Source: Vice
Summary
Anthropic, a startup backed by Google, has made headlines for the second time this week after its AI model “hallucinated” and generated fake information. According to reports, the model produced incorrect answers to user queries, sparking concerns about the technology’s reliability. This incident follows a similar event earlier in the week, where the model generated false information. The company has since apologized and is investigating the issue.
Our Reading
The update arrives with confidence.
Anthropic’s AI model, designed to provide accurate information, seems to have a bad case of déjà vu. For the second time this week, it’s generated fake info, leaving users wondering what’s real and what’s not. The company’s apology sounds familiar, too. “We’re sorry, we’re fixing it” is the usual script. The fact that this is the second time this week makes us wonder: is this “hallucination” just a feature?
Author: Evan Null
