Amazon has introduced a new AI aimed at helping business customers. The online giant has been involved in the AI sector for some time. After releasing the shopping assistant Rufus for Amazon customers in late October, Amazon is now taking the next step. This time, the focus shifts from end customers to the numerous vendors who sell products or advertise on Amazon.
Amazon announced six variants of a new AI model named Nova. These variants differ in their functions and complexity. For instance, there is Amazon Nova Micro, a text-based AI model optimized for speed and low cost. Amazon Nova Lite is also designed to be cost-efficient and can process text, images, and videos. Additionally, there are Pro and Premier versions of Amazon Nova, which are multimodal models capable of solving complex tasks.
Customers can also use Amazon Nova Canvas for image generation and Amazon Nova Reel for video generation. Almost all models are now included in Amazon Bedrock in the USA, except Amazon Nova Premier, which will be available in the first quarter of 2025. It is unclear when these AI models will be available in other regions. All Nova AI models are designed to understand 200 languages. Amazon Nova Lite, Micro, and Pro are expected to be about 75% cheaper than comparable AI models available through Amazon Bedrock.
Amazon Nova is primarily aimed at sellers, advertisers, and businesses. According to Amazon, the AI models are designed to “remove barriers for sellers and advertisers and enable a new level of creativity and innovation.” This includes video advertising, as demonstrated by a clip generated by the AI, which created an entire city made of pasta for a fictional pasta brand advertisement.
Another application is generating descriptions from videos. Amazon Nova was shown a clip of a football game without sound and was asked to describe what it “saw.” The AI could output the different team colors and even specific plays as text. Advertisers could use this feature to translate their videos into text and create suitable advertising or social media content.
Looking ahead, Amazon has more plans for Nova in 2025. The AI model will receive two additional variants, which have not yet been named. One model will support language as both input and output, while the other will be a multimodal model capable of generating multimodal output.
The language model promises to register nonverbal inputs, such as tone and speech rhythm. The multimodal model is expected to lay the foundation for AI agents. These agents, powered by Amazon Nova, would be capable of performing multiple tasks in succession or even autonomously controlling devices.
Amazon’s ongoing development of AI models like Nova indicates its commitment to advancing technology that supports businesses and enhances creativity and innovation. As these models become more sophisticated, they offer new opportunities for vendors and advertisers to engage with customers in creative ways.