← sources

Who should be held responsible for AI errors?

source: reddit (reddit:economy)published 2026-10-30 09:47ingested 2026-10-31 02:40

retained text

Financial Times: A [study by Wharton School](https://knowledge.wharton.upenn.edu/wp-content/uploads/2026/10/Wharton-GBK-AI-Adoption-Report-2026.pdf) published this month found no consensus among US managers on who was “responsible for AI errors”, with 28 per cent saying a governance or risk team, 26 per cent the “deploying employee’s manager”, 25 per cent the employees, and 9 per cent the vendor.

My Opinion: AI vendors are generally not held responsible for AI errors. Instead it is the business governance or risk team, or the employee using AI or his manager. So those developing AI models and selling them to businesses aren't held responsible. Rather those deploying and using AI models are held responsible. 

Who do you think should be responsible for AI errors? I think that it depends on how the vendor markets and sells its products. If it is clear about the shortcomings of its products, and doesn't misrepresent it, than the vendor should not be responsible. If the business trains the vendors AI models, on its own data, than they would be responsible.

There should be a human in the loop for high risk systems. And that human should be accountable for AI errors. 

document fetched 2026-10-31 00:46 · canonical https://reddit.com/r/economy/comments/1xjb110/who_should_be_held_responsible_for_ai_errors