123RF has launched a generative AI-driven video comprehension capability utilizing Amazon Web Services (AWS), heralding a significant transformation in how users search for and license creative assets. The Malaysia-based stock media platform, which boasts over 230 million assets for 12.4 million users globally, unveiled this feature on December 3 during the re:Invent conference. This innovation aims to transcend traditional keyword-based searches by enabling AI to interpret the actual visual content within videos and images.
The new system leverages Amazon Nova to analyze visual components directly, enhancing the accuracy of descriptors and search relevance. Initial tests on a dataset of five million videos yielded a twofold increase in the accuracy of video descriptors, facilitating more precise content matching. This advancement is particularly advantageous for e-commerce businesses, which rely on swift and accurate asset selection to effectively pair visuals with product listings and bolster daily output.
Traditional search methods often yield irrelevant results when metadata is lacking or inaccurately labeled. The latest upgrade mitigates this issue by focusing on visual content rather than solely relying on tags. For instance, when a user queries “green bag,” the system identifies authentic images or videos depicting green bags, circumventing unrelated assets that merely include the term in their metadata or feature a green background. Additionally, the AI can recognize trademarked patterns, logos, and branded items, facilitating automated flagging and appropriate licensing categorization.
Bernadine Michael, Chief Marketing Officer at 123RF, emphasized that adopting AWS technologies has fundamentally reshaped the company’s operations. “AWS has fundamentally changed how we operate at 123RF, transforming us into an AI-powered creative enabler. Our marketing teams now launch campaigns 35% faster. This efficiency allows our teams to focus less on repetitive work and more on serving the creator economy,” she stated. She highlighted that the primary impact lies in democratizing creativity across its global subscriber base, enabling designers and marketers to locate the right asset instantly, irrespective of language or cultural context.
123RF processes over three million image uploads each month, necessitating compliance checks, copyright verification, and quality assessments. Previously, this workload required a team of 30 to 40 reviewers managing approximately 3,000 images daily. The platform also aimed to enhance its multilingual support, as most metadata was submitted in English, despite accommodating 15 languages. The challenge extended beyond mere translation to encompass understanding idioms, cultural contexts, and nuances.
The integration of AI-driven visual understanding and automated checks has slashed the overall content review time by 92%. Customers can now discover relevant assets in 90% less time, with improved accuracy and a reduction in irrelevant or duplicate results. Phoebe Liew, Chief Technology Officer at 123RF, remarked that the new system allows the platform to interpret images in ways that closely mirror human perception. “Our AWS-powered AI technology now ‘sees’ images the way humans do,” she noted. “When someone uploads an image, our system instantly captures what makes it visually unique – like identifying a specific architectural style or recognizing the composition of a sunset scene. This visual understanding means we can match similar images automatically, regardless of what language the description is in.” She added that this approach lessens reliance on contributor-generated keywords and aids in avoiding copyright issues by analyzing visual elements directly.
In order to support its growth, 123RF has utilized various AWS generative AI services. The company has employed Amazon Bedrock combined with Anthropic’s Claude 3 Haiku model to create a translation system attuned to cultural nuance. This feature helps users steer clear of misinterpretations, such as the differing implications of terms like “freedom” in English versus “jiyū” in Japanese. The system is also capable of identifying quality issues, copyright icons, and logos without requiring manual intervention.
The Amazon Nova Pro and Nova Lite understanding models enable the platform to assess copyright risks before content is published, providing an additional layer of protection for users who depend on the accuracy and legality of creative assets. Hussein Mohd. Ali, Country Manager for AWS in Malaysia, remarked that this collaboration illustrates how AI is redefining creative workflows. “123RF demonstrates how AI transforms creative workflows – tasks that once took designers days now take minutes, enabling Malaysian companies to compete on a global stage,” he stated. Research from AWS’s Unlocking Ambitions survey reveals that nearly half of startups are already adopting AI solutions, with almost a third developing entirely new AI-driven products.
With these latest enhancements, 123RF aims to establish a new standard in the content licensing sector, enabling creators, marketers, and businesses to operate more efficiently through AI-driven content discovery.
See also
OpenAI Launches GPT-5.1 with Enhanced Features and 800 Million Weekly Users
Generative AI Model Identifies Blood Cell Abnormalities with 95% Accuracy
DifuzCam Unveils Lensless Camera Design Using Diffusion Models and Masks
Gemini Enables Multimodal Input with Image Uploads for Enhanced AI Analysis
OpenAI’s Sam Altman Issues ‘Code Red’ as Google’s Gemini 3 Tops AI LLM Rankings
