Chinese technology giant Alibaba has updated its open source video-generating artificial intelligence (AI) model.
This is described as part of the company's rapid pace of AI upgrades to keep up with the US-China competition, according to a report by Bloomberg News on Tuesday (August 26).
The update will turn portrait photos into “movie-quality avatars” and encourage them to talk, sing and perform, Alibaba said.
Bloomberg also noted that Alibaba has invested heavily in AI following the rise of Deepseek this year.
“We read their research papers and said, 'The Holy Cow… We're behind,'” Alibaba Chairman Joe Tsai said in a June fireside chat at the Viva Tech 2025 conference in Paris.
A few weeks later, Alibaba announced the Qwen series of LLMS and chose to open source many small models. Tsai said this will democratize access to AI, promote third-party innovation, and drive demand for the company's cloud computing infrastructure.
As Bloomberg pointed out, the company is facing competition from Google, AI startup Manus, Kuaishou technology and others, all of which have debuted new or updated video tools in recent months.
So far, the pivot has not been rewarded, the Bloomberg report added.
Alibaba reported a 7% increase in revenue in May as it faced a recession in consumer spending. The company is expected to release its latest quarterly profit on August 29th.
In other AI news, Pymnts wrote on Tuesday about the findings of the August edition of the CAIO report, “Tech on Technology: How Technology Secer promotes the adoption of agent AI.”
“Sectors such as software and financial services are moving forward, supported by engineering talent, an agile risk culture and flexible budgets,” writes Pymnts. Meanwhile, the goods and services industries, such as manufacturing, logistics, retail and hospitality, are lagging behind and restrained by structural fragmentation, operational complexity and more laidback paths (ROIs) to return to investment.
“Ultimately, what research highlights is that the evolution of AI's novelty to full AI autonomy turns on two axes: trust, industry tail, industry. Trust represents the confidence that the system functions as intended, and the tail speaks to the strategic and structural benefits tied to specific sectors that support rapid and risky adoption.
