Shutterstock’s first-ever partnership with a video-to-music platform licenses its music catalog for AI model training—a new benchmark for responsible AI music

Sonilo, the world’s first professionally licensed video-to-music AI platform, announced a strategic partnership with Shutterstock, one of the world’s largest providers of licensed creative content.

Through this partnership, Sonilo has licensed Shutterstock’s high-quality music catalog for AI model training—making this Shutterstock’s first-ever partnership with a video-to-music AI platform. Sonilo is headquartered in San Francisco and backed by B Capital, a venture capital firm affiliated with the Boston Consulting Group focused on emerging technology startups expanding globally.

Marketing Technology News: MarTech Interview with Miguel Lopes, CPO @ TrafficGuard

Redefining licensed AI music for video

The AI music industry is at a crossroads. By 2025, infringement cases filed against AI companies had surged by more than 130% year over year. The dominant response from AI music platforms has been to train on unlicensed content first and deal with the legal consequences later—producing AI-generated music that not only raises serious copyright concerns, but floods the internet with low-quality content that shortchanges creators and audiences alike.

Sonilo is built to address these problems. Powered by its proprietary Sonilo v1.0 model, Sonilo takes the video itself as input rather than relying on text prompts. It reads the footage, understands its pacing, structure and emotional arc, and composes an original soundtrack to match. No prompts, no library searches, and no manual syncing. Every soundtrack generated is original, production-ready, and cleared for commercial use—no additional licensing required, whether for social content, branded video, or broadcast use.

Most AI music platforms have treated licensing as an afterthought, training on copyrighted content without authorization and leaving artists without compensation. Sonilo was built differently. By securing professional licensing partnerships upfront, Sonilo ensures that every model it trains is built on content that artists have consented to and been compensated for. This partnership with Shutterstock is a direct expression of that commitment: the music industry doesn’t have to choose between innovation and integrity.

“Music has always been the last unsolved layer of video creation, and video has always carried its own soundtrack,” said Shawn Song, CEO, Sonilo. “We built Sonilo to hear it and compose from it, without a single text prompt. But how we build matters as much as what we build. While others have chosen to take artists’ work without permission and charge creators for the privilege, we’ve chosen a different path—one where artists are compensated from day one. Partnering with Shutterstock reflects that standard. Every model we train meets a bar the music industry can stand behind, because the most innovative AI platforms don’t have to come at the expense of the artists who make all of these possible.”

“AI innovation depends on access to high-quality, rights-cleared content and trusted licensing partnerships,” said Jessica April, Vice President, Data Licensing & AI Services, Shutterstock. “Sonilo’s approach reflects the growing demand for responsibly sourced training data and commercially safe AI workflows. We’re pleased to support companies building generative AI products with licensed content and scalable data solutions that help accelerate innovation while respecting creators and rights holders.”

Marketing Technology News: Disrupt or Be Disrupted: The AI Wake-Up Call for B2B Marketers

Scaling across platforms, built for every creator

This partnership is part of Sonilo’s broader momentum. Earlier this month, the platform launched as a native node inside ComfyUI, the open-source AI workflow tool used by over four million creators. Sonilo is also available via API for integration into video tools, creator platforms, game engines, and AI systems, and offers Free, Pro, and Premium tiers. Further platform integrations are planned throughout the rest of this year.

New addition to the Jina v5 model family delivers flexible, cost-efficient AI search across media types without rebuilding existing systems

AKOOL announced a major breakthrough in AI video infrastructure with the launch of its production-grade video inference engine, delivering 10–20× faster performance than conventional approaches and enabling real-time AI video applications at global scale.

Traditional AI video generation has been constrained by speed, often requiring tens of seconds to produce a single clip. AKOOL’s new system reduces that process to as little as one to three seconds, while also supporting real-time streaming with sub-30 millisecond latency per frame—enabling live, interactive video experiences.

AKOOL launches real-time AI video engine with 10–20× faster performance and sub-30ms latency at scale

“This is the moment AI video becomes truly usable,” said Jiajun (Jeff) Lu, CEO of AKOOL. “We didn’t just optimize one part of the pipeline—we rebuilt the entire stack. The result is a system that can generate and stream AI video in real time, reliably and at scale.”

