All analysis was generated autonomously, without human review. Scores are analytical opinions drawn from the cited public sources, without hands-on testing. They are not audits, certifications, investment reports, purchasing advice, or evaluations of quality.
Check Point reported buying Lakera, founded in 2021, for about $201.8 million in October 2025. Lakera makes software that security teams use to block data leaks and prompt injection, attempts to trick AI apps and agents into ignoring their instructions. Check Point sells Lakera’s products as Check Point AI Agent Security and Check Point AI Red Teaming, which tests AI apps with simulated attacks. Lakera had raised $30 million, most recently a $20 million Series A led by Atomico, BankInfoSecurity reported. Lakera’s named customers include Dropbox, AI startup Respell and Cohere. Researchers describe the Lakera Guard detector as trained on data from Lakera’s Gandalf prompt-injection challenges, which Check Point credits with over 80 million adversarial patterns a rival may have to gather.
| Description | Lakera is a security platform for GenAI and agent applications that provides runtime protection against prompt injections, data leakage, and jailbreaks, and red-teams AI systems to surface vulnerabilities. | [f1] |
|---|---|---|
| Acquisition | Check Point Software Technologies, announced 2025-09-16 | [f2] |
| Founded | 2021 | [f3] |
| HQ | Zurich, Switzerland | [f4] |
| Funding | $30M total | [f5] |
| Latest funding | $20M Series A led by Atomico, July 2024 | [f3] |
| Deployment | SaaS, Self-hosted | [f6] |
| Compliance | GDPR, SOC 2 | [f6] |
| Product | What it does |
|---|---|
| Lakera | Runtime guardrails plus adversarial testing for LLM and agent apps, screening prompts, responses, and tool calls for prompt injection, jailbreaks, and data leakage. |
AI Defense Matrix
| Govern | Identify | Protect | Detect | Respond | Recover | |
|---|---|---|---|---|---|---|
| AI-Workload Platforms Inference servers, training platforms, vector DB platforms, and the model-loading supply chain. | ||||||
| AI Coding and Orchestration Tools AI coding tools and agentic orchestration tools on user devices, plus their plugins, skills, hooks, system prompts, scaffolding, harnesses, configuration settings, and MCP clients. | ||||||
| AI-Generated Code Code produced by AI tools, AI-assisted reviews, AI-generated infrastructure-as-code and tests, and vibe-coded apps that bypass CI/CD. | ||||||
| AI Traffic Paths to external and self-hosted AI services, MCP tool channels, agent-to-agent calls, model-registry downloads, and egress to unapproved AI services. AI gateways, LLM routers, and MCP gateways steer traffic along those paths. The steering decisions and their inputs belong here too: routing policies, MCP server registries, and agent naming and discovery services. | ||||||
| AI Model Model weights, fine-tuning checkpoints, model cards, registries, an AI bill of materials (AIBOM), and the third-party LLMs your enterprise consumes. | ||||||
| Training Data Datasets used for training, fine-tuning, and continued learning. | ||||||
| Runtime AI Data User prompts, inference inputs, RAG content, vector DB content, persistent agent memory, and interaction history. | ||||||
| AI Agent Identities AI agents as non-human principals, plus credentials, keys, permission scopes, service accounts, and delegation chains across agents and tools. |
Lakera provides runtime guardrails and adversarial testing for LLM and agent apps, screening prompts, responses, and tool calls for prompt injection, jailbreaks, and data leakage. It is mapped to the AI Defense Matrix. [f7]
How well the company can compete in its security market, scored across eight dimensions against public evidence.
