A $1.2 Billion Startup Thinks AI Needs to Guard the Devices It Now Lives On

Key points

• Glow, a cybersecurity startup founded in 2025, raised $180 million in a Series A funding round in 2025 and is valued at $1.2 billion.
• Investors include Sequoia Capital, Cyberstarts, Greenoaks, and Redpoint Ventures, among others.
• The company uses AI models from Anthropic and Google to monitor and control software running on employee devices.
• Glow already has paying customers in healthcare, retail, and financial services, though it has not disclosed names or numbers.
• The startup employs nearly 100 people, roughly 70 percent of them based in Israel.

A new cybersecurity company called Glow stepped out of hiding on Wednesday, and it arrived with a striking number already attached: a $1.2 billion valuation before it had ever told the public it existed.

The Palo Alto startup raised $180 million in what is called a Series A round, meaning the first major outside investment a company takes in. Sequoia Capital, Cyberstarts, Greenoaks, and Redpoint Ventures led the deal, with Index Ventures and several smaller firms also joining.

So what does Glow actually do?

It watches the software running on employee laptops and other work devices, devices collectively called endpoints in security language. When an AI tool, a developer add-on, or a suspicious software package tries to install itself, Glow is supposed to catch it before it lands, not after the damage is done.

That distinction matters. Today's best-known endpoint security tools, from companies like CrowdStrike and Microsoft, are built mainly to detect threats that have already slipped inside. Glow's co-founder and chief executive Roi Tiger, a former vice president of engineering at Meta, says his platform is designed to stop risky software from entering in the first place.

"If you think of the past decade, everything was moving to the cloud," Tiger told TechCrunch. "Suddenly, AI lands on the endpoint in a way we've never seen."

He has a point. Employees now run AI coding assistants, AI writing tools, and automated AI agents, software programs that can carry out multi-step tasks on their own, directly on their work machines. Each one is a potential door that attackers could push open.

To power its monitoring, Glow uses large language models, the AI technology behind tools like ChatGPT and Claude, from Anthropic and Google's Gemini, accessed through Amazon's cloud service. It builds its own software on top to give those models the specific context they need to make accurate security calls inside a corporate network.

In practice, Tiger says the platform has already blocked malicious npm packages (small, reusable chunks of code that developers download to build apps) from reaching customer systems, and caught AI agents trying to pull in unsafe software.

Glow was co-founded by Tiger alongside Omer Singer, formerly head of cybersecurity strategy at Snowflake, Ophir Arie from security firm Claroty, and Arnon Joseph, also a Meta engineering alumnus.

What does this mean for ordinary employees?

For most workers, nothing changes today. Glow sells to businesses, not individuals. But if your employer deploys it, a quiet layer of AI will be watching what software loads onto your work laptop, aiming to stop attacks before your IT team even gets an alert. That is the pitch, anyway. The company has paying customers but has not yet shown public revenue figures to back up its billion-dollar price tag.