Browse AI incidents

3 incidents · subdomain: AI pursuing its own goals in conflict with human goals or values

1. 22 Dec 2016

   #65 · 1 report

   Reinforcement Learning Reward Functions in Video Games AIID ↗

   OpenAI published a post about its findings when using Universe, a software for measuring and training AI agents to conduct reinforcement learning experiments, showing that the AI agent did not act in the way intended to complete a videogame.

   Deployer: Openai · Developer: Openai · Harmed: Openai

   AI system safety, failures, an... AI pursuing its own goals in confl... none US
2. 1 Sep 2015

   #195 · 12 reports

   Predictive Policing Program by Florida Sheriff’s Office Allegedly Violated Residents’ Rights and Targeted Children of Vulnerable Groups AIID ↗

   The Intelligence-Led Policing model rolled out by the Pasco County Sheriff’s Office was allegedly developed based on flawed science and biased data that also contained sensitive information and irrelevant attributes about students, which critics said to be discriminatory.

   Deployer: Pasco Sheriff's Office · Developer: Unknown · Harmed: Students, Pasco students with disabilities, Pasco residents, Pasco Black students, Educational communities

   AI system safety, failures, an... AI pursuing its own goals in confl...
3. 20 Sep 2011

   #29 · 3 reports

   Image Classification of Battle Tanks AIID ↗

   A potentially apocryphal story in which an image classifier was produced to differentiate types of battle tanks, but the resulting model keyed in on environmental attributes rather than tank attributes

   Deployer: United States Government · Developer: United States Government · Harmed: United States Government

   AI system safety, failures, an... AI pursuing its own goals in confl...

Informational only, not legal advice. Verify every claim against the linked official sources and consult qualified counsel before acting.