AIMS project
AI Incidents in Marketing
AI Incident Database // Brand Communications & AI Governance

Duolingo’s AI-first announcement and the communications backlash

When an AI strategy statement contradicts a brand built on human warmth

Incident type Brand communications failure, AI labour narrative
Date April 2025
Risk Brand reputation, stakeholder trust, AI communications governance

Duolingo built one of the most distinctive brand voices in consumer technology: irreverent, warm, with a green owl whose passive-aggressive notifications became genuinely beloved internet content. The brand’s personality was central to its product: a human-feeling experience making language learning feel personal.

In April 2025, CEO Luis von Ahn published an internal memo, subsequently shared publicly on LinkedIn, declaring Duolingo to be an “AI-first” company. The memo outlined a plan to reduce reliance on contractors for work AI could handle, and stated that headcount would only increase if a team could not automate more of its work.

The timing compounded the problem. Duolingo had already laid off 10% of its contract translators and content writers in January 2024, replacing them with AI. The April 2025 memo arrived when those layoffs were already public. To users, educators, and labour advocates, the memo read less as a technology strategy and more as a confirmation of a pattern.

What went wrong

Social media erupted. Users threatened to delete the app. Educators questioned whether AI could genuinely replace the linguistic and pedagogical expertise of human translators. The backlash was strong enough that von Ahn subsequently walked back the framing, admitting the memo “did not give enough context”.

The decision to use AI in content production was not inherently the problem. The problem was announcing a philosophical commitment to replacing human work in public, without context, using a brand voice that had built its equity on warmth and humanity. The memo was written for an internal audience and was not prepared for external scrutiny.

Governance questions

  1. Who reviews external communications about your AI strategy (e.g. legal, brand, HR, communications?). Is there a clear owner for the public narrative?
  2. Has your organisation made any public statements about AI that could be read as dismissive of human creative or professional labour, and if so, how were they framed?
  3. How could your organisation test how an AI-related announcement might be interpreted by employees, customers, and other stakeholders before making it public?

Learning outcomes

After discussing this case, participants should be able to:

  1. Explain why AI strategy communications require the same scrutiny as advertising, particularly for brands whose equity depends on warmth, personality, or human connection.
  2. Identify the specific elements of the Duolingo memo that caused reputational damage: timing, tone, framing, and the absence of a human context narrative.
  3. Describe the communications governance questions a brand should ask before making a public statement about AI strategy: who is the audience, what is the context, and what will critics use this to say about us?
  4. Assess whether your brand’s public AI narrative is consistent with its brand values, and what the gap between them looks like to an external audience.

Discussion questions

  1. Has your organisation made any public statements about AI that could be read as dismissive of human creative or professional labour, and if so, how were they framed?
  2. Who reviews external communications about your AI strategy: legal, brand, HR, communications, and is there a clear owner for the narrative?
  3. If your brand is known for warmth, creativity, or human connection, how do you talk about AI in a way that reinforces rather than contradicts those values?