Responsible AI startup Fiddler secured $18.6 million in a Series B Prime funding round at the end of 2024. Starting the new year with a total investment of $68.6 million, the startup aims to help enterprise teams deliver responsible AI applications and ensure that people interacting with AI receive fair, safe, and trustworthy responses.
A pioneer in “AI Observability” and “AI Safety,” Fiddler’s investment was backed by Cisco Investments, Capgemini’s ISAI Cap Ventures, Dallas Venture Capital, Alpha Square Group, Samsung Next, and Essence Fund along with participation from existing partners Insight Partners, Lightspeed Venture Partners, Lux Capital, and Dentsu for the recent funding round.
READ: Gumloop secures $17 million in Series A funding to make AI automation (January 13, 2025)
“This funding will allow us to continue supporting enterprises in scaling AI responsibly, ensuring safety, trust, and performance at every step. Together, we’re building a future where AI works for everyone. Grateful for the support and excited for what’s next,” co-founder and CEO Krishna Gade wrote on X.
Fiddler enables the delivery of high-performance AI by allowing companies to launch and update models faster, leading to improved decision-making and revenue growth. It claims to reduce operational costs by accelerating the launch of AI applications, improving efficiency, identifying model issues, and minimizing downtime boosts productivity.
Headquartered in Palo Alto, California, the startup aims to ensure responsible AI governance by minimizing reputational risks, reducing bias, and improving customer satisfaction, leading to higher Net Promoter Scores (NPS).
When GenAI emerged, companies rapidly began prompt engineering, fine-tuning, testing, and evaluating large language models (LLMs). This gave rise to numerous GenAI vendors specializing in pre-production testing and evaluation. However, LLM testing remains stuck in pre-production, neglecting the most challenging part: monitoring LLMs in real-world environments, the company told The American Bazaar in a statement.
“We support enterprises at every stage of their AI journey in establishing long-term responsible AI practices. From accelerating [large language model operations] LLMOps and [machine learning operations] MLOps, driving business value, and mitigating risks to building customer satisfaction, we help establish responsible AI practices across all products as companies mature in their AI journey,” Fiddler’s team noted.
Without oversight in production, enterprises risk hallucinations, safety concerns, and privacy breaches are harming users and the companies involved. These risks limit their ability to explore AI innovation, keeping LLMs confined to testing environments. The Fiddler Trust Service and its Guardrails solution help enterprises alleviate this issue, the company claimed.
In 2024, the startup launched their Fiddler Trust Service, the industry’s first solution for scoring and monitoring LLM applications. The Fiddler Trust Service, powered by their proprietary, fine-tuned Fiddler Trust Models, is designed for task-specific, high accuracy scoring of LLM prompts and responses with low latency.
“Today, we’re excited to launch the Realtime Guardrails API for Gen AI applications. It’s powered by our Fiddler Trust Service, a groundbreaking platform built on a family of specialized small language models. These models detect and block malicious inputs like jailbreak attempts, measure response accuracy, and ensure LLM outputs are safe and reliable,” Gade announced Wednesday on X.
In addition to its proprietary API, Fiddler also partnered with Nvidia NIM (NVIDIA Inference Microservices) to safeguard applications in AI agents, co-founder Amit Paka shared on LinkedIn on Jan. 16.
Fiddler touts three key areas as its strength: monitoring, security, and governance. By monitoring a comprehensive set of LLM and ML metrics such as performance, accuracy, drift, personal identity information (PII) leakage, hallucination, safety, and custom metrics, Fiddler can alert teams for immediate action using explainability and diagnostics for root cause analysis. The platform also provides audit evidence and audit trail for regulatory and compliance reporting.
Founded in 2018, the startup boasts serving Fortune 500 organizations across pre-production and production to deliver high performance and responsible AI.
Prior to founding Fiddler, Gade led the Newsfeed Explainable AI Platform at Meta, previously known as Facebook. This platform aimed to increase transparency into Facebook newsfeed algorithms, and explain to Facebook users why they are seeing the particular posts they did on their newsfeed by implementing the “Why You’re Seeing This Post” feature.
READ: Three Indian startups selected for Walmart pilot programs (January 22, 2025)
The work and contributions of this platform was also shown to the White House to help explain the reasoning behind controversial posts that were made during the 2016 elections, Fiddler’s team told The American Bazaar.
In addition to shining a light on ML models, the move also underlined the sometimes-ill effects of personalized recommendations that pose harm and risks to companies and users, and emphasized the need in research, development, and advocacy around Explainable AI and Responsible AI.
Gade’s experience in explaining AI behavior and this growing concern in the market inspired him to team up with co-founder Amit Paka and launch Fiddler AI.
“Fiddler’s mission is to make the world a better place, as many of the decisions consumers make every day are influenced by AI,” Fiddler’s team told The American Bazaar.