The previous security of Google collects $ 13 million to fight e -mail threats before you reach you


like Amnesty International is increasingly assisting infiltrators To launch mass email attacks, former Google Security leaders have joined the creation of independent artificial intelligence agents who aim to stop hunting, harmful programs and e -mail to concessions for email threats before they reached the user’s inbox.

This is the task behind AjisayThe start of the new email security, which has just emerged from the ghost with $ 13 million in the financing of joint seeds by Access and Foundation Capital.

more than 90 % of successful electronic attacks start With an e -mail message, for all CISA. Modern crowds Ticket (PDF) It was also found that the fishing messages created by large language models (LLMS) had a 54 % click average in 2024, which is much higher than 12 % for human written email messages.

Aegisai aims to face this increasing threat through a group of independent artificial intelligence agents.

CY KHORMAEE and Ryan Luo, which was founded by Google Safe Browsing and Recaptcha, Cy Khormae and Ryan Luo, was established, a real -time network of artificial intelligence agents who analyze and neutralize e -mail threats independently, without relying on any specific set of rules. This approach challenges the typical e -mail security platforms that depend on fixed rules and often require intensive training for users.

“The total of all evil is a pdf attachment in an email. This is where all attacks have always started, so I really wanted to solve this problem,” Khormie said in an exclusive interview with Techcrunch.

A picture of the founders of Ajisay Ryan Low (left) and Say Khormi (right)
The founder of Aegisai, participants, Ryan Low (left) and Say Khormi (right)

KHORMAE was the head of the product and the director of Google Products Management for more than five years until July 2023. During that time, the security team responsible for protecting Google, its four billion users, four million webs of deception, harmful programs and fraud, using products such as safe browsing, web risk, and web risk. During this time, Loo first met, who spent nearly a decade in Google and was part of the safe browsing team.

He told Techcrunch, that Google gave experience in KHORMAE to build hunting detection techniques, a deep understanding of security from the company’s perspective, and how to develop and expand security work quickly.

Before Google, KHORMAE CONNCY established a sales intelligence platform, which was It was obtained by Sugarcrm in 2016. Later he held the position of Vice President for Products Management for more than a year and a half until November 2024, before Aegisai began.

Aegisai has built thinking factors, each of which is LLM specifically designed for a specific threat. Once the coordination worker recognizes a potential threat or threat, he calls on other agents in the network, which Khormae refers to as “Friends”. Then these agents operate the analysis, the mind with each other, and they respond to the coordination factor with the ruling.

The agents analyze in the actual time of each message component, including links, attachments, descriptive data, QR symbols and behavioral patterns.

A screenshot shows the AEGISAI information panel, showing the number of users and the banned email messages.
Aegisai dashboardImage credits:Ajisay

“What we know of building these tools in Google is what all the things you need to analyze around? Khormyi said.

While Aegisai has currently built more than 10 agents of this work, KHORMAE TECHRUNCH told that there could be 50 to 100 agents over time as opponents have become more intelligent and try to deceive the system.

He said: “I think completely that within two years, the opponents will understand what we do. It will re -prepare what we do, and then we will need to build more agents to stay in front of them.”

Unlike the typical email safety platform that uses a bases -based approach, artificial intelligence agents discover a set of attacks and run out to every possible variable for these actual time attacks, Jormie said. The startup company has developed multiple Amnesty International models designed for various threats and specific industries, including those in investment capital and financial services.

Besides discovering threats quickly, AEGISAI agents help reduce the wrong positives by up to 90 % compared to traditional solutions, as you claim to start.

It takes “no more than five minutes” for customers to install the AEGISAI system on Google or Microsoft Email 365 via API, for all KHORMAEE. Once prepared, the start starting will send a report within two days with the details about what is in the environment, including wrong positives and wrong negatives. It will be operated after that in reading mode for a week and then stimulating the quarantine.

“It is very difficult without this technology to solve this heterogeneous problem in the email,” Khormey said.

The startup company, along with offices in San Francisco and New York, is currently a pilot with customers in the United States and Europe and has already added three paid customers, including Lokker and Crypto Platform Connect. Upon starting, a team of six members is currently starting.

Through the new investment, Khormyi said that the start starting is planning to expand its technical experience and build a strong infrastructure to go to the market.

Leave a Reply

Your email address will not be published. Required fields are marked *