Marketing Technology News: MarTech Interview with Miguel Lopes, CPO @ TrafficGuard

The performance gains come from a full-stack approach spanning every layer of the AI pipeline, from algorithm design to hardware execution. By reducing computational steps, increasing parallel processing across GPUs, eliminating runtime overhead, and leveraging next-generation GPU architectures, AKOOL achieves order-of-magnitude improvements over standard frameworks.

The same infrastructure operates across cloud environments, real-time streaming systems, and on-device deployments, enabling developers and enterprises to build and scale AI video applications across platforms—from batch generation to live video and mobile experiences.

Marketing Technology News: Is the Traditional CDP Already Out of Date?

Beyond speed, the system is designed for production reliability, with real-time monitoring, automated quality controls, staged deployments, and per-model cost tracking—ensuring safe, scalable deployment without sacrificing performance.

The new inference engine is already powering AKOOL’s suite of products, including Akool Live Camera, enabling real-time digital avatars, live translation, and interactive video experiences. These capabilities mark a shift from AI video as a post-production tool to a live, interactive medium.

With this launch, AKOOL is establishing a new foundation for AI video—one that moves beyond generation into real-time communication, opening the door to new forms of content, collaboration, and digital presence.

VideoMagic Positioned at the Intersection of Generative AI, Content Intelligence, and Video Automation

VideoMagic, an AI-native platform pioneering the emerging “understanding engine” category, today announced the appointment of Alyson Young Castillo to its Advisory Board.

Castillo brings extensive experience advising and scaling growth-stage SaaS and technology-enabled companies, with a strong focus on go-to-market (GTM) strategy, revenue acceleration, and category development. In her role as an Advisory Board member, she will provide strategic guidance to support VideoMagic’s continued expansion, with an emphasis on refining GTM execution and accelerating market adoption.

Marketing Technology News: MarTech Interview With Jay H. Lee, Chief Marketing and Growth Officer @ Five9

VideoMagic operates at the convergence of generative AI, content intelligence, and video automation—enabling organizations to convert documents, presentations, and web content into high-quality, short-form videos within minutes. Beyond content transformation, the platform allows users to query, synthesize, and activate large volumes of unstructured data.

The platform is delivered as a Software-as-a-Service (SaaS) solution with a flexible subscription model, including a limited free tier, supporting both self-service adoption and enterprise-scale deployments.

VideoMagic’s AI-powered chat assistant seamlessly integrates directly into an organization’s digital properties—empowering users to engage with content, ask questions, and receive contextual, intelligent responses about products, services, and related information.

This combination of content ingestion, multimodal output (video), and interactive AI engagement positions VideoMagic as foundational infrastructure for how enterprises will communicate, distribute, and monetize knowledge in an AI-first world.
“Alyson brings a proven track record of helping companies define new categories and scale go-to-market execution,” said John Sculley, Chaiman-of-the-Board of VideoMagic. “Her ability to bridge product innovation with commercial strategy will be instrumental as we expand into new markets and deepen enterprise adoption.”

“I’m excited to join VideoMagic at this pivotal stage,” said Alyson Young Castillo. “We are witnessing a fundamental shift in how organizations interact with content—we are moving from information to understanding. VideoMagic is the layer that makes your data make sense. It converts complex data into dynamic short-form videos and powers a chat interface to explore, query and act on the underlying information in real time. This puts VideoMagic at the forefront of this transformation. I look forward to supporting the team as they scale.”

VideoMagic is gaining traction among enterprises, digital-first organizations, and content-rich businesses seeking to modernize customer engagement, streamline knowledge delivery, and unlock the value of their existing content at scale.

Training AI model on video data processed by Beamr’s content-adaptive technology made the model more resilient to compression, by lowering depth estimation error on safety-critical road users, including pedestrians and motorcyclists, by 30.7% 

Beamr Imaging Ltd. (NASDAQ: BMR), a leader in video optimization technology and solutions, released research demonstrating that machine vision models fine-tuned on video compressed by Beamr’s patented Content-Adaptive Bitrate (CABR) technology are more resilient than models trained on uncompressed data, while reducing the video data volumes that machine vision development depends on.

