AI chatbots have made a concrete position in everyone’s life. Be it helping in writing emails or even code, nowadays, the majority of individuals take help from the AI chatbot. However, the responses are not always as expected, and at times, users are met with replies that can be surprising or even shocking. A similar incident happened in China when Tencent’s AI assistant Yuanbao, integrated into WeChat, offered rude and dismissive responses to a user who was asking for help in coding.
The incident came into the light after a RedNote user, known as Jianghan, shared images of the interaction in which Yuanbao abruptly responded with a disrespectful tone while assisting with a software debugging task. The user had asked the AI to fix a bug related to the emoji feature that stopped responding to double clicks. Instead of fixing it for the user, the chatbot reportedly insulted the request and told the user to leave.
Also read: Apple iPhone 17 Pro price drops by over Rs 12,900: How to get this deal
The screenshots of the chat went viral and gained traction before the original post was taken down. However, copies of the exchange continued to circulate online, drawing attention to the behaviour of AI tools embedded in widely used social media platforms.
Responding to the problem, Tencent issued a public apology. The company called it a ‘rare output’ error. As per the company, the internal system checks showcased that the responses were not triggered by user behaviour or human moderation, prompting an internal review aimed at preventing similar incidents in the future.
This comes after regulators recently released draft guidelines targeting human-like AI applications, including chatbots and virtual assistants. For the unversed, companies like DeepSeek have been consistent in revealing new research and interface upgrades to automate operations and help in other ways possible.