Social influence tactics date back centuries, but modern psychological operations (PsyOps) took shape during World War II and the Cold War. Governments employed propaganda leaflets, radio broadcasts and covert media campaigns to shape enemy morale and public opinion. By the 2000s, state actors and private firms alike shifted to digital channels, using targeted ads, fake personas and algorithmic amplification to sway elections and consumer behavior.

Talking points:

Definition of PsyOps: combining psychology and operations to influence perceptions

Key examples: Operation Mockingbird (1950s), Voice of America, Cold War leaflets

Transition to digital: automated bots, microtargeting and data mining

The first couple of levels for developing social trust are around some platform native things :

Positive : Follows, User Lists, Likes

Negative : Label, Moderaton List, Unfollow, Block

Most social media users lack training in influence tactics, so they rely on everyday cues—likes, follower counts, badges and platform recommendations—to judge trustworthiness. These native tools serve as proxies for authority and popularity but also introduce biases. Users fall prey to echo chambers, hype cycles and viral misinformation when they don’t see the hidden hand of coordinated influence operations.

Talking points:

Heuristics of trust: social proof, authority bias, bandwagon effect

Platform signals: verification badges, trending tags, “Top Fan” icons

Algorithmic filters: curated feeds, “recommended for you” lists

Imagine each AT Protocol account maintaining a dynamic trust score for every other account it encounters. Starting at zero, interactions and third-party endorsements gradually increase that score toward 1.0, while unfollows instantly halve it and blocks reset it to zero. This decentralized trust ledger—stored in each user’s DID record—powers more nuanced content ranking, spam filtration and community governance.

Talking points:

Trust score lifecycle: initiation at 0, incremental rises, fast penalties

Interaction weights: recency, mutual follows, labeler endorsements

Unfollow penalty: score ÷ 2; block penalty: score → 0

User-controlled thresholds for content filtering and discovery