while ownership, rents, wages, and aggregate demand vanish from view. The speaker does not outright deny replacement, but refuses to adjudicate it and sells suspicion as analysis. The disclosed Ground News ad makes the loop almost comic: a video warning about influence ends by monetising the warning through another sponsor.