Abstract: As an emerging technology paradigm in the era of intelligent media, algorithmic recommendation has reshaped the practice of mainstream ideology with its unique operational mechanisms. At present, the algorithmic recommendation is widely embedded in the circulation field of information production, distribution, exchange and consumption, and it unfolds the logic of “covering” the mainstream ideology by hiding ideological attributes, being saturated with traffic capital logic, subverting interpersonal connectivity mechanisms, and altering the patterns of information acquisition. This results in multi-dimensional forms of “covering” characterized by spatiality, subjectivity, narrative, and valuation. To address these challenges, efforts should be made towards technical “goodness”, narrative “upgrading”, subject “boundary-breaking” and value “guarding”, and a diversified co-construction approach should be adopted to enhance the transmission, empathy, cohesion, and leadership of mainstream ideology, so as to guarantee its security.

Key words: algorithmic recommendation, mainstream ideology, covering, unveiling