| Dimension | Score | Rationale |
|---|---|---|
| Problem Clarity How precisely the company defines its problem, with evidence the problem exists at the scale claimed. | 3/5 | Lakera names a clear buyer, the security teams of enterprises building AI applications and agents, and a clear problem, prompt injection, data leakage and jailbreaks. Researchers in two independent arXiv studies tested Lakera Guard against prompt-injection attacks, and the cited sources contain no figure for what those attacks cost buyers. [s1, s2, s15, s16] |
| Capability Depth How specific the technical capabilities are, with evidence beyond marketing claims such as docs and third-party validation. | 4/5 | Lakera’s public documentation describes the Guard API, a self-hosted deployment and a dated release log, and independent researchers evaluated Lakera Guard in two arXiv studies. The InjecGuard authors found it good at avoiding false alarms on benign inputs containing attack-related words and weaker at detecting malicious inputs. The ChatInject authors described its reputation as relatively strong among detectors. [s6, s18, s15, s16] |
| Market Timing Whether the market is ready for this product, with evidence that buyers are actively seeking solutions. | 3/5 | Lakera’s products depend on enterprise adoption of generative AI and agents, which CSO Online in September 2025 described as an 18-month rush to embed language models and autonomous agents into core workflows. The buyer-side demand in the cited sources is indirect. CSO Online described demand that month from regulated sectors. CTech grouped the purchases of Aim, Prompt Security, Apex and Lakera in an acquisition wave. [s11, s10] |
| Team Credibility Demonstrated domain expertise with public signals such as prior exits, publications, and industry recognition. | 3/5 | BankInfoSecurity reports that CEO David Haber spent three and a half years leading AI at Daedalean AI, an autonomous flight control vendor, before Lakera, and TechCrunch names Haber, Matthias Kraft and Mateo Rojas-Carulla as founders of the company that built the Gandalf game. Independent researchers use Lakera’s PINT benchmark. The founders built and sold one company, Lakera, and the cited sources include no earlier founder exit and no sustained publication record by the team. [s12, s9, s15, s13] |
| GTM Proof Evidence of actual traction (customers, revenue signals, partnerships) beyond stated intentions. | 3/5 | Dropbox speaks for the product on Lakera’s homepage and AI Agent Security page, an unnamed bank describes a regulated deployment, and TechCrunch relayed in 2024 that Lakera claimed Respell and Cohere as customers. Check Point recorded immaterial revenue from Lakera between the October 2025 close and year-end and valued the customer relationships it acquired at $4.4 million. Lakera has not disclosed the size of its customer base. [s1, s2, s9, s13] |
| Funding Efficiency Whether funding matches go-to-market ambition, with signs of capital-efficient growth. | 4/5 | Lakera raised $30 million of outside funding, according to BankInfoSecurity, and Check Point paid about $201.8 million for all of its shares four years after its 2021 founding, according to Check Point’s annual filing. Lakera did not disclose revenue or margin, so the sale is the evidence of what that capital produced. [s12, s13, s9] |
| Category Clarity Whether the company creates or fits a recognizable category that buyers can quickly place in their stack. | 4/5 | Lakera fits the emerging category of runtime guardrails and red teaming for AI applications, and people outside the company place it there. Amit Jaju of Ankura Consulting told CSO Online that the Check Point deal adds AI-native runtime guardrails and continuous red teaming to Check Point’s stack. The InjecGuard authors list Lakera Guard among commercial prompt guard models, the ChatInject authors test it as a detector defense, and TechCrunch describes Lakera as building technology to protect generative AI applications from malicious prompts. [s11, s15, s16, s9] |
| Incumbent Defensibility How vulnerable the core value proposition is to absorption as a feature by a platform vendor. | 3/5 | Lakera has gathered attack data through its Gandalf game that a rival may need time to match, since Check Point cites more than 80 million adversarial patterns and the ChatInject authors describe Lakera Guard as trained on Gandalf challenges. A platform vendor absorbed the capability by acquisition, and Check Point now sells it inside its own AI security portfolio. [s8, s16, s13, s17] |
Lakera addresses runtime risk in AI applications and agents, which it describes as prompt injection, data leakage and jailbreaks, along with agents taking unauthorized actions through tools. On its AI Agent Security page, Lakera says the product detects and blocks prompt attacks, data leakage, unsafe tool use and unauthorized agent actions while agents operate.
The buyer is the security team of an enterprise that ships AI features or agents. Lakera publishes a Dropbox testimonial about safeguarding Dropbox’s LLM-powered applications and a quote from an unnamed bank that chose Lakera for its regulated banking environment.
Independent researchers treat the threat as real. The InjecGuard and ChatInject authors evaluated Lakera Guard against prompt-injection attacks. CSO Online described demand in September 2025 from regulated sectors, technology firms and large companies. [s1, s2, s15, s16, s11]
Lakera’s products now sell under Check Point names. Check Point’s acquisition announcement calls Lakera’s flagship solutions Check Point AI Red Teaming and Check Point AI Agent Security, and on its trust center Lakera says it builds those two products as part of Check Point.
AI Agent Security discovers agents across a customer’s environment, assesses each agent’s risk and inspects prompts, responses and agent actions at runtime. Its runtime layer, AI Guardrails, detects and flags prompt attacks, data leakage and content violations through the Guard API, and customers can also buy AI Guardrails on its own or run it self-hosted. Lakera released the agent discovery and risk assessment features as an early-access product on June 1, 2026, according to its changelog.
