Lakera
Lakera is a Zurich-based AI security company that provides real-time runtime protection for LLM and AI agent applications via Lakera Guard, Lakera Red, and the Gandalf crowdsourced threat-intelligence platform, serving Fortune 500 enterprises including Dropbox, Microsoft, Citi, and Cohere.
- Company typePrivate
- Founded2021
- HeadquartersSan Francisco, United States
- Headcount51–100
- GTM typeB2B
- OfferingSoftware
What Lakera does
Lakera is a real-time GenAI and LLM security company founded in 2021 in Zurich, Switzerland, with a US operating presence in San Francisco. The company builds a unified AI-native security platform whose core product, Lakera Guard, provides runtime detection of prompt injection, jailbreaks, indirect injections, data leakage, and policy violations across LLM inputs and outputs, with sub-50ms latency, a stated 0.01% production false positive rate, and multilingual/multimodal coverage across 100+ languages. Lakera's technical differentiation rests on a proprietary threat-intelligence flywheel driven by Gandalf, a crowdsourced red-teaming game with 1M+ players generating 100K+ new adversarial samples daily, which continuously trains its detection models and underpins the published D-SEC threat model and the open-source Backbone Breaker Benchmark (b3).
Lakera monetizes through a hybrid motion combining a free self-serve tier on platform.lakera.ai (product-led growth funnel), usage-based API pricing, and quote-based enterprise subscriptions for Lakera Guard, Lakera Red (AI red teaming), and Workforce AI Security. The company targets Fortune 500 enterprises and regulated industries, with named customers including Dropbox, Microsoft, Citi, Cohere, Nubank, Pearson, and Hinge Health, and has raised $30M in venture funding across a 2023 Redalpine-led seed and a 2024 Atomico-led Series A. The research team is anchored by PhDs from ETH Zurich, Cambridge, Max Planck, Sorbonne, EPFL, and Oxford.
In Q4 2025 Lakera was acquired by Check Point Software Technologies (NASDAQ: CHKP) for approximately $190 million (per Check Point's earnings disclosures; some media reports cite $300 million), forming the foundation of Check Point's Global Center of Excellence for AI Security in Zurich. Post-acquisition, Lakera's technology is embedded in Check Point's AI Defense Plane, Infinity Platform, and Quantum Firewall R82.10, and is slated for integration with Google Cloud's Gemini Enterprise Agent Platform in 2026.
Lakera firmographics
Firmographics- Name
- Lakera
- Legal name
- Lakera Inc
- Website
- https://lakera.ai
- Company type
- Private
- Founded year
- 2021
- Operating status
- Acquired
- Headcount range
- 51–100 employees
- Short description
- Lakera is a Zurich-based AI security company that provides real-time runtime protection for LLM and AI agent applications via Lakera Guard, Lakera Red, and the Gandalf crowdsourced threat-intelligence platform, serving Fortune 500 enterprises including Dropbox, Microsoft, Citi, and Cohere.
- Ownership category
- akta.pro rank
Lakera industry classification
Industry- Product category
- GenAI Security Software
- NAICS
- Software Publishers (5132), Computer Systems Design and Related Services (54151)
- SIC
- Services-Computer Programming Services (7371), Services-Computer Programming, Data Processing, Etc. (7370)
- akta.pro primary industry
- LLM/GenAI Guardrails & Safety Controls (policy, filtering, routing) (HDAAAKAD)
- akta.pro secondary industries
- Model Security Testing & Red Teaming (adversarial ML, jailbreaks) (HDAAAKAC), Attack Detection & Response for Cloud/SaaS (SOC for Cloud) (HDADAGAJ), Prompt Security & Injection Defense (HDAAAKAE)
Keywords
Where Lakera is headquartered
LocationHeadquarters
- HQ city
- San Francisco
- HQ country
- United States
- HQ region
- North America
Offices2 records
Markets served
Lakera business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Technology or R&D, Personnel, Marketing or Sales, Infrastructure, Operations
Revenue model
- Enterprise SaaS Subscriptions: Recurring subscription revenue from enterprise customers using Lakera Guard and Lakera Red. Quote-based enterprise contracts for Fortune 500 and regulated industries. SOC 2 Type I compliant SaaS deployment.
