TL;DR:
- Cambridge-based startup Unitary AI has secured a $15 million funding round led by Creandum, with participation from Paladin Capital Group and Plural.
- The company specializes in content moderation, with a focus on video content.
- Unitary AI employs a unique “multimodal” approach, combining text, sound, and visual analysis to effectively identify harmful content.
- The startup has seen substantial growth, with its daily video classification increasing from 2 million to 6 million, covering billions of images.
- It is expanding language support beyond English and plans to use the funding to expand into new regions and hire more talent.
- The content moderation market is highly competitive, with companies like OpenAI, Microsoft, and Amazon also vying for dominance.
Main AI News:
In the ever-evolving landscape of online media, content moderation remains a hot-button issue, with new regulations and growing public concern. Addressing this challenge is becoming increasingly complex, thanks to the rise of weaponized AI and other technological advancements. Cambridge-based startup Unitary AI is confident that it has found a superior approach to tackling content moderation, especially in the intricate realm of video content.
Today, Unitary announces a substantial $15 million funding round to capitalize on its recent market momentum. The Series A funding is spearheaded by prominent European venture capital firm Creandum, with participation from Paladin Capital Group and Plural. Unitary has witnessed a significant increase in video classification, now processing 6 million videos daily, up from 2 million, covering billions of images. Additionally, the platform is expanding language support beyond English. Although customer names remain undisclosed, Unitary reports annual recurring revenue (ARR) in the millions.
This funding infusion is enabling Unitary to expand into new regions and recruit top talent. While the startup has not disclosed its current valuation, it previously secured under $2 million in funding and an additional $8 million in seed funding, with notable investors including Carolyn Everson, a former executive at Meta.
Numerous startups in recent years have harnessed various aspects of artificial intelligence to develop content moderation tools. The sheer scale of the video content challenge is an ideal arena for AI solutions. The volume of data generated and shared on platforms such as YouTube, Facebook, Reddit, TikTok, dating sites, gaming platforms, and videoconferencing tools constitutes over 80% of all online traffic and is virtually impossible to handle manually.
Investors recognize the critical need for technology-driven solutions in an online world. Christopher Steed, Chief Investment Officer of Paladin Capital Group, states, “In an online world, there’s an immense need for a technology-driven approach to identify harmful content.“
However, the field of content moderation is highly competitive, with companies like OpenAI, Microsoft, Hive, Active Fence, Spectrum Labs, Oterlu (now part of Reddit), Sentropy (now a part of Discord), and Amazon’s Rekognition vying for a piece of the pie.
Unitary AI asserts that existing tools are less effective in the video domain because they primarily focus on analyzing one data type, such as text, audio, or images, but not all simultaneously. This approach leads to a high rate of false positives or, conversely, a failure to flag problematic content.
Unitary’s CEO, Sasha Haco, explains, “What is innovative about Unitary is that we have genuine multimodal models. Rather than analyzing just a series of frames, to understand the nuance and whether a video is, for example, artistic or violent, you need to be able to simulate the way a human moderator watches the video. We do that by analyzing text, sound, and visuals.”
Customers can customize parameters for content moderation and often use Unitary in conjunction with human moderation teams, reducing their workload and stress.
Haco addresses the question of why multimodal moderation hasn’t been done before, stating that visual-only models have been sufficient in the past. However, Unitary believes there is untapped potential in this approach.
The challenges of content moderation continue to plague social platforms, gaming companies, and other digital channels where user-generated media is shared. Recently, some social media companies have signaled a shift away from stricter moderation policies, fact-checking organizations have lost momentum, and questions linger about the ethics of moderation in handling harmful content.
Haco brings an interesting background to the table, having worked on black hole research with Stephen Hawking before founding Unitary AI. Her shift from studying spacetime phenomena to addressing complex real-world problems highlights the urgency of the content moderation issue.
James Thewlis, Unitary’s CTO, also contributes his expertise in computer vision and visual understanding with less manual annotation, a valuable asset to the company’s mission.
Unitary, as a name, not only symbolizes the startup’s goal of unifying diverse parameters for video understanding but also references Haco’s previous career in quantum physics, where unitary operators describe unpredictable quantum states—much like the unpredictable nature of online content and human behavior.
While multimodal research in AI has been ongoing for years, Unitary AI appears to be at the forefront of applying this concept in practical applications. Meta’s recent mention of multimodal AI in its Connect keynote further emphasizes the growing significance of this approach.
Gemma Bloemen, Principal at Creandum and board member, expresses her enthusiasm for Unitary AI, stating, “Unitary has emerged as clear early leaders in the important AI field of content safety, and we’re so excited to back this exceptional team as they continue to accelerate and innovate in content classification technology.“
Ian Hogarth, a partner at Plural and also a board member, highlights Unitary’s remarkable achievements, saying, “From the start, Unitary had some of the most powerful AI for classifying harmful content. Already this year, the company has accelerated to 7 figures of ARR, almost unheard of at this early stage in the journey.”
Conclusion:
Unitary AI’s successful funding round and its innovative multimodal approach to content moderation demonstrate a growing demand for advanced solutions in the market. As the challenges of online content moderation persist, companies that can effectively address these issues, especially in the realm of video, are likely to play a significant role in shaping the future of online media safety.