AI Red Teaming evaluates, scans and red teams AI applications and agents, and Lakera offers it as automated testing or as an expert-led service. Lakera also publishes a case study on building red teaming capabilities for NVIDIA’s NeMo Agent Toolkit.
Researchers outside the company have measured the detector. The InjecGuard authors found Lakera Guard good at avoiding false alarms on benign inputs containing attack-related words and weaker at detecting malicious inputs. The ChatInject authors measured lower attack success with detectors including Lakera Guard, although default multi-turn attacks succeeded more often than other tested variants. [s8, s5, s2, s18, s6, s3, s15, s16]
Lakera competes from inside Check Point. Check Point launched its AI Defense Plane in March 2026, built partly on technology from its Lakera and Cyata acquisitions, and sells Workforce AI Security, AI Application and Agent Security, and AI Red Teaming as its modules.
Several other AI security startups now sell from inside larger vendors. CTech listed Cato Networks buying Aim, SentinelOne buying Prompt Security and Tenable buying Apex in the same acquisition wave. F5 says it completed its CalypsoAI acquisition and introduced F5 AI Guardrails, and Palo Alto Networks says it completed its Protect AI acquisition.
Check Point chose Lakera after evaluating more than 20 AI security startups, according to CTech. On its trust center, Lakera says its products and Check Point’s own Workforce AI Security together form Check Point’s AI security portfolio. [s17, s10, s19, s20, s5]
Lakera publishes its customer references on its own pages. Dropbox gives a homepage testimonial and a Dropbox security engineer appears on the AI Agent Security page, and an unnamed bank describes a regulated deployment with Portuguese and Spanish requirements. TechCrunch relayed in 2024 that Lakera claimed Respell and Cohere as customers, and Lakera names a research partnership with Snyk.
Check Point disclosed in its annual filing what it paid and what it inherited. Check Point paid about $201.8 million for all of Lakera’s shares and assigned $150.1 million to goodwill that it attributes primarily to synergies, $44 million to core technology and $4.4 million to customer relationships. It recorded immaterial revenue from Lakera between the October 22, 2025 close and year-end.
Lakera still runs a self-serve entry point. Lakera offers free access through its Community plan and advanced features through its sales team, with SaaS or self-hosted deployment. [s1, s2, s9, s13, s7]
David Haber, Matthias Kraft and Mateo Rojas-Carulla founded Lakera in Zurich in 2021. TechCrunch identified them as CEO, CPO and CTO in July 2024. BankInfoSecurity reports that Haber spent three and a half years leading AI at Daedalean AI, an autonomous flight control company, and Check Point describes the founders as AI experts from Google and Meta.
Independent researchers use the team’s work. The InjecGuard authors drew on Lakera’s PINT benchmark and tested Lakera Guard, and the ChatInject authors wrote that Lakera Guard is often regarded as relatively strong among existing detectors.
Haber stayed with Check Point after the sale. In Check Point’s March 2026 launch release he appears as vice president of AI security. According to a March 2026 Swiss commercial register notice, he moved from board president to board member of Lakera AI AG, while Kraft and Rojas Carulla left the board. BankInfoSecurity reported 60 employees and CTech about 70 at the time of the deal. [s9, s12, s8, s15, s16, s17, s14, s10]
Lakera publishes attestations through a trust center branded for Lakera and Check Point AI Security. The trust center lists ISO 27001:2022, SOC 2 Type 2 and HIPAA and offers a 2025 SOC 2 Type II report for AI Guardrails and a 2025 HIPAA report. Lakera also says on it that Check Point maintains a broader compliance program available on request.
Lakera AI AG remains a registered Swiss company. Commercial register entries record the move of its registered seat from Zurich to Spreitenbach, an address it shares with Check Point Software Technologies (Switzerland) AG, according to Moneyhouse.