Machine vision teams handling petabyte-scale video data for autonomous vehicles (AV) and other video AI applications typically consider compression as a process for managing this scale. The findings reframe adaptive compression as an asset that strengthens AI model resilience, with the advantages of reducing storage and networking costs and infrastructure. This research extends Beamr’s ML-Safe benchmarks, validating a potential performance asset for AI models trained across machine vision applications.

Marketing Technology News: MarTech Interview with Stephen Howard-Sarin, MD of Retail Media, Americas @ Criteo

The research evaluated Depth Anything V2, a state-of-the-art monocular depth estimation model. The model was fine-tuned on AV video data compressed with Beamr’s technology that delivered 35.2% file-size reduction relative to baseline compression. The fine-tuned model demonstrated 30.7% reduction in depth estimation error on vulnerable road users, including pedestrians and motorcyclists, and 16.0% aggregate reduction across all object classes. Full methodology and results are available in the blog post.

“This research shows that compressed video data can produce models that are more robust, not less,” said Dani Megrelishvili, Beamr CPO. “That points to a different role for compression in our customers’ pipelines, from a cost they tolerate to a tool they deploy.”

Marketing Technology News: From MarTech Stack to MarTech Fabric: Weaving Brand, Content, and Conversion Into One Thread

“Machine vision teams have faced a structural trade-off: compress video data to manage scale, or face the escalating costs and infrastructure challenges of running AI models without compression,” said Ronen Nissim, ML Lead at Beamr. “Our research suggests this trade-off is more flexible than the industry may have assumed. By using compressed footage as augmentation during fine-tuning, we produced a model that performed better on the validation set than the equivalent model trained on uncompressed data.”

Beamr’s ML-safe benchmarks have previously validated content-adaptive compression across the AV development pipeline. The benchmarks demonstrated up to 50% file size reduction while preserving object detection accuracy at mean average precision of 0.96, with high fidelity across detection, localization, and confidence consistency. Subsequent testing for captioning workflows in world foundation model pipelines showed 41%–57% file size reduction with no measurable impact on the pipeline outputs.

Platform integrates generation, editing, asset management, and distribution tools to support structured AI content production

Cannon Studio announced ongoing development of its AI-powered platform designed to support video creators, filmmakers, and digital teams working with artificial intelligence tools.

The future of AI video is not random one-off clips. It is consistent worlds, better workflows, stronger communities, and tools that help creators finish what they start”

— Founder of Cannon Studio

As AI-generated video content becomes more widely adopted, creators are increasingly using multiple tools for ideation, generation, editing, and publishing. This workflow can introduce fragmentation, requiring users to manage assets and processes across separate platforms.

Cannon Studio is developing a system intended to consolidate these stages into a single environment, allowing users to manage content creation workflows from initial concept through final output.

Marketing Technology News: MarTech Interview with Stephen Howard-Sarin, MD of Retail Media, Americas @ Criteo

Structured Approach to AI Content Creation

Cannon Studio includes a system for organizing content into reusable components.

Users can create and manage assets such as characters, locations, objects, and visual references, which can be reused across different projects. This approach is designed to help maintain consistency across scenes and outputs without requiring repeated setup.

The platform also includes a workflow framework referred to as Creator Flow, which allows users to organize projects into scenes and individual shots. Each shot can include prompts, dialogue, and reference elements, providing a structured format for generating and refining content.

Integration of Generation and Production Tools

In addition to content generation, Cannon Studio incorporates tools for editing and post-production within the same environment.

These tools include:

– Video trimming and cropping
– Format conversion and compression
– Basic audio and narration workflows
– Output preparation for publishing

The platform also provides access to multiple AI video generation models within its interface, allowing users to select different options depending on project requirements.

Development of Distribution and Asset Systems

Cannon Studio is also developing features related to content distribution and asset management.

Marketing Technology News: From MarTech Stack to MarTech Fabric: Weaving Brand, Content, and Conversion Into One Thread

The platform includes Cannon TV, an internal environment where users can share video content. This includes support for longer-form content as well as shorter video formats.