- Usage-Based API Pricing: API-first deployment with usage-based pricing likely tied to volume of secured prompts/transactions. Self-serve sign-up via platform.lakera.ai indicates tiered usage pricing for developers and smaller deployments.
- Freemium / Free Tier: Free tier available via platform.lakera.ai for evaluation and small-scale use. Functions as PLG funnel into enterprise sales.
- Professional Services (Red Teaming): Revenue from Lakera Red AI red teaming engagements, including custom threat snapshots, risk-based vulnerability management, and collaborative remediation guidance for enterprise customers.
- Acquisition / Strategic Value (Check Point): Lakera was acquired by Check Point for approximately $190M-$300M in Q4 2025. Revenue now consolidated within Check Point Software Technologies' broader security subscription revenue (Q4 2025 security subscriptions up 10% YoY to $1,219M).
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Freemium | Pay-as-you-go | Free self-serve tier for evaluation |
| Subscription | Annual | Enterprise quote-based subscription |
| Usage-based | Pay-as-you-go | Usage-based API pricing |
Go-to-market motion4 records
Distribution channels5 records
Marketing channels14 records
Lakera product offering
Product offeringCore offering
Lakera sells an AI-native security platform for enterprises deploying generative AI applications and LLM-powered agents. Its flagship product, Lakera Guard, delivers real-time runtime protection against prompt injection, jailbreaks, data leakage, toxic content, and policy violations in LLM inputs and outputs via an API that integrates with a single line of code. The platform is complemented by Lakera Red (risk-based GenAI red teaming), Workforce AI Security (shadow AI discovery and employee AI usage governance), and Gandalf (a crowdsourced adversarial-data and security training game) that feeds proprietary threat intelligence back into the product.
Product overview
Lakera offers a unified AI-native security platform organized as a platform-plus-modules architecture. At its core is Lakera Guard, the runtime security product for AI agents and LLM applications, which provides real-time prompt-injection, data-leakage, content-moderation, and policy-violation detection. Lakera Guard is complemented by Lakera Red (AI Red Teaming for proactive vulnerability testing), Workforce AI Security (governing how employees use AI / shadow AI discovery), and feature modules including DLP, Content Moderation, Custom Detectors, and the Policy Control Center. Adjacent to the platform, Lakera also operates Gandalf, an AI security education game that continuously feeds crowdsourced adversarial data into the platform's threat intelligence, and publishes the AI Model Risk Index (built on the open-source Backbone Breaker Benchmark / b3) as an industry benchmark for backbone LLM security.
Differentiator
Problem solved
Functional benefit
Brands
- Lakera Guard: Real-time enterprise-grade security product for LLM and AI agent applications, providing prompt attack prevention, data leakage protection, and content moderation.
- Lakera Red
- Gandalf
- AI Model Risk Index
Products and services
- Lakera Guard Lakera's core runtime security product providing real-time, context-aware detection of prompt injections, jailbreaks, data leakage, toxic content, and policy violations in LLM inputs and outputs, with sub-50ms latency, 0.01% production false positive rate, multimodal and multilingual coverage across 100+ languages, and API-first deployment.
- Workforce AI Security Module that secures how employees use AI across apps and browsers, providing shadow AI discovery, context-aware data protection in prompts, and granular policy controls by user, app, and action.
- Lakera Red Risk-based GenAI red teaming service that performs direct and indirect attack simulations, provides vulnerability management, and delivers collaborative remediation guidance to secure AI systems before attackers do.