Lakera’s security page now redirects to Check Point’s AI security page. Lakera offers SaaS or self-hosted deployment and EU or US data residency. [s5, s14, s4, s7]
| Company | Relationship | Note | Compare |
|---|---|---|---|
| Prompt Security | competes with | Competes for the same enterprise AI security buyer, and CTech lists SentinelOne’s purchase of Prompt Security in the same acquisition wave as Check Point’s purchase of Lakera. | |
| CalypsoAI | competes with | Competes for runtime guardrails and red teaming budgets. F5 says it completed its CalypsoAI acquisition and introduced F5 AI Guardrails, so CalypsoAI competes from inside a larger vendor, as Lakera does inside Check Point. | N/AWe captured the evidence for these companies under different evidence-model versions (v1 vs v2), so the totals were scored under different conditions and are not directly comparable. |
| Aim Security | competes with | Competes for the same enterprise AI security buyer, and CTech names Cato Networks’ purchase of Aim alongside Check Point’s purchase of Lakera. | |
| Protect AI | competes with | Competes for AI application security budgets. Palo Alto Networks says it completed its Protect AI acquisition, so Protect AI also sells from inside a larger vendor. | |
| HiddenLayer | competes with | Competes for the same buyer of AI application and agent security. | |
| Noma Security | adjacent | Adjacent, competing for the enterprise AI security buyer that the agent discovery features in Lakera’s AI Agent Security also address. |
Add analyzed competitors to compare them side by side with Lakera.
A closer look at the company's product strategy, measuring how defensible it is against market forces and examining the eight areas behind it.
press the advantage
Lakera has gathered proprietary attack data through its Gandalf games, which Check Point puts at more than 80 million adversarial patterns. The ChatInject researchers, who study prompt injection, cite Lakera’s use of Gandalf and red-team data to train Lakera Guard. A rival running its own attack games over time may gather comparable data. Customers buy software their own teams operate, including an API that screens prompts and agent actions, agent discovery and automated red teaming, and Lakera also sells an expert-led red-teaming service. Lakera lists SOC 2, ISO 27001 and HIPAA attestations for procurement reviews. In its annual filing, Check Point valued Lakera’s core technology at $44 million and attributed most of the $201.8 million price to goodwill from expected synergies.
| Dimension | Score | Rationale |
|---|---|---|
| Value Delivery Does the product sell software as the product, or judgment, trust, or accountability with software as the delivery mechanism. | 2/3 | Lakera blends software with expert services. Customers configure and operate AI Guardrails and automated red teaming themselves, and Lakera also sells an expert-led red-teaming service in which its specialists assess the customer’s AI applications and return findings with remediation guidance tailored to that environment. The runtime products remain software the customer’s team runs and owns. |
| Switching Cost How expensive leaving is for a customer: data portability, integrations, learned workflows, network effects, regulatory data residency. | 2/3 | Customers call the Guard API from their applications and tune policies and sensitivity levels, which Lakera supports with a Policy Impact Simulator, so leaving means rewiring those calls and re-tuning policies in another product. Lakera documents that integration and tuning work but publishes no estimate of what migrating away would cost. |
| Compliance Moat Whether certifications, liability acceptance, or audit trails block an easy replacement. | 1/3 | Lakera lists ISO 27001:2022, SOC 2 Type 2 and HIPAA, attestations a funded competitor can obtain through ordinary enterprise preparation, so they ease procurement without blocking a replacement. |
| Problem Complexity Whether the product requires ML, optimization, real-time systems, or years of specialized expertise. | 3/3 | Lakera has retrained its prompt-injection model and improved its GPU model for text and audio. It says it observes more than 100 languages in real time. Check Point cites latency under 50 milliseconds. The InjecGuard authors measured how prompt guard models, Lakera Guard included, trade catching attacks against over-blocking benign input. |
| Buyer Profile Whether buyers are SMB operators, mid-market IT teams, or regulated enterprises and governments with procurement gates. | 3/3 | Lakera builds and sells its products for the security teams of enterprises, and its customer references include Dropbox and an unnamed bank that deployed it in a regulated banking environment. Lakera also says on its homepage that industry leaders from Fortune 500 companies to startups trust it. |
| Layer Whether the product is an end-user application, a platform with application features, or infrastructure other applications depend on. | 3/3 | Customer applications send model requests through Lakera’s guardrails. Lakera says AI Agent Security blocks prompt injection, jailbreaks and adversarial instructions before they reach the model, and blocks indirect injection through connected tools before agents act. Teams can buy AI Guardrails alone as the runtime layer embedded in their own applications. On its homepage, Lakera quotes Dropbox saying it uses AI Agent Security to safeguard its LLM-powered applications. These are Lakera’s own descriptions. |
| Proprietary Data, Content, or IP Whether the product accumulates datasets, content licenses, or IP that a rival cannot recreate from scratch. | 2/3 | Lakera retains the attack data its Gandalf games collect, which Check Point puts at more than 80 million adversarial patterns and CTech at more than 80 million attack data points. Lakera says on its homepage that it learns from more than a million hackers. The ChatInject authors, citing Lakera, describe Lakera Guard’s proprietary detectors as trained on that Gandalf data. A rival that ran its own attack games and red teaming over time may gather comparable data. |
Lakera sells to the security teams of enterprises that build AI applications and agents. On its homepage Lakera says Fortune 500 companies and startups use it, and its customer references include Dropbox, whose testimonial covers safeguarding LLM-powered applications, and an unnamed bank that chose Lakera for its regulated banking environment.
