Glow, a cybersecurity startup based by former Meta and Snowflake executives, emerged from stealth as a unicorn, betting that synthetic intelligence is reshaping how enterprises safe worker gadgets.
The Palo Alto-headquartered startup on Wednesday mentioned it raised $180 million in an all-equity Sequence A funding spherical that valued it at $1.2 billion, with backing from Sequoia Capital, Cyberstarts, Greenoaks, and Redpoint Ventures, alongside participation from Index Ventures, Swish Ventures, Lux Capital, Operator Collective, and Holly Ventures. The funding made Glow one of many newest cybersecurity startups to attain unicorn standing earlier than publicly disclosing income metrics.
As extra enterprises deploy AI instruments and attackers more and more use generative AI to automate phishing, develop malware, and launch extra subtle cyberattacks, firms are rethinking how they safe endpoints — from worker laptops to servers and different linked gadgets. Issues have intensified since Anthropic unveiled its Mythos AI model, which the corporate mentioned demonstrated superior capabilities in figuring out and exploiting software program vulnerabilities, prompting broader debate over AI-assisted cyberattacks. Glow is betting that this shift requires a brand new method to endpoint safety.
Based in 2025, Glow is constructing an endpoint safety platform that helps enterprises monitor and management the software program, AI brokers, and developer instruments working on worker gadgets. The startup mentioned the platform makes use of specialised AI brokers to repeatedly map enterprise environments, assess threat in actual time, and implement safety insurance policies.
“If you happen to consider the previous decade, all the pieces was transferring to the cloud and SaaS. All of the sudden, AI lands on the endpoint in a approach we’ve by no means seen,” co-founder and chief govt Roi Tiger (pictured above, middle) mentioned in an interview.
Tiger, a former Meta vice chairman of engineering, co-founded Glow alongside former Snowflake cybersecurity technique head Omer Singer (pictured above, left), former Claroty vice chairman of analysis and growth Ophir Arie (pictured above, proper), and former Meta engineering chief Arnon Joseph. The startup’s management group additionally consists of chief working officer Emily Heath, a former chief data safety officer at United Airways and DocuSign who served on the board of Wiz by means of its $32 billion acquisition by Google and was beforehand a accomplice at Cyberstarts.
Although it has solely simply emerged from stealth, Glow mentioned it already has paying prospects throughout industries together with healthcare, retail, and monetary providers. Nonetheless, it declined to reveal buyer names and numbers. The startup’s typical deployments, Tiger mentioned, span tens of 1000’s of worker gadgets throughout international organizations.
To energy the platform, Glow makes use of AI fashions from Anthropic and Google’s Gemini by means of Amazon Bedrock, whereas constructing its personal software program to offer the fashions with enterprise context and enhance their reliability for safety duties, Tiger informed TechCrunch.
Glow’s platform, Tiger mentioned, has already prevented malicious npm packages, third-party software program elements used to construct purposes, from being put in in buyer environments, recognized AI brokers trying to tug in such software program, and detected worker gadgets the place endpoint detection and response instruments had been lacking or working with diminished performance.
Glow enters a crowded endpoint safety market dominated by firms together with CrowdStrike, Microsoft, SentinelOne, and Palo Alto Networks. Tiger mentioned current endpoint detection and response merchandise focus totally on detecting threats after they emerge, whereas Glow is designed to stop dangerous software program, AI brokers, and developer instruments from getting into enterprise environments within the first place.
The startup employs practically 100 folks, about 70% of whom are in Israel and the rest within the U.S. Whether or not AI-native endpoint safety platforms change into a definite class stays to be seen, as enterprises are solely starting to grapple with the safety implications of more and more succesful AI fashions.
If you buy by means of hyperlinks in our articles, we may earn a small commission. This doesn’t have an effect on our editorial independence.
