Could an AI Essay Writer Write a Viral Manifesto? We Tried It on Twitter

Manifestos that tap into strong emotions—particularly righteous anger, hope, or a sense of belonging—spread more readily. Content that makes people feel something deeply is more likely to be shared than purely intellectual arguments.

Successful manifestos often help people express or affirm their identity. When sharing becomes a way to signal group membership or values alignment, content spreads faster through social networks of like-minded individuals.

Viral manifestos typically employ simple, memorable language with short sentences and clear structure. They avoid jargon and academic complexity in favor of accessible language that can be quickly understood and readily quoted.

The most shareable manifestos present familiar ideas in seemingly novel ways. They offer the excitement of a fresh perspective while confirming existing beliefs, creating an optimal balance between cognitive comfort and stimulation.

On Twitter specifically, manifestos must balance brevity with substance. The most successful examples can be shared in a thread format (typically 5-15 tweets) that delivers enough depth to feel substantial without requiring excessive time investment from readers.

Manifestos that connect to current cultural moments or ongoing social conversations have natural distribution advantages. They feel relevant and urgent rather than abstract or theoretical.

Focused on internet privacy, digital rights, and technology autonomy. We instructed the AIs to create a compelling case for greater individual control over personal data and technology.

Addressing modern workplace culture, work-life integration, and productivity philosophy. We asked for provocative but constructive ideas about reshaping professional norms.

Calling for environmental awareness and climate responsibility. We specified that this should avoid political polarization while still conveying urgency and personal responsibility.

Addressing screen addiction, social media effects on mental health, and healthier technology habits. This aimed to be motivational rather than judgmental.

Challenging conventional education paradigms and advocating for lifelong, self-directed learning. This needed to balance critique with constructive alternatives.

The most successful AI manifestos articulated common frustrations that people experience but rarely see explicitly acknowledged in public discourse. This created powerful moments of recognition and validation.

Manifestos that combined criticism of existing systems with hopeful visions for alternatives performed better than those that were purely critical or purely utopian. This balance helped readers feel both validated in their discontent and inspired to imagine alternatives.

Successful AI manifestos avoided vague platitudes in favor of concrete examples and specific claims. This specificity made the manifestos feel more authentic and thoughtful, even when generated by AI.

The least successful manifestos relied heavily on generic language and familiar platitudes. This made them feel uninspired and algorithmic, failing to create the emotional connection necessary for virality.

AI systems often produced overly cautious content with excessive hedging language ("some may argue," "it could be said") that diminished the manifesto's rhetorical power. The most successful examples required significant prompting to reduce this tendency.

AI systems struggled most with highly polarized topics. They either produced blandly neutral content that appealed to no one or leaned into one perspective in ways that alienated significant portions of the potential audience.

The most viral human-written manifestos typically emerge from lived experience and genuine urgency. AI systems can simulate this quality but struggle to match the authentic emotional resonance that comes from real frustration, hope, or indignation.

Truly viral manifestos often tap into nascent cultural moments that are just beginning to crystallize. AI systems, trained on past data, are better at reflecting established cultural patterns than identifying emerging ones or creating genuinely novel cultural framings.

Manifestos often gain credibility from their author's personal experience or authority. A tech manifesto from a prominent developer or a workplace manifesto from someone who transformed their company carries inherent social proof that AI-generated content lacks.

While our experiment was limited in scope, it demonstrated that AI systems can produce moderately persuasive content at scale. This raises concerns about potential use in coordinated influence campaigns that could flood social media with consistent messaging disguised as organic content.

Combined with audience analysis tools, AI manifesto generators could potentially create highly targeted persuasive content optimized for specific demographic, psychographic, or ideological segments—a powerful tool for manipulation if misused.

The drop in engagement after revealing the AI origin of our manifestos demonstrates how attribution matters. Bad actors could leverage AI-generated content while falsely attributing it to fictional or real people to increase its perceived authenticity and impact.

While individual AI manifestos might be less powerful than the most viral human-created content, their ease of production could enable quantity-over-quality approaches that saturate social media environments and crowd out authentic human voices.