In addition, the company is working on an asset marketplace where users will be able to manage, and potentially exchange, reusable creative elements such as characters, environments, and visual components within the platform.

Mobile Access and Platform Expansion

According to the company, a mobile application is currently in development. The app is expected to provide access to projects, content review, and sharing capabilities outside of desktop workflows.

The platform is designed to support creators working across different environments, including individual users, small teams, and content production groups.

Evolving AI Content Workflows

The development of platforms like Cannon Studio reflects a broader shift in how AI tools are used in content creation.

While early AI tools focused primarily on generating individual outputs, newer systems are increasingly incorporating workflow management, asset organization, and production features.

This shift is being driven by the need for more structured processes as creators move from experimentation to repeatable content production.

Trusted by the industry’s leading advertisers and publishers, the solution leverages real-time contextual video intelligence and pod-level targeting to connect viewers to commerce in high-attention and high-intent moments

V4b.AI is proud to announce the launch of its proprietary AI-driven video solution, designed to meet the growing demands of modern businesses for high-quality, scalable video content.

The newly launched system addresses a significant barrier in generative AI applications by maintaining product accuracy, apparel fidelity, and the consistency required for real-world business environments. By blending multiple foundational AI models, 3D-based accuracy layers, and structured post-production workflows, V4b.AI provides a reliable solution for creating video content tailored to ecommerce, fashion, advertising, and digital platforms. As demand for short-form video content escalates across industries like ecommerce, apparel, and D2C brands, the traditional approach to video production has proven costly, time-consuming, and challenging to scale. V4b.AI’s innovative system offers an alternative by enabling businesses to produce high-quality, tailored content faster and more efficiently. This eliminates reliance on physical shoots while empowering brands to create content effectively for global markets.

Marketing Technology News: MarTech Interview with Miguel Lopes, CPO @ TrafficGuard

“Technology has always amplified expertise, not replaced it. From Hollywood to high-end post-production studios, the model has been consistent, with controlled systems operated by specialists to deliver precision and scale. What we are building follows the same principle. Businesses should not have to become content production experts to use AI. They need reliable, high-quality outputs that help them scale, while we handle the complexity, control, and accuracy behind it,” said Manish Kumar, Founder of V4b.AI.

Founded by Manish Kumar, who brings over 20 years of experience across content, marketing, and technology, and who has built and run a successful digital creative agency over the past decade, V4b.AI represents a natural evolution of solving content challenges at scale.

Marketing Technology News: Is the Traditional CDP Already Out of Date?

V4b.AI’s AI-powered video solution is poised to transform the way businesses approach commercial video creation, offering a scalable, accurate, and efficient alternative to traditional production methods.

New platform verifies claims across YouTube Live, Discord, and social media with structured verdicts and real sources.

ClaimCrush, an AI-powered fact-checking platform, today announced the launch of new capabilities that allow users to verify claims across videos, images, text, and live streams in real time.

Designed for content creators, journalists, and researchers, ClaimCrush analyzes spoken and written statements, extracts individual claims, and evaluates them against credible sources.

The platform returns structured verdicts such as Supported, Partially Supported, or Contradicted, alongside clear reasoning.

A key innovation is its live fact-checking capability, enabling users to verify claims instantly during YouTube Live streams, Discord calls, Instagram Live sessions, and other real-time environments with a single click.

Marketing Technology News: MarTech Interview with Miguel Lopes, CPO @ TrafficGuard

The company also introduced a debate feature, allowing users to engage in discussions while the AI evaluates arguments, fact-checks claims, and determines a winner based on evidence and logical consistency.

Unlike traditional tools that summarize content, ClaimCrush focuses on breaking information into discrete claims and verifying each individually, enabling deeper analysis and clearer understanding of what is true.

Marketing Technology News: Is the Traditional CDP Already Out of Date?

The platform is available on web, Android, and as a Chrome extension. Early traction includes strong engagement and adoption by content creators with large audiences.

ClaimCrush is also expanding into B2B custom fact-checking solutions, offering configurable source sets and workflows for organizations.
The company’s mission is to become the go-to platform for verifying information online and reducing misinformation.