{"slug":"lakera","url":"https://www.altairalabs.ai/vs/lakera/","markdown":"https://www.altairalabs.ai/vs/lakera.md","name":"Lakera","vendor":"Check Point","category":"Gateways and guardrails","checked":"2026-10-22","summary":"Omnia checks each tool call’s arguments and caller before it runs, and refuses what it can’t approve. Lakera, now Check Point AI Guardrails, detects prompt attacks and returns a flag.","headline":"Lakera spots prompt attacks. Omnia stops the tool call they’re after.","scene":{"setup":"A tool result carries a hidden instruction to export the customer list.","omnia":"Whatever the model now believes, the export tool’s policy only allows exports by named roles, to approved destinations. The call is refused before it runs and the audit says why. Adversarial callers tried the same trick before release.","omniaCaps":["per-call-policy","claim-based-access","audit-log","adversarial"],"them":"Lakera returns a flagged result. You decide within your application what mitigating action to take.","source":"https://docs.lakera.ai/docs/api/guard","omniaEvidence":"https://omnia.altairalabs.ai/how-to/security/configure-tool-policies/"},"differences":[{"title":"The action is stopped, not just flagged","body":"If policy says no, or the broker can’t answer, the tool call doesn’t run. The block happens in the platform, with no handler to write in your application.","caps":["per-call-policy","claim-based-access"],"them":"Detection returns a flag; mitigation is your application’s job.","source":"https://docs.lakera.ai/docs/api/guard","omniaEnterprise":true,"omniaEvidence":"https://omnia.altairalabs.ai/how-to/security/configure-tool-policies/"},{"title":"Sensitive data never reaches the model","body":"Named fields in a tool response are replaced before the model sees them, and sessions and memory are redacted before they’re stored.","caps":["tool-response-redaction","pii-sessions","pii-memory"],"them":"PII is detected and flagged.","source":"https://docs.lakera.ai/docs/defenses","omniaEnterprise":true,"omniaEvidence":"https://omnia.altairalabs.ai/how-to/privacy/configure-privacy-policies/"},{"title":"Red teaming that gates the release","body":"Adversarial callers are part of every version’s test suite, run through its real tools, and the release waits on them.","caps":["adversarial","deploy-gate"],"them":"Red teaming is a separate product and service.","source":"https://www.lakera.ai/","omniaEnterprise":false,"omniaEvidence":"https://promptarena.altairalabs.ai/arena/how-to/voice/voice-red-team/"}],"rows":[{"topic":"Prompt injection detection","omnia":"Adversarial tests before release.","caps":["adversarial"],"them":"Direct and indirect prompt attacks.","coverage":"yes","source":"https://docs.lakera.ai/docs/defenses","omniaEnterprise":false,"omniaEvidence":"https://promptarena.altairalabs.ai/arena/how-to/voice/voice-red-team/"},{"topic":"Tool allow and deny lists","omnia":"Per agent, plus a rule per call.","caps":["agent-policy","per-call-policy"],"them":"Tool allow and deny list.","coverage":"yes","source":"https://docs.lakera.ai/docs/defenses","omniaEnterprise":true,"omniaEvidence":"https://omnia.altairalabs.ai/explanation/security/policy-engine/"},{"topic":"Blocking","omnia":"Fails closed.","caps":["per-call-policy"],"them":"Your application acts on the flag.","coverage":"partly","source":"https://docs.lakera.ai/docs/api/guard","omniaEnterprise":true,"omniaEvidence":"https://omnia.altairalabs.ai/how-to/security/configure-tool-policies/"},{"topic":"Redaction before the model","omnia":"Tool responses, sessions, memory.","caps":["tool-response-redaction"],"them":"Detection.","coverage":"partly","source":"https://docs.lakera.ai/docs/defenses","omniaEnterprise":false,"omniaEvidence":"https://www.altairalabs.ai/privacy"},{"topic":"Runs and releases the agent","omnia":"Yes.","caps":["progressive-rollout"],"them":"Not in scope.","coverage":"undocumented","source":"https://www.lakera.ai/","omniaEnterprise":false,"omniaEvidence":"https://omnia.altairalabs.ai/tutorials/progressive-rollouts/"}],"faq":[{"q":"Who owns Lakera?","a":"Check Point acquired Lakera in 2025; Lakera Guard is now Check Point AI Guardrails."}]}