Insider Brief
- White Circle, an AI monitoring and safety platform, has raised $11 million in seed funding from a roster of prominent industry angels including executives from OpenAI, Anthropic, Mistral, Hugging Face, and Datadog, with capital earmarked for product development and team expansion across the US, UK, and Europe.
- Founded by Denis Shilov, who went viral in 2024 after bypassing the safety filters of every major AI model with a single prompt, the platform monitors AI inputs and outputs in real time, detecting harmful content, hallucinations, prompt injection attacks, and model drift through a single API with support for 150 languages.
- White Circle has also published notable research including CircleGuardBench, a benchmark for AI moderation performance, and the KillBench study, which ran over one million experiments across 15 AI models to expose hidden biases related to nationality, religion, and body type that intensified under standard production integration methods.
PRESS RELEASE — White Circle — the platform trusted by global brands to keep their AI models in check — has raised $11m from some of the biggest names in the industry including Romain Huet (OpenAI); Dirk Kingma (ex-OpenAI, now Anthropic); Guillaume Lample (Mistral); Thomas Wolf (Hugging Face); Olivier Pomel (Datadog); François Chollet (Keras); Mehdi Ghissassi (ex-DeepMind); Paige Bailey (DeepMind); and David Cramer (Sentry). The Seed funding will be used to further accelerate product development, expand its team across the US, UK and Europe and grow White Circle’s global customer base.
As firms look to capitalise on the AI boom; with vibe coding letting anyone ship AI products in an instant; and with increased reports of AI models breaking free from their safety rails, White Circle is the only all-in-one, enterprise-grade platform that tests, protects, observes and improves AI in real time.
It was founded by engineer Denis Shilov who went viral in 2024 when he bypassed the safety filters of every major AI model with a single prompt. Shilov’s universal jailbreak meant he could unlock instructions on how to make drugs and weapons, access dangerous or illegal information and extract anything that models including ChatGPT, Claude and more were explicitly built to refuse.
After hitting 1.4M views, the post drew attention from Anthropic, OpenAI and Hugging Face, leading to Shilov being invited to join Anthropic’s bug bounty program and later building the White Circle platform to meet growing demand from AI providers.
A single source of truth and trust
White Circle’s proprietary models monitor AI inputs and outputs in real time. Based on each company’s custom policies, they detect harmful content, catch hallucinations, prevent prompt injection attacks, flag model drift and identify abusive or malicious users. The data from these scans helps teams understand how their models perform in different use cases, making it easier to choose the right models and improve them over time. Everything runs through a single API.
In practice, White Circle can detect a fintech model leaking sensitive customer data, block an AI agent from executing malicious instructions like deleting files or disrupting systems, flag and ban users attempting to abuse the system, and spot rising user frustration when a model starts to fail. Teams can define custom policies to control exactly what is allowed based on their needs, including automated enforcement actions like rate limiting or banning bad actors. The models also improve over time by learning from labelled user feedback, becoming more accurate and better adapted to each use case. Support for 150 languages ensures it works reliably across global products.
The platform itself is led by Head of Design Elena Iumagulova. White Circle is also SOC 2 Type I and Type II certified and HIPAA compliant.
“AI is moving faster than our ability to guide it,” said Denis Shilov, founder and CEO of White Circle. “We already trust it with decisions that touch millions of people — from hiring to healthcare, finance and security — and with the rise of vibe coding, anyone can ship an AI product without knowing what that model is actually doing once it’s live. Until now there’s not been a platform purpose-built to monitor AI’s behaviour, catch it when it goes wrong, or shape how it acts. With White Circle, we’re finally giving companies everything they need to hold their AI accountable and optimise their models in a single place, without sacrificing security, compliance or risk.”
Elena Iumagulova, Head of Design at White Circle said: “We wanted any team, technical or not, to be able to see exactly what their AI is doing in one place — whether they’re getting visibility into their models for the first time or running AI at scale and need to catch problems fast. Everything from real-time flagging to analytics and optimisation lives in one place, connected via a single API.”
Keeping AI accountable
As part of its mission to close AI’s accountability gaps, White Circle’s research continues to shape how the industry understands model risk. In May 2025, the company published CircleGuardBench, a benchmark that tests how well AI moderation models actually perform under real-world conditions. This benchmark lets anyone search by model and see their performance.
More recently, the company published its KillBench study in which it ran more than a million experiments across 15 AI models, including those from OpenAI, Google, Anthropic, and xAI, to uncover hidden biases. When forced to make decisions about human lives, all models showed preferences depending on nationality, religion, body type, and even phone brand. The research also found that structured output, the standard integration method for production AI, repeatedly caused refusal rates to collapse and biases to amplify.
“Denis and the White Circle team have an unusual combination of deep technical credibility and a clear commercial instinct,” said Ophelia Cai, Partner at Tiny VC. “The KillBench research alone shows what’s possible when you approach AI safety empirically rather than ideologically and the team is building the infrastructure the industry genuinely needs.”
ENDS
About White Circle
White Circle is an AI control platform that helps companies monitor, protect, and improve AI models in production. Its platform checks inputs and outputs against custom policies, detects hallucinations and prompt injection, and provides real-time analytics on model behaviour. White Circle is SOC 2 and HIPAA compliant and supports more than 150 languages. Visit whitecircle.ai
Contacts
Aimee Soulsby
aimee@burlington.cc
