Dropzone AI
by Dropzone AI
AI SOC analyst agents that investigate every security alert without playbooks, with threat hunting in beta
Dropzone AI sells autonomous AI SOC agents for security operations teams and MSSPs. Its AI SOC Analyst investigates every alert across SIEM, EDR, identity, cloud and email tools without playbooks or code, then hands analysts an evidence-backed verdict, cutting the manual triage that buries most SOCs.
Dropzone AI, founded in 2020 by CEO Edward Wu, builds AI agents that do the triage and investigation work of a tier-one SOC analyst. The flagship AI SOC Analyst connects to a customer's existing tools through more than 100 integrations, including Splunk, CrowdStrike Falcon, Microsoft Defender for Endpoint, SentinelOne, Google Security Operations, Okta, Wiz, Amazon GuardDuty, Datadog and Cortex XSOAR, then investigates phishing, endpoint, network, cloud, identity and insider-threat alerts by querying those systems itself. The company says it needs no playbooks, code or prompts, no data migration or normalization, and no new endpoint agent; analysts steer it in natural language and the corrections persist. Each investigation produces a verdict with an auditable trail of every tool queried, and automated containment is limited to blocking malicious IPs and disabling compromised accounts on threats the agents have confirmed. The product line now includes an AI Threat Hunter (in beta), an AI Detection Engineer that drafts detections for analyst review, and an AI Threat Intel Analyst listed as arriving in 2027. Dropzone raised a $37 million Series B led by Theory Ventures in July 2025, with Madrona, Decibel, Pioneer Square Labs and In-Q-Tel, bringing total funding past $57 million. In January 2026 it reported 11x ARR growth for 2025 and more than 300 enterprises in production; its site now claims 400+ security teams. Customers named in coverage include UiPath, Zapier, Avalara, Kwik Trip and the MSSPs CBTS and ECS. It competes in the crowded AI SOC category with Torq, Exaforce and Prophet Security.
A SOC manager or CISO at a mid-size or large enterprise, or an MSSP, whose analysts cannot investigate every alert and who wants coverage without writing SOAR playbooks.
Every alert gets investigated with an evidence trail; Zapier reports 85% less manual investigation and ECS automated 70% of alerts.
At a Glance
- Category
- Governance & Security
- Pricing
- Contact for pricing, Subscription
- Target Market
- CISOs, SOC Managers, MSSPs, Federal Agencies
- Deployment
- Cloud-only
- Founded
- 2020
- Customers
- 400+ security teams in production (vendor claim, November 2026)
Key Features
- ✓AI SOC Analyst
Investigates every incoming alert across phishing, endpoint, network, cloud, identity and insider threats, so low-priority alerts stop going unread.
- ✓No playbooks or data migration
Queries existing SIEM, EDR, identity and email tools directly, which removes SOAR playbook authoring and log normalization from deployment.
- ✓Evidence-backed verdicts
Every investigation ships with an auditable trail of each tool queried, so analysts can check the reasoning before acting.
- ✓Scoped automated containment
Can block malicious IPs and disable compromised accounts, but only on threats the agents have confirmed, limiting blast radius from errors.
- ✓AI Threat Hunter (beta)
Runs hypothesis-driven hunts from prebuilt packs or analyst prompts; one customer cut a 40-hour hunt to about one hour.
- ✓AI Detection Engineer
Reviews and drafts detection rules, sending each one to analysts for approval before it is deployed to production.
- ✓Natural-language coaching
Analysts correct the agent in plain language and the correction persists, tuning investigations to local context without code.
Capabilities
Use Cases
- •Alert backlog elimination
A SOC routes all SIEM and EDR alerts to the AI SOC Analyst so every alert is investigated and analysts only review verdicts.
- •MSSP alert offload
A managed security provider like CBTS uses Dropzone to offload 30% to 50% of alert volume across customer tenants.
- •Phishing triage
Reported and detected phishing emails are investigated against email, identity and endpoint data, and confirmed compromises trigger account disablement.
