Users reported that Claude AI chatbot was becoming "lazy", but the official response was that no adjustments were made

Technology media The Decoder published a blog post yesterday (August 28),It is reported that Clause AI, a subsidiary of Anthropic ChatbotsBecoming more and more "lazy" and no longer as capable as before.

Netizen NextgenAITrading posted a message in the Reddit community 11 days ago, with the title "This is definitely not your illusion!Claude I have become more and more lazy recently". The post has received 485 likes and 290 replies.

Netizens said that Claude easily forgot tasks and struggled with basic programming, which ultimately led him to cancel his subscription.

Alex Albert of Anthropic responded, saying their investigation "did not reveal any widespread issues" and confirmed that they had not modified the Claude 3.5 Sonnet model or inference pipeline.

statement:The content of the source of public various media platforms, if the inclusion of the content violates your rights and interests, please contact the mailbox, this site will be the first time to deal with.
Information

AI threatens outsourced call centers in the Philippines, replacing 300,000 jobs in five years

2024-8-29 9:43:20

Information

The AI customer service market is huge! Bland AI raises $16 million to disguise itself as a human to provide telephone customer service

2024-8-29 9:49:45

Search