Since the sale, Check Point sells Lakera’s products as part of a wider AI security portfolio. On its trust center, Lakera says it builds AI Agent Security and AI Red Teaming, which together with Check Point’s own Workforce AI Security, a product for governing employee use of AI, form that portfolio.
The products serve the teams that build and deploy AI. AI Agent Security discovers the agents an organization builds and deploys and connects to platforms such as Amazon Bedrock, Microsoft Copilot Studio and Salesforce Agentforce, according to Lakera’s changelog. CSO Online described demand in September 2025 from regulated sectors such as finance and healthcare, technology firms and large companies.
Lakera’s runtime product, now called AI Guardrails, detects and flags prompt attacks, data leakage, content violations and off-policy agent behavior through the Guard API. AI Guardrails is the runtime layer of AI Agent Security, which adds agent discovery and per-agent risk assessment. Lakera released the discovery and risk assessment features in early access on June 1, 2026.
Lakera trains its own detection models and keeps retraining them. Lakera’s changelog entries cover retrained prompt-injection models for new attack variants, an improved GPU model for text and audio, and a Policy Impact Simulator that estimates how different sensitivity levels would have affected a customer’s past traffic.
Lakera draws its attack data partly from Gandalf. Check Point says Lakera’s platform draws on more than 80 million adversarial patterns from Gandalf, and the ChatInject authors, citing Lakera, describe Lakera Guard as trained on red-team data and Gandalf challenges. The Gandalf address now redirects to Agent Breaker, a Lakera by Check Point game in which players use prompt attacks to hack GenAI applications.
Independent researchers have measured the detector. The InjecGuard authors found Lakera Guard good at avoiding false alarms on benign inputs containing attack-related words and weaker at detecting malicious inputs. The ChatInject authors described its reputation as relatively strong among existing detectors.
Lakera combined a free self-serve tier with enterprise sales. Lakera still offers a free Community plan, and TechCrunch wrote at the July 2024 Series A that Lakera claimed customers in North America, including Respell and Cohere, and was primed to expand, particularly in the U.S.
Check Point bought all of Lakera’s shares on October 22, 2025, for about $201.8 million. In its annual filing, Check Point attributed most of the price to goodwill, which it says arises primarily from synergies. For the period between the close and year-end, Check Point recorded Lakera’s revenue and operating results as immaterial to its consolidated financial statements.
Check Point now takes Lakera’s products to market under its own names. It announced the AI Defense Plane in March 2026, built partly on technology from the Lakera and Cyata acquisitions, with Workforce AI Security and AI Application and Agent Security available immediately and AI Red Teaming in limited release.
Lakera publishes plan packaging and a free tier but no enterprise price. The Community plan costs nothing and includes 10,000 requests a month with an 8,000-token prompt limit on SaaS hosting, and the Enterprise plan offers flexible request volumes, configurable limits and SaaS or self-hosted deployment through a sales conversation.
Lakera packages the plans by request allowance and does not publish how it bills enterprise customers. The Enterprise plan adds single sign-on, role-based access control, SIEM integration and US as well as EU data residency, and Check Point also sells the products inside its AI security portfolio.
Lakera delivers AI Guardrails as a cloud service or as a self-hosted container. Lakera documents an Enterprise SaaS option and an on-premises deployment for maximum compliance control, and lists versioned container releases in its changelog.
Lakera kept shipping through the ownership change. Lakera published changelog entries from November 2025 through June 8, 2026, including retrained detection models, audio improvements and the early-access AI Agent Security release, under Check Point AI Security branding.
Check Point’s acquisition announcement cites detection rates above 98 percent, latency under 50 milliseconds and false positives below 0.5 percent. These are vendor figures, and the InjecGuard and ChatInject authors measured accuracy trade-offs in Lakera Guard without testing those headline numbers.
Lakera publishes its attestations on a trust center branded for Lakera and Check Point AI Security. Lakera lists ISO 27001:2022, SOC 2 Type 2 and HIPAA and offers a 2025 SOC 2 Type II report for AI Guardrails and a 2025 HIPAA report. It also says on the trust center that Check Point maintains a broader compliance program available on request.
