Two cybersecurity startups focused on securing the growing use of artificial intelligence launched today with new funding rounds as interest in protecting enterprise endpoints and autonomous AI agents continues to grow.

Bold Security Ltd. and Onyx Security Inc. each revealed $40 million in funding as organizations deploy AI across their infrastructure and, in doing so, create new attack surfaces and operational risks that traditional security tools struggle to address.

Bold is developing a platform that transforms enterprise endpoints into active security agents powered by AI. The company’s approach shifts security processing directly onto devices such as laptops rather than relying on centralized cloud monitoring.

It’s focused on protecting the endpoint layer, long considered one of the most common entry points for cyberattacks. Differing from other solutions, Bold says, its system runs AI models locally on devices to analyze user behavior, application usage and data interactions in real time to detect anomalies and intervene before a threat spreads.

The company argues that existing endpoint security tools rely heavily on cloud analysis and siloed monitoring systems that can generate excessive alerts while failing to understand user context. By running models locally, Bold is aiming to provide faster response times, reduce latency and keep sensitive data from being transmitted externally.

The funding round was led by Red Dot Capital Partners Pte. Ltd., with Bessemer Venture Partners LP and Picture Capital also participating. Bold plans to use the funding to expand its go-to-market strategy and scale its technology globally.

Onyx Security, on the other hand, is tackling a different but rapidly emerging security challenge around agents: the governance and control of autonomous AI agents operating inside enterprise systems.

The company is building what it calls a secure “AI control plane” that is designed to monitor, manage and govern how autonomous AI agents interact with corporate systems and data.

The problem Onyx is taking on is a real one: As organizations increasingly deploy AI agents across engineering, operations and customer service workflows, those systems are gaining direct access to internal platforms and business processes. Onyx argues that this shift creates new risks ranging from reasoning errors and hallucinations to prompt-based attacks that manipulate agent behavior.

The Onyx platform discovers AI agents operating within an organization, tracks their decision-making processes and approves, blocks, or modifies actions when necessary to enforce security and governance policies.

Onyx’s funding, consisting of a $5 million seed round and $35 million Series A, came from Conviction Capital and Cyberstarts.

Image: Onyx