The co-founders of EDR supplier Cybereason have regrouped with a brand new safety startup, 7AI, to assist organizations shift the burden of performing repetitive and routine safety duties at the moment carried out by human analysts onto AI. 7AI’s Agentic AI Platform frees safety professionals from time-consuming duties, corresponding to triaging alerts, decoding alerts, correlating telemetry, and trying to find recognized threats, says Lior Div, one of many co-founders.
Div and Yonatan Striem-Amit left Cybereason two years in the past after Softbank took a majority stake within the firm; they based 7AI in April 2024. The startup, which emerged from stealth on Thursday, says greater than a dozen corporations, principally massive and midsize enterprises, are already utilizing its Agentic AI Platform. 7AI additionally obtained $36 million in seed funding from Greylock Companions, Spark Capital, and CRV.
Div describes agentic AI as “swarms of AI brokers” able to autonomously taking over routine safety duties. Not like remoted generative AI brokers, these swarms can allow autonomous operations by pooling and speaking their intelligence to analyze and prioritize threats whereas optimizing system sources. A swarm of brokers working in tandem signifies that one agent might be configured to find suspicious telemetry in an endpoint detection and response (EDR) system whereas one other might be configured to validate the potential menace by correlating cloud logs. One more agent might be configured to look at consumer habits patterns in identification and entry administration (IAM) methods.
“As an alternative of spending their time on repetitive work to answer alerts, our early clients are capable of begin their work with full context, drastically fewer false positives, and the outcomes of full investigations,” Div defined in a weblog publish asserting the corporate’s new platform. The platform paperwork how every agent reached its conclusions and may be reviewed at any time by human analysts.
7AI’s agentic AI capabilities, which is hosted within the Amazon Internet Providers cloud, is constructed with generative AI instruments from Open AI and Anthropic.
“Relating to reasoning, we’re utilizing Open AI,” Div tells Darkish Studying. “However in relation to truly implementing and writing code, we’re utilizing Anthropic.”
A Alternative for SOAR?
The platform will not be designed to exchange safety directors and analysts however moderately permit them to take mundane duties off their plates to allow them to allocate their time to extra strategic capabilities.
“AI will take away 90% of the boring, toiling work,” Div says.
In addition to dealing with repetitive duties, 7AI’s platform is designed to correlate telemetry with out transferring knowledge into one other system. For instance, in a typical menace looking state of affairs, the info must be pushed right into a safety data and occasion administration (SIEM). As an alternative, 7AI correlates the data at its supply. The platform can even detect menace exercise and anomalies in IAM methods corresponding to Okta, Div says.
“We imagine our AI will meet the info the place the info was born,” he says. “You do not have to ship plenty of these items to the SIEM anymore.”
This might additionally scale back organizations’ reliance on managed safety and repair suppliers or managed detection and response suppliers, Div suggests.
“We do not suppose that you will want a SOAR upon getting our system as a result of it is going to determine on the fly what’s the proper playbook to run and what sort of investigation to conduct with out the necessity for human beings to specify it step-by-step,” Div says.