Lakera AI AG remains a Swiss company inside Check Point. Lakera AI AG moved its registered seat from Zurich to Spreitenbach, an address it shares with Check Point Software Technologies (Switzerland) AG according to Moneyhouse, and Lakera’s security page now redirects to Check Point’s AI security page.
Check Point built its AI Defense Plane partly on Lakera’s technology. Check Point says the AI Defense Plane is built on its AI Security platform with technology from ThreatCloud AI and the Lakera and Cyata acquisitions, and it pairs Lakera’s products with its own Workforce AI Security.
Lakera’s integrations face outward to the platforms where customers build agents. AI Agent Security connects to Amazon Bedrock, Google Cloud, Microsoft Copilot Studio, Salesforce Agentforce and other agent platforms, and in its documentation Check Point says it works with any AI model or provider.
Lakera also uses Gandalf commercially. CTech reports that Gandalf is used both as a research tool and as a commercial service for Fortune 500 clients. At RSA Conference 2026, Check Point said, it would debut Gandalf: The Agent Gauntlet, an experiential showcase of how agentic systems can be attacked.
David Haber, Matthias Kraft and Mateo Rojas-Carulla founded Lakera in Zurich in 2021. BankInfoSecurity reports that Haber spent three and a half years leading AI at Daedalean AI before Lakera, and Check Point describes the founders as AI experts from Google and Meta.
Haber moved into a Check Point role after the sale. He appears as Check Point’s vice president of AI security in its March 2026 launch release, and according to a March 2026 Swiss register notice, Kraft and Rojas Carulla left the Lakera AI AG board while Haber stayed on it.
Independent researchers use the team’s work. The InjecGuard authors used Lakera’s PINT benchmark and evaluated Lakera Guard, and the ChatInject authors tested it among the defenses they studied.
| Id | Source | Tier | Accessed |
|---|---|---|---|
| f1 | Lakera: The AI-Native Security Platform to Accelerate GenAI | official | 2026-10-26 |
| f2 | SEC EDGAR: Check Point Software Technologies Form 20-F for fiscal year 2025 | regulatory | 2026-10-26 |
| f3 | TechCrunch on Lakera’s $20M Series A | press | 2026-10-26 |
| f4 | Calcalist Tech on Check Point’s $300M Lakera acquisition | press | 2026-06-11 |
| f5 | BankInfoSecurity on Check Point’s Lakera purchase | press | 2026-10-26 |
| f6 | AI Defense Matrix Catalog entry | other | 2026-06-08 |
| f7 | AI Defense Matrix Catalog mapping | other | 2026-10-26 |
| Id | Source | Tier | Accessed |
|---|---|---|---|
| s1 | Lakera homepage “Learning from 1M+ hackers around the world” | official | 2026-10-26 |
| s2 | Lakera: AI Agent Security product page “Secure AI Agents from Discovery to Runtime” | official | 2026-10-26 |
| s3 | Lakera: AI Red Teaming product page “AI Red Teaming gives teams a continuous workflow to evaluate, scan, and red team AI applications and agents.” | official | 2026-10-26 |
| s4 | Check Point: AI Security solutions page (lakera.ai/security redirects here) “Check Point delivers one platform to discover, protect, and govern every AI interaction from employees to agents.” | official | 2026-10-26 |
| s5 | Lakera - Check Point AI Security Trust Center “Lakera is part of Check Point Software Technologies. We build AI Agent Security and AI Red Teaming, helping security teams protect AI agents and applications from discovery through runtime enforcement.” | official | 2026-10-26 |
| s6 | Lakera docs changelog (Check Point AI Security branding) “Changelog | Check Point AI Security” | official | 2026-10-26 |
| s7 | Lakera Guard Platform pricing page (Community and Enterprise plans) “The default pricing plan (Community) provides free access to a limited number of requests. Contact sales to update your plan and get access to advanced features.” | official | 2026-10-26 |
| s8 | Check Point: announcement of the agreement to acquire Lakera (September 16, 2025) “today announced it has entered into an agreement to acquire Lakera, one of the world’s leading AI-native security platforms for Agentic AI applications.” | official | 2026-10-26 |
| s9 | TechCrunch: Lakera raises $20M Series A (July 24, 2024) “Founded out of Zurich in 2021, Lakera officially launched last October with $10 million in funding” | press | 2026-10-26 |
| s10 | CTech by Calcalist: Check Point acquires Lakera “The value of the deal was not disclosed, but it is estimated at $300 million.” | press | 2026-10-26 |
| s11 | CSO Online: Check Point acquires Lakera (deal expected to close in Q4 2025) “Expected to close in Q4 2025 for an undisclosed sum, the deal is expected to boost Check Point’s AI security stack, strengthening enterprise defenses as organizations accelerate their adoption of AI.” | press | 2026-10-26 |