- •Threat hunting at lower cost
Hunters hand hypotheses to the AI Threat Hunter, turning a multi-day manual hunt into roughly an hour of review.
Ideal For
Best For
- ✓Lean SOC teams that leave low-severity alerts uninvestigated and want full coverage without adding headcount
- ✓MSSPs that need per-tenant separation and per-agent capacity pricing across many customer environments
- ✓Security teams with an established SIEM and EDR stack who do not want to migrate or normalize data into a new platform
- ✓US federal agencies needing a FedRAMP High authorized AI SOC offering
Not Ideal For
- ✗Teams that want broad autonomous response, since containment is limited to blocking IPs and disabling accounts on confirmed threats
- ✗Buyers who need transparent list pricing or a hands-on free trial before a sales cycle; pricing is by request and the trial is a self-guided demo
- ✗Organizations wanting a mature threat-intelligence agent today, since the Threat Intel Analyst is listed as arriving in 2027 and the Threat Hunter is in beta
Deployment
Market & Ratings
400+ security teams in production (vendor claim, November 2026)
Market Analysis
Pros
- ✓Fast reported traction: 11x ARR growth in 2025 and net revenue retention over 370%, per the company
- ✓Broad integration coverage of 100+ security tools, so most SOC stacks connect without new data pipelines
- ✓Cloud Security Alliance study with 148 participants found 22-29% better accuracy and 45-61% faster investigations than manual analysis
- ✓Analysts keep ownership of verdicts, and containment actions are narrow by design
Cons
- ✗No public pricing, and absolute revenue was not disclosed behind the 11x growth claim
- ✗Several agents are not finished: the Threat Hunter is in beta and the Threat Intel Analyst is listed for 2027
- ✗SOC 2 and ISO 27001 status could not be confirmed; the trust center page rendered no compliance content
- ✗No independent user reviews or practitioner discussion surfaced on G2 or Hacker News, so most performance figures are vendor-reported
Pricing
Enterprise
Contact for pricing
- ✓Sized on employee count
- ✓All-inclusive agents
- ✓Unlimited investigations under fair use
- ✓Unlimited programmatic actions
- ✓GreyNoise and ReversingLabs threat feeds included
MSSP
Contact for pricing
- ✓Choose per agent and capacity level
- ✓Tenancy separation
Federal
Contact for pricing
- ✓Per agent and capacity
- ✓FedRAMP High (Class D)
No prices are published; every tier says request pricing. Enterprise is sized on company employee count rather than alert volume and includes unlimited investigations under fair use, unlimited programmatic actions and bundled GreyNoise and ReversingLabs feeds. MSSP and Federal tiers are priced per agent and capacity level. A self-guided demo is offered in place of a free trial.
Security & Compliance
Sources
This page was written from 7 sources, 4 on domains other than dropzone.ai.
- 1.dropzone.ai — dropzone.aivendor
- 2.dropzone.ai — pricingvendor
- 3.dropzone.ai — integrationsvendor
- 4.securityweek.com — dropzone ai raises 37 million for autonomous soc analyst
- 5.siliconangle.com — dropzone ai raises 37m expand ai soc analyst development int
- 6.techintelpro.com — dropzone ai 11x arr growth 37m series b in 2025
- 7.securitybrief.co.uk — dropzone ai hires leaders to drive emea apac push
Stay Ahead of the Curve
Weekly enterprise AI insights for technology leaders. No spam, no vendor pitches—unsubscribe anytime.
SubscribeRelated Products
Rubrik Agent Cloud
AI agent discovery, runtime policy and rollback from the Rubrik data security platform
Pretxt
Runtime AI agent supervision that enforces plain-language business policies in milliseconds, without an LLM judge
Portal26
AI adoption management platform for shadow AI discovery, AI governance and Claude usage control
Akeyless Agentic Runtime Authority
Intent-based runtime access control for AI agents: credentialless access, per-action policy checks and a session kill switch