Need smarter insights in your inbox? Join our weekly newsletters to get solely what issues to enterprise AI, knowledge, and safety leaders. Subscribe Now
Prophet Safety, a startup growing autonomous synthetic intelligence methods for cybersecurity protection, introduced Tuesday it has raised $30 million in Collection A funding to speed up what its founders describe as a basic shift from human-versus-human to “agent-versus-agent” warfare in cybersecurity.
The Menlo Park-based firm’s funding spherical, led by enterprise capital agency Accel with participation from Bain Capital Ventures, comes as organizations wrestle with an amazing quantity of safety alerts whereas subtle attackers more and more leverage AI to scale and automate their operations. Prophet’s method represents a marked departure from the “copilot” AI instruments which have dominated the market, as a substitute deploying absolutely autonomous brokers that may examine and reply to threats with out human intervention.
“Each safety operations crew is confronted with a twin mandate of decreasing threat whereas driving operational effectivity,” stated Kamal Shah, Prophet Safety’s co-founder and CEO, in an unique interview with VentureBeat. “Our Agentic AI SOC Platform addresses each challenges by automating guide, repetitive duties in safety operations with velocity, accuracy and explainability.”
The funding announcement coincides with Prophet’s launch of what it calls the business’s most complete Agentic AI SOC Platform, increasing past its preliminary Prophet AI SOC Analyst to incorporate Prophet AI Risk Hunter and Prophet AI Detection Advisor. The platform represents a big evolution from conventional Safety Operations Heart (SOC) automation instruments, which generally depend on inflexible, pre-programmed playbooks.
The AI Impression Collection Returns to San Francisco – August 5
The subsequent section of AI is right here – are you prepared? Be part of leaders from Block, GSK, and SAP for an unique take a look at how autonomous brokers are reshaping enterprise workflows – from real-time decision-making to end-to-end automation.
Safe your spot now – house is restricted: https://bit.ly/3GuuPLF
Safety groups drowning in 960 each day alerts face unprecedented capability disaster
The cybersecurity business faces a disaster of capability and functionality. Shah, who beforehand served as CEO of container safety firm StackRox earlier than its acquisition by Crimson Hat, skilled these challenges firsthand. In accordance with his observations, organizations obtain a mean of 960 safety alerts each day, with as much as 40% going uninvestigated because of useful resource constraints.
“The primary criticism that I see from clients each single day is just too many alerts, too many false positives,” Shah defined. “If you consider the world that we reside in right this moment, on common, an organization will get 960 alerts a day from all the safety instruments that they’ve of their setting, and 40% of these alerts are ignored as a result of they simply don’t have the capability to go and examine all these alerts.”
The issue is compounded by a extreme scarcity of expert cybersecurity professionals. Shah factors to what he calls a crucial expertise hole, noting there are 5 million open positions in cybersecurity globally, making a state of affairs the place even organizations with funds to rent can not discover certified personnel.
Prophet’s answer straight addresses this capability crunch. Over the previous six months, the corporate’s AI SOC Analyst has carried out greater than 1 million autonomous investigations throughout its buyer base, saving an estimated 360,000 hours of investigation time whereas delivering 10 instances quicker response instances and decreasing false positives by 96%.
How autonomous AI brokers differ from reactive copilot methods remodeling cybersecurity
The excellence between Prophet’s “agentic” AI and the copilot fashions deployed by bigger cybersecurity distributors like CrowdStrike, Microsoft, and Sentinel One is prime to understanding the corporate’s worth proposition. Conventional copilot methods require human analysts to provoke queries and interpret responses, basically serving as subtle search interfaces for safety knowledge.
“Copilot is reactive,” Shah defined. “You’ve got an alert are available in and a safety analyst has to go and write questions, ask the query to say, hey, what does this imply? And it’s a must to know what inquiries to ask. The analyst continues to be within the loop for each single alert that is available in as a result of they’re interacting with it.”
In contrast, Prophet’s agentic AI proactively initiates investigations the second an alert is triggered, autonomously gathering proof, reasoning via the information, and reaching conclusions with out human intervention. The system paperwork each step of its investigation course of, creating an audit path that enables safety groups to know and confirm its reasoning.
“What Prophet AI is ready to do is instantly, as soon as an alert is triggered, it proactively goes and completes the investigation,” Shah stated. “Inside a matter of minutes, your investigation is full and it is aware of what inquiries to ask, and it’s been educated to behave like an professional analyst.”
Constructing enterprise belief via clear AI decision-making and knowledge safety
Prophet’s system leverages a number of frontier AI fashions, together with choices from OpenAI, Anthropic, and others, deciding on essentially the most acceptable mannequin for every particular process. The corporate has constructed what Shah describes as an “evals framework” to make sure accuracy, repeatability, and consistency whereas stopping AI hallucinations—a crucial concern in safety contexts the place false data can result in inappropriate responses.
“In safety, you’re in a belief constructing train with the safety groups, and for those who hallucinate, you’re going to lose belief they usually’re not going to make use of your product,” Shah emphasised. The corporate employs a retrieval-augmented era (RAG) structure mixed with rigorous analysis processes to take care of what Shah calls “a excessive bar for safety groups.”