| s12 | BankInfoSecurity: Check Point adds AI application defense with Lakera purchase “Lakera, founded in 2021, employs 60 people and has raised $30 million of outside funding, having most recently completed a $20 million Series A round led by Atomico in July 2024.” | press | 2026-10-26 |
| s13 | SEC EDGAR: Check Point Software Technologies Form 20-F for fiscal year 2025 “In October 2025, we acquired 100% of the share capital of Lakera AI AG, a privately held Swiss company, a leading AI-native security platforms for Agentic AI applications.” | regulatory | 2026-10-26 |
| s14 | Moneyhouse: Lakera AI AG commercial register entry (CHE-153.024.821) “The company Lakera AI AG is registered under the UID CHE-153.024.821.” | regulatory | 2026-10-26 |
| s15 | arXiv: InjecGuard, benchmarking over-defense in prompt injection guard models (2410.22770) “Furthermore, LakeraGuard LakeraAI (2024a) is a commercial guard model developed by LakeraAI, with undisclosed model details, such as scale and architecture, and available solely through an API.” | research | 2026-10-26 |
| s16 | arXiv: ChatInject, chat-template prompt injection against LLM agents (2509.22830) “The Lakera Guard model is a proprietary prompt-injection detector developed by Lakera AI.” | research | 2026-10-26 |
| s17 | Check Point: AI Defense Plane launch announcement (March 23, 2026) “said David Haber, VP, AI Security at Check Point Software Technologies.” | official | 2026-10-26 |
| s18 | Lakera docs: The Check Point AI Defense Plane overview “AI Guardrails, the runtime layer of AI Agent Security: real-time detection and flagging of prompt attacks, data leakage, content violations, and off-policy agent behavior through the Guard API” | official | 2026-10-26 |
| s19 | F5 blog: F5 completes acquisition of CalypsoAI, introduces F5 AI Guardrails and F5 AI Red Team “Today, F5 is pleased to announce that we have completed our acquisition of CalypsoAI , a pioneer at the frontier of AI security, and we are introducing the latest addition to securing our digital roads: F5 AI Guardrails” | official | 2026-10-26 |
| s20 | Palo Alto Networks press release: Palo Alto Networks Completes Acquisition of Protect AI “Palo Alto Networks® (NASDAQ: PANW) announced it has completed its acquisition of Protect AI, an innovative leader in securing AI applications and models.” | official | 2026-10-26 |
| Id | Source | Tier | Accessed |
|---|---|---|---|
| s1 | Lakera homepage “Learning from 1M+ hackers around the world” | official | 2026-10-26 |
| s2 | Lakera: AI Agent Security product page “Secure AI Agents from Discovery to Runtime” | official | 2026-10-26 |
| s3 | Lakera: AI Red Teaming product page “AI Red Teaming gives teams a continuous workflow to evaluate, scan, and red team AI applications and agents.” | official | 2026-10-26 |
| s4 | Check Point: AI Security solutions page (lakera.ai/security redirects here) “Check Point delivers one platform to discover, protect, and govern every AI interaction from employees to agents.” | official | 2026-10-26 |
| s5 | Lakera - Check Point AI Security Trust Center “Lakera is part of Check Point Software Technologies. We build AI Agent Security and AI Red Teaming, helping security teams protect AI agents and applications from discovery through runtime enforcement.” | official | 2026-10-26 |
| s6 | Lakera docs changelog (Check Point AI Security branding) “Changelog | Check Point AI Security” | official | 2026-10-26 |
| s7 | Lakera Guard Platform pricing page (Community and Enterprise plans) “The default pricing plan (Community) provides free access to a limited number of requests. Contact sales to update your plan and get access to advanced features.” | official | 2026-10-26 |
| s8 | Check Point: announcement of the agreement to acquire Lakera (September 16, 2025) “today announced it has entered into an agreement to acquire Lakera, one of the world’s leading AI-native security platforms for Agentic AI applications.” | official | 2026-10-26 |
| s9 | TechCrunch: Lakera raises $20M Series A (July 24, 2024) “Founded out of Zurich in 2021, Lakera officially launched last October with $10 million in funding” | press | 2026-10-26 |
| s10 | CTech by Calcalist: Check Point acquires Lakera “The value of the deal was not disclosed, but it is estimated at $300 million.” | press | 2026-10-26 |
| s11 | CSO Online: Check Point acquires Lakera (deal expected to close in Q4 2025) “Expected to close in Q4 2025 for an undisclosed sum, the deal is expected to boost Check Point’s AI security stack, strengthening enterprise defenses as organizations accelerate their adoption of AI.” | press | 2026-10-26 |
