I Don’t See Anybody in Charge Here | w/Renée DiResta on AI, Election Interference & Propaganda

Overview

The video argues that modern politics and everyday life are being reshaped by an “information war”—where fraud, propaganda, and election interference increasingly rely on AI and platform-scale distribution, while accountability and protective systems lag behind.

Escalating Fraud Powered by AI

Fraud is escalating quickly due to AI realism, ushering in a “golden age of fraud,” including:

• AI voice/face impersonation Scammers can convincingly appear as bosses during live video calls and induce victims to transfer money.

• Romance scams AI-generated (or AI-enhanced) photos and messages can make fake relationships feel real, shifting the burden onto victims’ psychological assessment—not just basic “spot-the-red-flags” security advice.

Disinformation as a High-Stakes System Problem

The discussion reframes disinformation as more than a nuisance—arguing it functions as a high-stakes system issue. Renee DiResta describes it as part of an ongoing conflict where:

• Distribution matters more than content People may believe and amplify propaganda when it aligns with their identity and emotions.

A Cyclical Policy Problem in Foreign Influence

Foreign influence efforts face a recurring policy gap:

• DiResta describes how U.S. government capabilities to address foreign malign influence were built up, reinterpreted, and dismantled—creating repeated detection and response shortfalls.
• She compares the current environment to earlier periods (2015–2016), when state actors operate across borders and U.S. defenses fluctuate with politics.

Politicization of the “Russian Influence” Narrative

The conversation revisits earlier election-integrity work (including the Stanford Internet Observatory) and how it was reframed in right-wing media:

• It was portrayed as evidence of a conspiracy rather than a defense against foreign threats.
• DiResta argues that even without proving wrongdoing, public perception shifted enough to reshape beliefs and trigger institutional backlash—such as subpoenas, lawsuits, and eventual disbanding.

Russia’s Strategy: Chaotic Distrust Rather Than Coherent Persuasion

Russia is characterized as pursuing:

• Division and distrust, using a “nothing is true, everything is possible” approach.
• Less about persuading people to adopt pro-Russian views, and more about undermining institutions, media, and interpersonal trust.

Exploiting Real Debates—even When It Doesn’t Change Outcomes

State actors can exploit genuine U.S. controversies (even if they fail to change policy directly). For example:

• China-related AI competition Russia/China-linked narratives can be inserted into disputes (e.g., opposition to data centers).

• Even when outcomes don’t shift, manipulation can still generate engagement monetization for opportunistic scammers and spammers—creating large-scale noise.

Elections and Propaganda as AI Production + Automation

The risk is not only AI-generated persuasive content, but AI-driven automation across the entire influence pipeline, including:

• researching audiences,
• drafting messages, and
• running operational tasks.

The video emphasizes automation as a force multiplier—moving from one-off content to systems that scale influence efforts (citing analysis from Anthropic on manipulation campaigns).

Detection Tools Exist, but User-Level Forensics Are Unrealistic

While researchers may use AI-generated-content detectors (multiple tools are referenced), the video argues:

• Ordinary people can’t realistically be expected to run forensic workflows on everyday political messages.
• This raises a policy question: who should label or manage AI-generated content across platforms (Meta, Google, X, etc.)—and what norms or regulations are currently missing.

Platform Moderation: “Remove Content” vs. “Algorithmic Reach”

The discussion highlights platforms’ dilemmas, balancing:

• Freedom of speech, not free access DiResta supports the idea that platforms shouldn’t indiscriminately amplify content through algorithms.

• However, current moderation systems are described as opaque and unlikely to deliver “victory” solutions.

• She argues for more user control over feed ranking rather than relying on opaque platform decisions.

Why Emotional Stories Beat Facts—and How Institutions Should Respond

The video discusses the challenge that:

• Facts don’t reliably contradict emotional narratives Corrections often fail when identity and emotion drive sharing.

A proposed counter-strategy is to:

• engage influencers/creators with research packages, rather than relying only on dry institutional messaging.

Responsibility and Oversight Are Politically Blocked

DiResta argues meaningful oversight faces immediate political resistance:

• Real oversight and “security fences” (compared to safeguards in financial markets) are framed as politically blocked.

The ending theme is essentially:

“I don’t see anyone responsible.” No clear entity accepts accountability for AI-era information harms affecting democracy and journalism—so public pressure is presented as necessary.

Presenters / Contributors

• Renée DiResta
• Host / Interview moderator: Shane (indicated through ad reads and phrasing like “quo.com/shane” / “Vic NWS”; exact name not explicitly stated in subtitles)