Information privateness and safety symbolize paramount issues for Prophet’s enterprise clients. The corporate employs a single-tenant structure making certain buyer knowledge stays remoted, and maintains contractual agreements with AI mannequin suppliers stopping buyer knowledge from getting used to coach or fine-tune fashions.
Early clients report dramatic effectivity beneficial properties as AI handles 1000’s of safety alerts
Prophet’s buyer base consists of Docker, which supplied a testimonial for the funding announcement. Tushar Jain, Docker’s EVP of Engineering and Product, famous that “Prophet AI is already serving to streamline components of our safety workflow, and we’re simply getting began. With the current launch of Risk Hunter and rising integration with our methods, we see a transparent path to quicker response instances, lowered noise, and a extra centered safety crew.”
The corporate has additionally revealed case research demonstrating dramatic enhancements in SOC effectivity. Eric Wille, CISO at Cupboard Works, reported decreasing his crew’s alert quantity from 33,200 down to simply six alerts requiring human consideration, successfully permitting his small crew to function with the effectivity of a a lot bigger group.
“Prophet AI minimize our alert queue from 1000’s to dozens,” Wille stated in a video testimonial. “It’s a drive multiplier that removes investigation bottlenecks, improves analyst focus, and helps us reply to actual threats quicker.”
Rising cyber threats and evolving assault strategies drive demand for AI-powered protection
Prophet’s emergence happens towards a backdrop of quickly evolving cyber threats. CrowdStrike’s 2025 International Risk Report documented a 150% enhance in China-nexus cyber exercise and a 442% progress in voice phishing operations, whereas noting that 79% of detected threats have been malware-free, making them more durable to determine via conventional signature-based detection strategies.
The corporate’s method to integration throughout current safety instruments gives a key aggressive benefit. Fairly than requiring organizations to switch their present safety stack, Prophet integrates with current Safety Info and Occasion Administration (SIEM) methods, Endpoint Detection and Response (EDR) platforms, and different safety instruments.
“Should you’ve obtained to go get 5 – 6 totally different copilots to make use of inside your group, it’s going to be very complicated,” Shah defined. “What clients are telling us is that, hey, I would like an unbiased AI SOC platform that may assist me triage, examine and reply to alerts from all of my safety instruments, not only one or two.”
Accel’s preemptive funding indicators rising confidence in autonomous safety methods
Eric Wolford, Associate at Accel, emphasised the mixture of technical innovation and confirmed market traction that drove the funding determination. “What stood out to us about Prophet wasn’t simply the technical ambition, however the real-world traction: they’re delivering autonomy and velocity whereas displaying their work—a crucial differentiator in an business that runs on belief,” Wolford stated in a press release.
Accel’s cybersecurity funding portfolio consists of CrowdStrike, Tenable, and BlackPoint Cyber, offering the agency with deep experience in evaluating safety applied sciences. The preemptive nature of the funding spherical — Prophet was not actively searching for capital — underscores investor confidence within the firm’s trajectory.
The funding will primarily assist engineering growth and go-to-market acceleration as Prophet scales its platform capabilities. The corporate plans to proceed increasing its agentic AI platform, doubtlessly including new modules for added safety operations workflows.
Trade consultants predict widespread adoption of AI brokers will reshape cybersecurity panorama
Prophet’s success displays broader tendencies reshaping cybersecurity. Deloitte’s 2025 cybersecurity forecasts predict widespread adoption of agentic AI methods, with 40% of enormous enterprises anticipated to deploy such methods of their SOCs by 2025. The consulting agency characterizes this shift as transferring from “automation that follows directions to automation that thinks.”
The corporate’s “function elevation” philosophy — enhancing fairly than changing human analysts — addresses issues about AI displacing cybersecurity professionals. Shah emphasised that automation ought to free analysts from repetitive duties to give attention to higher-value safety work.
“This isn’t about eliminating jobs,” Shah stated. “It’s about making certain an analyst doesn’t must spend time triaging and investigating alerts, as a result of who needs to try this all day, daily? As an alternative, they’ll give attention to the 4% of points that really matter to a company. They’re advancing their careers and doing extra higher-order safety work.”
As cyber threats proceed evolving and incorporating AI capabilities, the arms race between attackers and defenders more and more depends on technological sophistication fairly than human capability alone. Prophet’s method suggests a future the place cybersecurity turns into primarily a contest between AI methods, with human experience centered on strategic oversight and complicated decision-making.
The corporate’s skill to show measurable enhancements in SOC effectivity whereas sustaining transparency and explainability positions it to seize market share as organizations grapple with the twin pressures of accelerating threats and chronic expertise shortages. With the brand new funding, Prophet Safety goals to speed up this transition, doubtlessly setting the usual for the way organizations defend towards AI-powered assaults in an period the place the velocity and scale of threats exceed human capability to reply manually.
However maybe essentially the most telling indicator of this shift isn’t Prophet’s expertise or funding — it’s what occurred when Shah’s crew wasn’t actively searching for funding. Accel approached them anyway, recognizing that in a world the place attackers launch AI-powered assaults at machine velocity, the outdated playbook of human-driven protection isn’t simply inadequate — it’s out of date.
