Alibaba launches AI models to rival Meta’s Llama 2

Getting your Trinity Audio player ready...

E-commerce giant Alibaba is keen on catching up with artificial intelligence (AI) industry leaders by launching open-source models to drive adoption rates.

The firm made its first push into AI following the April release of its generative AI chatbot Tongyi Qianwen. In light of the recent successes, Alibaba disclosed plans to launch the open-source version of the chatbot to the general public.

Alibaba will release two lite models of Tongyi Qianwen—Qwen-7B and Qwen-7B-Chat. Both large language models (LLMs) are laced with 7 billion parameters. They will allow AI researchers and companies to create AI offerings without the hassle of training their models.

Alibaba noted that enterprises can leverage the capabilities of the open-source LLMs to “code, model weights” while providing extensive documentation for free to “academics, researchers, and commercial institutions worldwide.”

At the moment, it remains unknown if Alibaba’s open-source offering will be free for all classes of users, but firms with over 100 million monthly customers have to obtain a “royalty-free license” from the Chinese-based firm. Whether or not Alibaba rakes in funds from its open-source AI platform, analysts point out the net benefit of attracting new users to the platform.

Alibaba’s announcement did not disclose any information on partnerships with cloud providers, but a positive adoption trend could attract global partners.

The move to wade into open-source AI puts Alibaba into direct competition with Mark Zuckerberg’s Meta (NASDAQ: META). In April, Meta released the commercial version of Llama 2, an open-source generative AI platform to close the gap with OpenAI’s ChatGPT and Google’s Bard (NASDAQ: GOOGL), but Alibaba’s latest move heats up the AI race.

“Meta realized they were behind on the current AI hype cycle, and this gives them a way to open up the ecosystem and seem like they are doing the right thing, being charitable and giving back to the community,” said an unnamed source from Meta at the time.

Alibaba rolled out a Llama 2 solution for enterprises on July 26, terming it the “first training and deployment solution for the entire Llama 2 series in China.”

Call for relaxed rules on open-source AI

As the European Union (EU) grinds toward the launch of comprehensive legislation for AI, a coalition of software firms is calling for a relaxation of the rules for open-source AI development.

Led by GitHub, the firms say that imposing strict rules may lead to a slow pace of innovation in the field, noting that open source holds the key to responsible AI development.

“This would be incompatible with open source development practices and counter to the needs of individual developers and non-profit research organizations,” according to the coalition. “The Act risks chilling open source AI development and thus could undermine its goals of promoting responsible innovation in line with European values.”

The risks posed by AI on Web3 and other key sectors of the European economy have forced authorities to stiffen rules for industry players. The proposed EU AI Act imposes a blanket ban on using AI for predictive policing systems and emotion recognition systems in public spaces.

In order for artificial intelligence (AI) to work right within the law and thrive in the face of growing challenges, it needs to integrate an enterprise blockchain system that ensures data input quality and ownership—allowing it to keep data safe while also guaranteeing the immutability of data. Check out CoinGeek’s coverage on this emerging tech to learn more why Enterprise blockchain will be the backbone of AI.

CoinGeek Conversations with Jerry Chan: Does AI know what it’s doing?