- OpenAI released their chatbot in late 2022, followed by Google's chatbot, Bard, in March.
- Italy became the first Western country to ban ChatGPT earlier this month under suspicions the software may breach Europe’s strict privacy policy.
- The Cyberspace Administration of China released guidelines for AI on Tuesday, including a security assessment before any AI product can go live and a requirement that all products must represent the core values of socialism.
- The U.S. launched a public consultation on AI regulation on Tuesday, allowing public comments on AI regulation for 60 days, which will then be used to help lawmakers approach AI development.
- In 2021, the Biden Administration banned the export of high-powered AI accelerator chips to China, forcing the use of multiple low-powered chips instead.
The use of chatbots has become increasingly popular in recent years, as companies try to enhance their customer service capabilities and improve operational efficiency. Several major tech companies, including Facebook parent company Meta, Alibaba, SenseTime Group, and Baidu, have all introduced AI-powered chatbot programs to capitalize on this trend.
Meta’s Segment Anything Model (SAM) is a recent release that can identify specific writing in pictures and videos, removing unwanted elements by clicking on them, drawing a box around them, or entering their name into a text prompt. SAM runs on “the largest ever segmentation dataset” created, and was trained on 11 million images and has over one billion masks. Another chatbot from Meta, called Llama, was released as an open-source package for the AI community.
Alibaba has also recently unveiled its AI chatbot, Tongyi Qianwen, which is available for enterprise customers to test in both Chinese and English. The chatbot can create business pitches, write emails, draft travel itineraries, and jot down meeting minutes, and will be initially embedded in the workplace messaging platform Dingtalk and in its Tmall Genie smart speakers.
SenseTime Group has released its user-facing AI chatbot, SenseChat, which includes features like writing a children’s story, computer code, and emails. Meanwhile, Baidu has introduced its AI chatbot, Ernie Bot, which is based on the company’s deep learning model Ernie and can calculate mathematical problems, generate pictures and videos, and speak in Chinese dialects.
While these chatbots have the potential to improve customer service and operational efficiency, they are not without their controversies. SenseTime Group was recently sanctioned by the US government for using facial recognition to surveil the Uyghurs in China, while Baidu sued Apple for placing fake versions of Ernie Bot on the App Store. Despite these challenges, the use of AI-powered chatbots is likely to continue to grow, as companies seek to stay ahead in the chatbot race.
inputs from // forbes
