Malevolent Soft Power

In the AI Story

Kamarck's immediate subject was Russian interference in the 2016 American presidential election, which she characterized as the exercise of malevolent soft power. The operations did not attempt to convince Americans that Russian values were attractive. They attempted to inflame American divisions, undermine trust in American institutions, and produce polarization that served Russian strategic interests by weakening the democratic functioning that generates American soft power. The mechanism was indirect: damage the source of soft power rather than compete with it.

Kamarck warned that the 2016 operation was the first but most certainly not the last example of malevolent soft power being used to influence a campaign, and that all of that paled in comparison to what AI could do to democratic systems. The warning acquired new dimensions with the AI tools The Orange Pill documents. The aesthetics of the smooth — Byung-Chul Han's concept of the cultural tendency to eliminate friction from every human experience — has an information-security dimension that neither philosophy nor commercial analysis fully explores. Smoothness is not merely aesthetic preference or cultural pathology. It is a vulnerability, because AI has collapsed the cost of producing persuasive content to near zero while leaving the human capacity to evaluate content unchanged.

Before AI, producing polished, professional analysis required expertise, time, and institutional backing. A credible think-tank report required researchers, editors, peer reviewers, and the institutional reputation giving authority. This production friction served as a rough quality signal — not perfect, but creating a cost differential between credible and incredible content that gave consumers a heuristic for distinguishing the two. AI eliminates this cost differential. A state actor, non-state organization, or single individual can now produce content indistinguishable in surface quality from the output of the most credible institutions. The polished prose, structured argument, appropriate citations, measured tone signaling expertise — all can be generated in minutes at negligible cost.

The poisoned well problem is the structural consequence. A single bad actor's use of AI-generated disinformation does not merely damage the credibility of the bad actor's output. It damages the credibility of the entire information environment, including legitimate analysis, genuine expertise, and authentic cultural production. Citizens cannot selectively distrust the manufactured when they cannot distinguish it from the authentic. They develop generalized distrust — a corrosive skepticism treating all information as potentially manipulated, all expertise as potentially fabricated. This is precisely the condition authoritarian influence operations seek to create: not to convince foreign populations of their narratives but to undermine the concept of shared truth itself. AI accelerates this objective by orders of magnitude. The asymmetry disadvantages democracies: open information environments generate soft power but create attack surfaces closed societies do not present.

Origin

Kamarck introduced the concept in "Malevolent Soft Power, AI, and the Threat to Democracy," Brookings Institution, November 2018. The analysis emerged from the recognition that Russian election interference operations employed the mechanisms of soft power (cultural influence, information channels) while inverting its purpose (manipulation rather than attraction).

Key Ideas

Inverted channels. Malevolent soft power uses the same cultural and informational channels as legitimate soft power but delivers manipulation rather than attraction.

Division exploitation. The mechanism targets divisions in the victim society rather than promoting the attractiveness of the manipulator's values.

AI amplification. Artificial intelligence accelerates malevolent soft power by orders of magnitude, enabling production of persuasive content at scale previously impossible.

Poisoned well effect. Individual disinformation campaigns damage the entire information environment by undermining heuristics for distinguishing credible from manufactured content.

Democratic asymmetry. Open information environments generate soft power but create vulnerabilities closed societies do not present, producing systematic disadvantage for democracies.

Debates & Critiques

Critics argue that concerns about malevolent soft power risk producing overreactions — censorship regimes, surveillance expansions — that themselves damage democratic soft power more than the original manipulation. Kamarck's framework concedes this concern but insists that recognition of the threat must come before debate over proportionate response, and that underreaction poses greater long-term danger than overreaction.