Designed for Outrage: How Algorithms Learned to Profit From Hate

Hate, Sped Up

Propaganda and misinformation are not new — the film cites Nazi Germany's boycotts of Jewish businesses and its dehumanizing depictions of Jews as a reminder of that. What has changed, this Nous Explainers video argues, is speed. Where hate once traveled by letter, newspaper or word of mouth, it now reaches millions with a single tap. Drawing on geographer David Harvey's idea of "time-space compression" — his argument that capitalism has steadily collapsed the distance and time separating people — the video suggests that when everything happens in real time, the sense of truth itself starts to blur. In that blur, it argues, hate becomes infinitely scalable, and in India it has found a rising target: the country's Muslim minority.

The Business Behind the Feed

To explain why, the video turns to how social media platforms actually make money. "If you're not paying for the product, then you are the product," it notes, describing an advertising-based business model in which every click, hesitation and pause is recorded and fed back into a loop built to keep users watching. Mark Zuckerberg is heard recalling that he built the first version of Facebook simply as a directory for Harvard friends to connect — a modest origin for a company now worth billions on advertising alone. The video traces the underlying logic to economist Herbert Simon's 1971 warning that "a wealth of information creates a poverty of attention": once attention becomes scarce, whoever captures it owns the market.

Why Outrage Wins

The video argues that this business model has a specific consequence: polarizing content has, in its words, a "home field advantage." Platforms reward conspiracy theories and outrage because the algorithm understands performance, not politics — and polarization performs best. It cites a Pew Research Center study of 10,000 American adults finding personal and political polarization at a 20-year high, and a 2018 MIT study of 126,000 stories shared by three million users, which found false news is 70% more likely to spread than true news and reaches people six times faster. The video also references an internal Meta experiment that classified content as "good for the world" or "bad for the world" and found users spent more time on the latter — a finding the company, by its own account, chose not to act on. The video frames that choice, paired with the brain's evolved negativity bias, as the mechanism turning ordinary engagement into an outrage machine: attention generates data, data trains the algorithm, and the algorithm optimizes for whatever provokes the strongest reaction.

From Attention to Consent

The video situates this inside a longer history of opinion-shaping, invoking Edward Bernays's 1920s idea of "engineering consent" and Edward Herman and Noam Chomsky's later concept of "manufacturing consent" — the argument that media doesn't simply inform but filters information to serve power and profit. What has changed, it argues, is that propaganda no longer needs a state broadcaster or a mass newspaper: anyone with a smartphone and an algorithm that rewards outrage can now produce it, and an entire economy of creators, influencers and troll networks has grown up around that incentive.

Myanmar's Warning

The video's central case study is Myanmar, where it says Facebook let extremist speech spread unchecked, fueling the 2017 Rohingya genocide. Peer-to-peer networks portrayed Rohingya Muslims as violent outsiders while glorifying the military, fake military-run pages quietly spread incendiary rumors — including one from the monk Wirathu that directly incited violence — and researcher warnings about the platform's role went unheeded until Facebook admitted in 2018 that it had not done enough to prevent the platform's misuse. The video treats this as evidence of a broader pattern: platform design decides which voices are amplified, often in ways that favor dominant power. It cites 2021 data showing X amplified right-wing content over left-leaning sources, and a 2023 Human Rights Watch finding that Meta suppressed pro-Palestinian content while allowing anti-Palestinian rhetoric to circulate — even as both platforms rolled back hate-speech and fact-checking rules by early 2025.

India's Digital Hate Factory

The video argues India has become a particularly acute case of platform capitalism reinforcing Hindutva nationalism. It cites a 2018 Observer Research Foundation study recording a 19-30% rise in digital hate incidents targeting Muslims over cow protection, beef consumption and interfaith marriage, and a 2020 Wall Street Journal investigation reporting that Facebook's top India public-policy executive, Ankhi Das, shielded a BJP legislator's calls for Muslims to be shot, reportedly citing business concerns. An inquiry after the 2020 Delhi pogrom found Facebook complicit in letting hate content go unchecked. In 2024, the video says, India Hate Lab documented 1,165 offline hate events — a 74% jump from 2023 — with nearly 1,000 originating as social-media posts, roughly half of them on Facebook, and 98% still online. It notes the irony that X, while hosting such content, blocked the accounts of hate-crime trackers India Hate Lab and Hindutva Watch on Indian government orders — not for spreading hate, but for documenting it — even as figures like Yati Narsingh Anand, who has called for genocide at a Haridwar religious assembly, continue circulating freely on WhatsApp and YouTube.

Hate as Infrastructure

The video's closing argument is that none of this is accidental: it is what happens when political power, corporate profit and platform design align. Citing 2022 findings that India topped the world in online anti-Muslim hate posts, it concludes that hate has become infrastructure — "built into the code, rewarded by the market, and weaponized by politics" — with consequences that reach beyond the screen to reshape democracy itself. The problem, it argues, is not the algorithm as such but who it is designed for, since digital platforms tend to replicate and amplify the inequalities already present in the societies that build them.