- Gandalf (AI Security Training Game)
Quantifiable outcome
- 3-4 orders of magnitude risk reduction via context-aware approach
- +6 more outcomes
Companies that use Lakera
Customer profileNamed customers10 records
Segments5 records
Ideal customer profiles4 records
Lakera technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration5 records
AI capability12 records
Feature10 records
Lakera partnerships and signals
Strategic signalPartnerships
Eight partnerships are on record, tiered flagship acquirer / parent company, strategic, core customer/partner and core customer/investor.
- Check Point Software Technologiesflagship acquirer / parent companyCheck Point acquired Lakera in Q4 2025 for approximately $190-300M. Lakera forms the foundation of Check Point's Global Center of Excellence for AI Security. Lakera technology is integrated into Check Point's AI Defense Plane, Quantum Firewall R82.10, and Infinity Platform. Co-developed the Backbone Breaker Benchmark (b3).
- NVIDIAstrategicNVIDIA and Lakera AI jointly proposed a unified framework for agentic system safety (A Safety and Security Framework for Real-World Agentic Systems, arxiv 2511.21990). Case study of NVIDIA's AI-Q Research Assistant. Released dataset of 10,000+ attack and defense execution traces.
- Google CloudstrategicCheck Point (powered by Lakera) integrates with Google Cloud's Gemini Enterprise Agent Platform through Agent Gateway and Agent Registry. Provides Discover, Govern, and Protect capabilities for agentic AI systems. Integration available in late June 2026.
- UK AI Security InstitutestrategicCollaborated with Lakera and Check Point on the open-source Backbone Breaker Benchmark (b3) for evaluating LLM security in enterprise AI agents. Uses 19,000+ adversarial prompts dataset.
- SnykstrategicLakera and Snyk have partnered on AI agent security research, including joint research on agent hijacking attacks. Co-published research findings.
- Coherecore customer/partnerCollaboration with Cohere to define new enterprise LLM security standards. Lakera and Cohere have worked together on prompt injection attacks and other AI security threats. Cohere's Head of Safety is a public advocate.
- MicrosoftstrategicMicrosoft features Lakera's Gandalf game in their PyRIT (Python Risk Identification Toolkit) for AI security. Microsoft also uses Lakera's platform for securing GenAI applications.
- Dropboxcore customer/investorDropbox uses Lakera Guard as their security solution to help safeguard LLM-powered applications, secure and protect user data, and uphold reliability and trustworthiness of intelligent features. Dropbox Ventures also invested in Lakera's Series A.
Scale indicators16 records
Recent moves7 records
Expansion highlights6 records
Lakera competitors and assessment
Company assessmentDirect peers
- Cisco AI Defense (formerly Robust Intelligence): Direct competitor offering AI model firewall and runtime protection for LLMs. Cisco acquired Robust Intelligence in 2024 and integrated its technology into the Cisco AI Defense product line, targeting the same Fortune 500 enterprise customers as Lakera with similar prompt injection and model security capabilities.
- Protect AI: Direct competitor providing AI/ML security posture management, LLM scanning, and red teaming. Acquired by Palo Alto Networks, Protect AI focuses on the same enterprise GenAI security use cases (prompt injection, data leakage, model vulnerabilities) and competes head-to-head with Lakera Guard.
- HiddenLayer: Direct competitor offering AI security posture management and runtime protection for LLMs and ML models. HiddenLayer provides adversarial threat detection, model scanning, and AI risk management capabilities that overlap directly with Lakera Guard and Lakera Red.
- Prompt Armor: Direct competitor specializing in LLM prompt injection prevention and data leakage protection for enterprise AI applications. Prompt Armor offers similar runtime security capabilities to Lakera Guard with a focus on chatbot and agent protection.
- Pillar Security: Direct competitor providing AI security posture management and runtime protection for enterprise AI applications. Pillar Security offers similar prompt injection, data leakage, and AI red teaming capabilities targeting Fortune 500 customers.