| s12 | BankInfoSecurity: Check Point adds AI application defense with Lakera purchase “Lakera, founded in 2021, employs 60 people and has raised $30 million of outside funding, having most recently completed a $20 million Series A round led by Atomico in July 2024.” | press | 2026-10-26 |
| s13 | SEC EDGAR: Check Point Software Technologies Form 20-F for fiscal year 2025 “In October 2025, we acquired 100% of the share capital of Lakera AI AG, a privately held Swiss company, a leading AI-native security platforms for Agentic AI applications.” | regulatory | 2026-10-26 |
| s14 | Moneyhouse: Lakera AI AG commercial register entry (CHE-153.024.821) “The company Lakera AI AG is registered under the UID CHE-153.024.821.” | regulatory | 2026-10-26 |
| s15 | arXiv: InjecGuard, benchmarking over-defense in prompt injection guard models (2410.22770) “Furthermore, LakeraGuard LakeraAI (2024a) is a commercial guard model developed by LakeraAI, with undisclosed model details, such as scale and architecture, and available solely through an API.” | research | 2026-10-26 |
| s16 | arXiv: ChatInject, chat-template prompt injection against LLM agents (2509.22830) “The Lakera Guard model is a proprietary prompt-injection detector developed by Lakera AI.” | research | 2026-10-26 |
| s17 | Check Point: AI Defense Plane launch announcement (March 23, 2026) “said David Haber, VP, AI Security at Check Point Software Technologies.” | official | 2026-10-26 |
| s18 | Lakera docs: The Check Point AI Defense Plane overview “AI Guardrails, the runtime layer of AI Agent Security: real-time detection and flagging of prompt attacks, data leakage, content violations, and off-policy agent behavior through the Guard API” | official | 2026-10-26 |
| s21 | Lakera: Agent Breaker game (gandalf.lakera.ai redirects here) “Lakera by Check Point” | official | 2026-10-26 |
This site is an experimental research aid created by Zeltser Security Corp. All its data gathering and analysis was performed autonomously without human review, and it can contain errors of fact, interpretation, and judgment that a human reviewer might catch.
The analyses are statements of opinion, not statements of fact. Machine analysis produced the scores, summaries, and matrix placements by weighing the public sources each page cites, and reasonable people can weigh the same sources differently. Where a page states a fact, it cites the public source and the date it was checked, and the statement is only as accurate as that source. Unless a profile expressly says otherwise, the analysis involves no hands-on testing and no independent validation of any company's products or services.
Nothing here is professional, security, legal, financial, investment, or purchasing advice, and nothing here is a recommendation to invest in, do business with, or avoid any company. Inclusion of a company is not an endorsement, and absence of a company is not a judgment about it. Reading this site creates no advisory or client relationship. Verify any detail you plan to act on against the vendor's current materials.
The content is provided "as is" and "as available," with all warranties disclaimed, express or implied, including merchantability, fitness for a particular purpose, accuracy, and non-infringement. No entry is warranted to be complete, current, or correct. Companies change, vendors update their claims, sources can be wrong, and automated analysis can misread them.
To the fullest extent permitted by law, the operator, Zeltser Security Corp, is not liable for any damages that arise from using this site or relying on its content, including direct, indirect, incidental, special, and consequential damages and lost profits, even if advised that such damages were possible. If you are dissatisfied with the site or disagree with these terms, your remedy is to stop using it.
Entries link to vendor pages, press coverage, and other external sites that Zeltser Security Corp does not control and is not responsible for. A link is not an affiliation with the destination or an endorsement of it. Product and company names and trademarks are the property of their owners, used here nominatively to identify the companies described. Short quotations from cited sources appear for identification and commentary.
Use, quotation, automated retrieval, and redistribution of the content are governed by the Terms of Use at cybercompanyprofiles.com/terms, which permit personal and internal business use with attribution and prohibit republication and resale.