Kuaishou Technology, a Chinese AI and technology company, launched a new text-to-video model called Kling this year. Su Hua, a Chinese billionaire internet entrepreneur, is the co-founder and CEO of the video platform Kuaishou, known outside of China as Kwai.
Several AI enthusiasts shared their creations from Kling on X that captured the hearts of internet users worldwide. A series of animals and objects were featured enjoying a meal of noodles. From a panda munching on a bowl of ramen to a kangaroo slurping up some udon, the videos are both hilarious and heartwarming.
A few others include blueberries turning into puppies and a tray of apples turning into guinea pigs, which mess with your head.
The level of detail and realism in the videos is a testament to the capabilities of Kling and the progress made in the field of AI.
Known for the creation of TikTok competitor, Kuaishou joined the race with other Chinese tech companies to rival OpenAI’s Sora.
With simple text prompts, it can generate highly realistic videos in 1080p high-definition resolution. The videos can be up to two-minutes long. Sora, on the other hand, makes 60-second videos with text prompts.
Kling boasts the ability to produce realistic motions on a large scale, simulate the attributes of the physical world, and weave concepts and imagination together setting a new benchmark in AI-powered video creation.
However, as impressive as Kling AI may be, its accessibility is primarily limited to a few users even though the company claimed it would be available worldwide. This poses significant challenges for its global adoption.
For some users who were looking forward to accessing its offerings, this situation may feel like a huge letdown.
A worldwide release creates expectations of inclusivity and accessibility; when these are unmet, it can harm the company’s reputation. Despite Kling AI’s impressive features, its primary hurdle is limited availability.
Currently, its access is mostly limited to invited beta testers, with some users in China able to experience a limited demo version through the Kuaishou app, as claimed by ChatGPT on Quora.
At a time when the US is heavily debating AI ethics and incorporating ‘Responsible AI’, China seems unperturbed and is likely responding to these AI ethicists with a Kling.
The AI company hit the headlines recently by announcing the global launch of its International Version 1.0, a platform designed to revolutionise industries worldwide. This milestone release features advanced machine learning, multilingual support, and enhanced data analytics, promising unparalleled efficiency and innovation across sectors.
AI Video Generator War Begins!
While systems like OpenAI’s Sora and Kuaishou’s Kling have showcased impressive capabilities, they remain accessible only to a select group of users. Similarly, Luma AI’s Dream Machine also boasts remarkable features but is limited to a restricted audience.
Interestingly, Kuaishou’s AI tool entered the market shortly after Vidu AI, another Chinese text-to-video AI model known for producing HD 1080p 16-second videos.
This model’s launch coincides with a flurry of activity in the generative AI sector, as startups and tech giants compete to develop advanced tools that create realistic images, audio, and video from text inputs.
It has a user-friendly interface that supports text-to-video or text-to-image generation.
Unlike Runway, Haiper, and Luma Labs, it prompts up to 2,000 characters, enabling highly detailed descriptions. It performs better with lengthy, well-crafted prompts.
This cutting-edge AI model employs variable resolution training, enabling users to produce videos in various aspect ratios. Remarkably, it can showcase full expression and limb movement from a single full-body image.
AI video creation seems like the next battleground for tech companies with contenders like OpenAI’s Sora, Microsoft’s VASA-1, Adobe’s Firefly, Midjournery, and Pika Labs, already in the game.
Furthermore, Google recently introduced Veo, a new text-to-video AI model, at Google I/O to compete with OpenAI’s Sora. Veo improves on previous models, offering consistent, high-quality over-a-minute-long 1080p videos.
While some were impressed with Veo’s capabilities, others argue that it may not exactly be state-of-the-art in its latency or abilities compared to Sora.
Now that Kling is here, the benchmark of making cinematically impressive and real-world-like videos has gone up.
Why is Kling a big deal?
This month, Runway introduced Gen-3, which offers enhanced realism and the ability to generate 10-second clips. Last month, Luma Labs unveiled the impressive Dream Machine.
These new model updates were initially spurred by the release of Sora earlier this year, which remains the benchmark for AI video generation. Recently, a series of short films on YouTube showcased Sora’s full potential. Additionally, Kling played a significant role in the wave of updates.
It also adopts a unique approach to AI by incorporating generative 3D in its creation process. It provides Sora-level scene changes, clip lengths, and video resolution. Given that OpenAI only grants a limited number of select creators access to Sora, Kling AI might just be the top choice for now.
Capabilities of Kling AI
Kling AI is accessible via the Kuaishou app, available on both iOS and Android platforms. This mobile app puts Kling AI’s advanced video generation capabilities directly at users’ fingertips, enabling them to create high-quality, realistic videos from their smartphones.
For users outside China, accessing Kling AI often requires navigating around these barriers. Some have resorted to emailing Kuaishou directly to request access, explaining their interest in becoming beta testers.
The competitive landscape is evolving, but the restrictions on access can hinder Kling’s ability to gain traction outside China.
Chinese attempts to lure domestic developers away from OpenAI – considered the market leader in generative AI – will now be a lot easier, after OpenAI notified its users in China that they would be blocked from using its tools and services.
“We are taking additional steps to block API traffic from regions where we do not support access to OpenAI’s services,” said an OpenAI spokesperson.
OpenAI has not elaborated about the reason for its sudden decision.
ChatGPT is already blocked in China by the government’s firewall, but until this week developers could use virtual private networks to access OpenAI’s tools in order to fine-tune their own generative AI applications and benchmark their own research. Now the block is coming from the US side.
The OpenAI move has “caused significant concern within China’s AI community”, said Xiaohu Zhu, the founder of the Shanghai-based Centre for Safe AGI, which promotes AI safety, not least because “the decision raises questions about equitable access to AI technologies globally”.
