BitcoinWorld AI Safety: Datumo Secures $15.5M to Revolutionize AI Evaluation for Enterprises In the rapidly evolving landscape of artificial intelligence, particularly with the rise of generative AI , the conversation around safety and responsible deployment has become paramount. For those in the cryptocurrency space, understanding the underlying technology and its ethical implications is just as crucial as tracking market trends. A recent McKinsey report highlighted a significant challenge: most organizations feel unprepared to use generative AI safely. A core concern is AI explainability – the ability to understand how and why an AI makes certain decisions. While 40% of respondents view this as a major risk, only a fraction are actively addressing it. This is where companies like Seoul-based Datumo are stepping in, aiming to build a more trustworthy AI ecosystem. Why is AI Safety a Growing Imperative? The urgency for robust AI safety measures stems directly from the complex and often opaque nature of advanced AI models. As these models become more integrated into critical business operations and daily life, the potential for biased, incorrect, or even harmful outputs increases. Explainability is not just a technical hurdle; it’s a foundational requirement for trust and accountability. If businesses cannot understand their AI, they cannot truly control it or mitigate its risks. Datumo, originally an AI data labeling company, recognized this critical gap and strategically pivoted to address it head-on, focusing on tools that enable comprehensive testing, monitoring, and improvement of AI models without requiring deep technical expertise from the end-user. How is Datumo AI Reshaping Model Evaluation? Datumo’s journey began with a practical frustration: the time-consuming nature of traditional data labeling. David Kim, CEO of Datumo and a former AI researcher, envisioned a novel approach – a reward-based app allowing anyone to label data in their spare time. This innovative concept, validated at a KAIST startup competition, laid the groundwork for what would become a formidable player in the AI space. Co-founded in 2018 alongside five KAIST alumni, Datumo (formerly SelectStar) quickly secured pre-contract sales and surpassed $1 million in revenue in its first year. Today, Datumo boasts an impressive client roster including major Korean conglomerates like Samsung, LG, Hyundai, Naver, and SK Telecom. However, as clients’ needs evolved, Datumo’s mission expanded beyond simple data labeling. The Evolution from Data Labeling to Full-Stack AI Evaluation Several years ago, Datumo’s clients began requesting more than just data annotation. They sought assistance in scoring AI model outputs and comparing them against alternatives. As Michael Hwang, co-founder of Datumo, explained, this was a pivotal moment: they realized they were already performing AI model evaluation without explicitly knowing it. This insight prompted Datumo to double down on this critical area, leading to the release of Korea’s first benchmark dataset specifically focused on AI trust and safety. CEO David Kim further elaborated on this strategic expansion, noting, “We started in data annotation, then expanded into pretraining datasets and evaluation as the LLM ecosystem matured.” This evolution underscores Datumo’s adaptability and foresight in anticipating the needs of the burgeoning AI market. Understanding the Nuances of AI Evaluation and Market Dynamics The importance of robust AI evaluation is underscored by significant industry movements. Meta’s substantial investment in data-labeling giant Scale AI highlights the intense competition for high-quality AI training data. Interestingly, shortly after this deal, OpenAI, a Meta competitor, reportedly ceased using Scale AI’s services, signaling a dynamic and competitive landscape. Datumo shares common ground with companies like Scale AI in pretraining dataset provisioning, and with Galileo and Arize AI in AI evaluation and monitoring. However, Datumo distinguishes itself through its unique licensed datasets, particularly data crawled from published books, which CEO Kim notes offer rich, structured human reasoning but are notoriously challenging to clean. Furthermore, Datumo offers a full-stack evaluation platform called Datumo Eval. This innovative platform automatically generates test data and evaluations, checking for unsafe, biased, or incorrect responses without the need for manual scripting. Crucially, Datumo Eval is a no-code tool, making advanced AI evaluation accessible to non-developers on policy, trust and safety, and compliance teams. Strategic Funding: Fueling the Future of Generative AI Trust Datumo’s recent achievement of raising $15.5 million, bringing its total funding to approximately $28 million, is a testament to its compelling vision and market potential. This round saw participation from prominent investors including Salesforce Ventures, KB Investment, ACVC Partners, and SBI Investment. The story behind attracting Salesforce Ventures is particularly insightful. CEO David Kim’s decision to share a fireside chat with Andrew Ng, founder of DeepLearning.AI, on LinkedIn caught the attention of Salesforce Ventures. This organic outreach led to a series of meetings and a soft commitment, culminating in an eight-month funding process. This significant capital injection will accelerate Datumo’s R&D efforts, especially in developing automated evaluation tools for enterprise AI. It will also be instrumental in scaling global go-to-market operations across South Korea, Japan, and the U.S. With 150 employees in Seoul, Datumo has already established a presence in Silicon Valley since March, signaling its global ambitions. The Path Forward: A Safer, More Explainable AI Future Datumo’s strategic evolution from a data labeling specialist to a leader in AI evaluation and safety positions it at the forefront of a critical industry need. By providing accessible, no-code tools and unique datasets, Datumo is empowering businesses to build and deploy AI models with greater confidence and responsibility. Their focus on explainability and trust directly addresses the major concerns highlighted by industry reports, paving the way for more widespread and ethical adoption of artificial intelligence. As AI continues to permeate every aspect of technology and business, Datumo’s contributions will be vital in ensuring that innovation is matched by an unwavering commitment to safety and accountability. To learn more about the latest AI market trends, explore our article on key developments shaping AI models’ features and institutional adoption. This post AI Safety: Datumo Secures $15.5M to Revolutionize AI Evaluation for Enterprises first appeared on BitcoinWorld and is written by Editorial Team
