Movements Need the Critical Thinking That AI Destroys

Verdict: Jacobin opinion piece delivers a sharp philosophical warning on AI's risk to human subjectivity and leftist movements, but it weakens its argument through unverified expert attributions, selective citation of a preliminary study, and source stacking that implies undue consensus.

Key Techniques and Evidence

The article employs strong philosophical framing from Kant, Žižek, and Marx to argue AI outsources "judgment," potentially eroding critical thinking vital for politics. This is executed thoughtfully, linking "cognitive debt" to broader existential risks.

However, several issues stand out:

• Unverified attribution: Credits AI ethics expert Zinnya del Villar with specific LLM examples, like associating "nurse" with women and "scientist" with men, to illustrate bias reinforcement.

"As AI ethics expert Zinnya del Villar has shown..."

*Evidence*: No public statements from del Villar match these examples; her work covers AI ethics generally but lacks this nurse/scientist link (searches across her publications and interviews).

• Unverified paraphrase: Attributes to Derek Thompson the idea that chatbots affirm users ("could tell us that we’re always right"), unlike humans, to explain preference for "decaffeinated" AI.

*Evidence*: Thompson writes on AI productivity but no matching quote or phrasing found in his Atlantic pieces or interviews.

• Overstating preliminary research: Cites an MIT preprint on "cognitive debt" (reduced brain activity in chatbot users) as providing "initial support," treating the term as established.

"A recent MIT study that found significantly reduced brain activity among regular users of chatbots..."

*Evidence*: arXiv:2506.08872 is unreviewed, n=54 Boston adults aged 18-39; EEG showed temporary connectivity differences during tasks, partially reversing in a control swap—not permanent atrophy.

• Source asymmetry: Relies on AI-skeptic philosophers (Eisikovits, Vallor) and left critics; brief economic nod to productivity but no counterbalancing voices.

*Why notable*: Builds toward consensus on AI's threat to "emancipatory politics" without diverse input.

Verifiable Omissions and Impact

• Study limitations: Omits preprint status, small/local sample, and lack of long-term data. *Matters*: Readers infer robust science backing "gradual loss of cognitive capacities," when findings are tentative.
• EEG specifics: No mention that effects were task-specific and reversible. *Matters*: Supports claim of erosion without noting nuances from the study itself.
• Mixed evidence on cognition: Excludes studies showing AI can enable higher-order thinking via offloading (e.g., Frontiers in Psychology 2025 on "cognitive offloading paradox"). *Matters*: Claim of uniform "destruction" of critical thinking lacks balance on verifiable productivity-cognition links.

These gaps don't negate the thesis but amplify its alarm via incomplete science.

Author and Outlet Context

Florian Maiwald, a University of Bonn philosophy research associate with a PhD there, specializes in political theory (e.g., book *Regressive Illusions*). He contributes to Jacobin and similar outlets, applying critical theory to liberalism and capitalism. As an opinion piece in socialist-leaning Jacobin, the anti-capitalist AI frame aligns with its mission—transparent advocacy, not hidden agenda.

Coverage Variations

Other outlets handled the MIT study differently:

• Primary source (MIT) stays empirical, detailing EEG results without policy alarmism.
• Time.com notes limitations explicitly while emphasizing education risks.
• The Conversation adds nuance, critiquing hype around "brain rot."

Bottom Line

Strengths: Compelling synthesis of philosophy and tech critique; rightly flags real risks like over-reliance on AI for judgment. Weaknesses: Unverified claims and study overreach erode credibility, turning skepticism into selective advocacy. Solid for leftist readers seeking ideological ammo, but demands scrutiny for broader audiences.

Further Reading

• MIT Media Lab: Your Brain on ChatGPT – Researchers' neutral empiricism on findings.
• Time.com: ChatGPT May Be Eroding Critical Thinking Skills – Alarmist take with methods details and limitations noted.
• The Conversation: MIT researchers say using ChatGPT can rot your brain. The truth is a little more complicated – Nuanced skepticism tempering hype.
• Undark.org: Are We Offloading Critical Thinking to Chatbots? – Balanced inquiry citing related studies.

(Word count: 612)