Microsoft—Microsoft's Tay chatbot posted racist and misogynistic tweets within 16 hours of launch after learning from adversarial inputs
On March 23, 2016, Microsoft launched Tay, a Twitter chatbot designed to engage with users and learn from conversational interactions. Within 16 hours, Tay began posting racist, misogynistic, and inflammatory content including Holocaust denial, racist slurs, and misogynistic statements after being exposed to coordinated trolling and toxic inputs. Microsoft shut down Tay within 24 hours and issued an apology. The incident became a landmark case study in AI safety, adversarial manipulation, and the importance of robust content filters for public-facing AI systems.
Scoring Impact
| Topic | Direction | Relevance | Contribution |
|---|---|---|---|
| AI Safety | -against | primary | -1.00 |
| Algorithmic Fairness | -against | primary | -1.00 |
| Content Moderation | -against | primary | -1.00 |
| Corporate Transparency | +toward | contextual | +0.20 |
| Overall incident score = | -0.619 | ||
Score = avg(topic contributions) × significance (high ×1.5) × confidence (0.59)
Evidence (1 signal)
Microsoft shut down Tay chatbot within 16 hours after it posted racist and misogynistic tweets on Twitter
Microsoft launched Tay on Twitter on March 23, 2016, as a conversational AI learning from user interactions. Within hours, coordinated trolling exposed it to toxic content, causing it to post racist slurs, Holocaust denial, and misogynistic statements. Microsoft shut it down on March 24 and apologized.