Philosophy
What feels like freedom from endless choice is also a quiet loss of “encounter with difference”, as feeds seal us inside familiar worlds.
Platforms as Complete Systems
The efficiency is real. Just as varna-based occupational sorting eliminated the anxiety of choosing life paths, algorithmic curation eliminates the paralysis of infinite choice. We do not have to search through thousands of movies; Netflix surfaces what we probably enjoy. We do not have to scan millions of job postings; LinkedIn highlights opportunities matching our profile. We do not have to evaluate countless perspectives; our news feed presents stories aligned with our existing frameworks. This is genuinely useful. The cognitive load reduction is substantial. The satisfaction rate is high. The system works, which is exactly why it is so concerning.
It is concerning because it eliminates an ‘encounter with difference’. A person confined to scholarly pursuits rarely engaged with the realities of agricultural labour or mercantile complexity (as an example). Their worldview, shaped by prescribed boundaries, mistook partial perspective for complete truth. The farmer’s wisdom, the merchant’s insights, and the manual labourer’s knowledge all remain inaccessible across caste barriers. Our algorithmic filter bubble operates identically. It fundamentally limits our worldview. The algorithm cannot be said to be lying; it only accurately identified patterns. But in optimising for those patterns, it has systematically excluded encounters that might disrupt them.
The Karma of Data
The parallel extends to how both systems explain inequality. In religious frameworks, karma provided the explanation: current circumstance reflects accumulated past actions. If born into disadvantage, previous karmas explain this. If enjoying privilege, earned through past merit. The system isn’t unjust; it’s perfectly just across vast temporal scales. This logic neutralised questions about structural inequality. Why challenge caste hierarchy when it reflects cosmic justice? Why question gender restrictions when they manifest karmic truth? The framework transformed social arrangements into natural law, making contingent power structures appear as necessary cosmic order.
Algorithmic systems employ remarkably similar logic through the language of data and patterns. If certain demographic groups receive lower credit scores, this reflects their actual risk patterns. If job recommendation systems suggest lower-paying positions to women, this matches their demonstrated career trajectories. If content moderation flags certain communities more frequently, this corresponds to their actual violation rates. The algorithm isn’t biased; it’s merely recognising patterns that exist in the data. This data-karma logic performs the same mystification as religious frameworks. It makes structural inequality appear as technical neutrality. It transforms historical discrimination into predictive accuracy. It presents systemic disadvantage as an individual pattern.
Consider how this operates in criminal justice algorithms. Systems predicting recidivism risk show higher scores for Black defendants. The algorithm defenders argue: we are not imposing bias; we are recognising patterns in the data. But the data reflects centuries of discriminatory policing, prosecution, and sentencing. Past injustice becomes the algorithm’s training material, then gets projected into the future as “objective risk assessment.” The person flagged as high-risk receives more severe sentencing, more intensive supervision, and fewer opportunities for rehabilitation. This treatment increases actual recidivism risk; lack of employment, housing instability, and constant surveillance make successful reintegration nearly impossible. The algorithm’s prediction becomes self-fulfilling. The data-karma cycle completes: our algorithmic assignment shapes our outcomes, our outcomes generate data confirming the assignment; the pattern perpetuates.
Just as karma explained inequality while obscuring its structural production, data patterns explain algorithmic sorting while obscuring how past discrimination becomes future prediction. Just as challenging karma meant questioning cosmic justice itself, challenging algorithmic bias requires questioning our faith in data as neutral truth.