AI News Flash: OpenAI Redefines Core Values, vivo Set to Release Self-Developed Large Model

Recently, a series of significant developments have occurred in the field of artificial intelligence. OpenAI has adjusted its core values, prioritizing general artificial intelligence (AGI) as its primary goal. Meanwhile, several tech companies are laying out plans for AI large models, with vivo announcing the launch of its self-developed AI large model matrix. Additionally, the U.S. government has stated it will further tighten restrictions on AI chip exports to China.

OpenAI Redefines Core Values, Focusing on AGI as the Primary Goal

OpenAI recently revised its core values, placing general artificial intelligence (AGI) at the forefront, demonstrating a strong pursuit of AGI. The new core values include: focusing on AGI, being eager and motivated, scaling, creating products that users love, and team spirit. OpenAI states its commitment to building safe and beneficial AGI, which will have a tremendous positive impact on the future of humanity.

vivo is about to release its self-developed AI model matrix

vivo announced that it will release its self-developed AI large model matrix at the vivo Developer Conference on November 1, covering five self-developed large models with different parameter scales of one billion, ten billion, and one hundred billion, fully covering core application scenarios. It is worth mentioning that vivo's self-developed large model currently ranks first in the C-Eval global Chinese leaderboard.

Qualcomm Snapdragon Summit Focuses on AI, Black Shark Smartwatch Equipped with Baidu Large Model

Qualcomm will hold the 2023 Snapdragon Summit on October 25-26, expected to be themed around AI. At the same time, Black Shark announced its new smartwatch S1 Pro, which is equipped with Baidu's "Wenxin Yiyan" large model, supporting AI Q&A functionality, map navigation, and will offer a large number of free AI watch faces.

The United States will strengthen export restrictions on AI chips to China

A U.S. official stated that the U.S. will take further measures to prevent American chip manufacturers from circumventing government restrictions on exporting semiconductor products to China, in order to "plug policy loopholes." New measures are expected to be updated this week, which will block the export of certain artificial intelligence chips that just meet the current technical parameters, while requiring companies to report the export volume of other chips.

SenseTime Upgrades Medical Health Large Model, Huawei May Launch New AI Chat Assistant

SenseTime Technology has released an upgraded medical health large language model "Daiyi", supporting 13 preset scenarios and personalized features. Huawei recently applied to register the trademark "Huawei eKitChat", which may indicate that Huawei's Kunling brand will launch AI chat assistants or AI customer service products.

Multi-party Jointly Established Artificial Intelligence Safety Governance Special Committee

The China Cybersecurity Association officially established a Professional Committee for Artificial Intelligence Security Governance. This committee was jointly initiated by 58 organizations including the National Internet Emergency Center, Beijing Academy of Artificial Intelligence, Shanghai Artificial Intelligence Laboratory, and several technology companies, aiming to enhance the capability of artificial intelligence security governance.

"Jinjing Yunhua" Completes Series B Financing, Focuses on Cybersecurity Large Models

Beijing Jinjing Yunhua Technology Co., Ltd. recently completed nearly 100 million yuan in Series B financing, with the funds primarily used to develop large model products for the cybersecurity industry, expand the R&D team size, and broaden the product market.