Broad incumbents
- Microsoft Azure AI Content Safety: Broad incumbent offering native LLM content moderation and prompt injection detection as part of Azure OpenAI Service. Bundles AI safety controls into existing Azure enterprise contracts, creating a low-friction alternative that competes with Lakera Guard for the same workload protection use cases.
- Google Cloud Sensitive AI Protection: Broad incumbent offering native AI safety controls within Google Cloud and Gemini Enterprise Agent Platform. Provides data loss prevention, content moderation, and model safety features bundled with Google Cloud enterprise relationships.
- AWS Bedrock Guardrails: Broad incumbent offering native prompt injection, content moderation, and sensitive information filtering as part of Amazon Bedrock. Bundles AI safety controls into AWS enterprise contracts, providing a cost-effective alternative for customers already on AWS.
Emerging players
- Adversa AI: Emerging AI security research and product company offering red teaming, adversarial testing, and AI model hardening. Focuses on continuous security evaluation of AI systems, overlapping with Lakera Red's red teaming capabilities.
- Enkrypt AI: Emerging AI safety and security platform offering LLM guardrails, data privacy protection, and AI red teaming. Provides comparable runtime protection to Lakera Guard with a focus on enterprise compliance use cases.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat6 records
Key risks6 records
Key highlights7 records
Customer concentration
Lakera social profiles
Digital presenceLakera compliance and trust
Trust signalCompliance4 records
Lakera financial estimates
Financial estimateRevenue estimate
Valuation estimate
Lakera leadership team
Management profileNumber of profiles
Profiles11 records
Lakera funding detail
Funding detailFunding overview
Funding rounds3 records
Investors6 records
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Lakera M&A and investment
M&A and investmentM&A
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Lakera
What does Lakera do?
Lakera sells an AI-native security platform for enterprises deploying generative AI applications and LLM-powered agents. Its flagship product, Lakera Guard, delivers real-time runtime protection against prompt injection, jailbreaks, data leakage, toxic content, and policy violations in LLM inputs and outputs via an API that integrates with a single line of code. The platform is complemented by Lakera Red (risk-based GenAI red teaming), Workforce AI Security (shadow AI discovery and employee AI usage governance), and Gandalf (a crowdsourced adversarial-data and security training game) that feeds proprietary threat intelligence back into the product.
Is Lakera a public or private company?
Lakera is a private company. It is classified as corporate owned and is currently acquired.
When was Lakera founded?
Lakera was founded in 2021. It employs 51 to 100 people.
Where is Lakera based?
Lakera is headquartered in San Francisco, United States, in the North America region.
How does Lakera make money?
Five revenue lines are on record. Enterprise SaaS Subscriptions are the primary driver. The others are usage-Based API Pricing, freemium / Free Tier, professional Services (Red Teaming) and acquisition / Strategic Value (Check Point).
Who are Lakera's main competitors?
Direct peers on record are Cisco AI Defense (formerly Robust Intelligence), Protect AI, HiddenLayer, Prompt Armor and Pillar Security. Broad incumbents are Microsoft Azure AI Content Safety, Google Cloud Sensitive AI Protection and AWS Bedrock Guardrails. Emerging players are Adversa AI and Enkrypt AI.
Does Lakera have an API?
Yes. Lakera provides a developer-facing API (the Lakera Guard API) that enables real-time, low-latency screening of LLM inputs and outputs for prompt injection, data leakage, toxic content, and policy violations. The API supports per-project policies, customizable detectors, and integrates via a single line of code, with SaaS and self-hosted deployment options. A public documentation portal, implementation guides, and a changelog are available at docs.lakera.ai / platform.lakera.ai. Developer documentation is at docs.lakera.ai/docs/api.
What industry is Lakera in?
Lakera's product category is GenAI Security Software. Its primary akta.pro industry code is HDAAAKAD, LLM/GenAI Guardrails & Safety Controls (policy, filtering, routing), with a secondary code of HDAAAKAC, Model Security Testing & Red Teaming (adversarial ML, jailbreaks). Its NAICS code is 5132 and its SIC code is 7371.