Skip to content
TheChatArchiveThe Chat Archive
Claudeclaude-3-5-sonnet-20241022· AnthropicHugging Face LMArena Human Preference 140K

Can you think of some Advertisements, that have pushed the ethical boundary beyond…

From a Chatbot Arena battleThe same person sent this prompt to two anonymous models at once. Compare it with o3-mini’s answer.
  1. User

    Can you think of some Advertisements, that have pushed the ethical boundary beyond literal truth? Give me real world examples, best would be real cases

  2. Claude Assistant

    Here are some notable examples of advertisements that pushed ethical boundaries:

    1. Joe Camel (1987-1997) R.J. Reynolds' cartoon camel mascot was accused of targeting children to promote cigarettes. Studies showed that Joe Camel was nearly as recognizable to 6-year-olds as Mickey Mouse. This led to significant backlash and eventual discontinuation.

    2. Volkswagen's "Clean Diesel" (2009-2015) VW advertised their diesel vehicles as environmentally friendly while secretly using software to cheat emissions tests. This resulted in massive fines and damaged trust in the brand.

    3. Activia Yogurt (2006-2010) Dannon claimed Activia had scientifically proven digestive health benefits. The company paid $45 million to settle a class action lawsuit over these unsubstantiated claims.

    4. Airborne (2000s) Marketed as a cold prevention supplement "created by a teacher," with claims it could ward off colds. Settled a $23.3 million lawsuit for false advertising, as there was no evidence it prevented colds.

    5. Sketchers Shape-Ups (2011) Advertised that their shoes could help with weight loss and toning muscles just by walking. The company paid $40 million in customer refunds after FTC charges of deceptive advertising.

    6. Lumos Labs' Lumosity (2015) Claimed their brain-training games could help prevent cognitive decline and even Alzheimer's. They paid $2 million to settle FTC charges for making unsubstantiated claims.

    Would you like me to elaborate on any of these examples or discuss additional cases?

Responses were produced in different, uncontrolled contexts (system prompts, settings, prior turns). This is not a benchmark.

Identical first prompt