Audits were never designed for agents that ship weekly
On 17 September 2026, TechCrunch reported that Comp AI raised a $34 million Series A led by Roo Capital and Grand Ventures (about $37.5 million total funding). Founded by Lewis Carhart, Claudio Fuentes, and Mariano Fuentes after lessons from LeapAI’s SOC 2 grind, Comp AI builds agentic workflows for policies, evidence collection, continuous control monitoring, vendor assessments, and AI-assisted penetration testing — while stressing that agents draft and humans still approve, and that independent auditors are not replaced.
CEO Carhart’s sharpest line for product teams: imagine finishing SOC 2, then two weeks later deploying an AI agent that can access customer data or change permissions. The audit did not become invalid; it simply was not designed to tell you in real time what changed afterward. That is the gap continuous agentic compliance tries to close — and the same gap MENA scale-ups hit when enterprise buyers ask for fresh evidence mid-deal.
Engineering and UX implications
1. Evidence as a product surface. Treat control status, agent permission diffs, and audit trails as first-class UI — not PDF exports at quarter-end. Design systems should show “what changed since last attestation” in plain language for security and sales alike.
2. Human approval scales with consequence. Comp AI’s founders argue safeguards should increase as agents take more consequential actions. Encode that in workflow UX: low-risk drafts auto-queue; high-risk permission changes require named approvers.
3. Pen-test agents are dual-use careful. Automated testing of codebases and infrastructure helps startups move faster, but keep scope contracts and change windows explicit — AdSense-safe editorial note: this is about defensive continuous assurance, not offensive playbooks.
iFynx takeaway
Point-in-time compliance is a lagging metric in the agent era. Build products that emit continuous evidence and bilingual control narratives — or your next enterprise deal will stall while someone screenshots Confluence.
Acceptance checklist for continuous agentic compliance
Audit theater collapses when agents ship weekly. Continuous control monitoring needs product UX that auditors and engineers both trust.
- Every control has an owner, evidence source, freshness SLA, and last-pass timestamp
- Agent-collected evidence is labeled distinctly from human-uploaded evidence
- Failed controls open tickets with bilingual remediation steps — not only red badges
- Vendor assessment workflows store questionnaire versions and who attested
- Penetration-test assists never auto-close findings without human sign-off
- Kill switch can pause agents that mutate policies while freezing read-only evidence collection
If an auditor asks “when did this control last prove itself?” the answer must be one click away.
UX pitfalls after the logo wall of certifications
- Dashboards celebrate “SOC 2 ready” while evidence freshness is weeks stale
- Engineers drown in false-positive control failures and mute alerts
- Policy agents rewrite docs in English only; Arabic ops teams drift out of sync
- Vendor risk scores lack explainability, so procurement ignores them
- AI pen-test noise floods Jira without severity calibration
Acceptance includes a surprise auditor drill: reconstruct last quarter’s evidence pack in under two hours.
Ops playbook: four weeks to continuous, not ceremonial, compliance
- Week 1: inventory controls, owners, and evidence freshness gaps
- Week 2: ship agent-vs-human evidence labels + ticket templates
- Week 3: bilingual remediation for the top ten failing controls
- Week 4: freeze-agent drill + auditor reconstruction timer
- Success metric: % of critical controls with evidence fresher than SLA
Agentic compliance wins when Monday’s deploy does not orphan Friday’s audit story.
MENA enterprise notes
Regional banks and fintechs will ask about residency of evidence stores and Arabic auditor packs.
- Clarify where screenshots, logs, and configs live relative to customer data regions
- Provide Arabic executive summaries even if control technical detail stays English
- Map frameworks (SOC 2, ISO, local circulars) without pretending one agent covers all
- Keep human attestation visible for high-risk vendor and pen-test closures
Continuous means explainable every week — not a scramble before renewal.
Treat this as an operating practice, not a one-off redesign workshop. Schedule a monthly review of the metrics and checklists above, assign a single accountable owner, and keep a short changelog of product fixes shipped since the last review. When stakeholders ask for more visual polish, redirect the conversation to the open defects that still block conversion, trust, or operator adoption. Depth compounds when the same checklist is reused across markets with only locale-specific payment, language, and compliance adjustments.
What continuous compliance actually involves
Traditional compliance works in cycles: prepare for an audit, collect evidence, pass, then repeat a year later. Continuous compliance replaces the scramble with ongoing checks. In practice that means:
- Automated evidence collection from cloud accounts, identity providers, code repositories and HR systems, stored with timestamps.
- Control monitoring that flags drift, such as a disabled MFA policy, a public storage bucket or an offboarded employee who still has access.
- Policy management with versioning, approvals and employee acknowledgements.
- Vendor risk reviews that track which suppliers process which data and when their reviews expire.
- Change tracking that records significant changes, including new AI agents and their permissions, between audits.
Where AI agents fit, and where humans must stay
Agents are well suited to drafting policies from templates, mapping existing controls to multiple frameworks, chasing evidence owners and summarising what changed since the last review. Humans must remain responsible for approving policies, accepting risks, signing attestations and working with independent auditors. A product that blurs this line creates risk for its customers rather than reducing it.
Frameworks MENA scale-ups are asked about
Gulf enterprise buyers and regulators often ask vendors about international frameworks such as SOC 2 and ISO/IEC 27001, alongside local requirements. In Saudi Arabia, the National Cybersecurity Authority’s Essential Cybersecurity Controls apply to many organisations and their suppliers, and the Personal Data Protection Law, enforceable since 14 September 2024, covers personal data processing. In the UAE, the federal Personal Data Protection Law (Federal Decree-Law No. 45 of 2021) applies, with separate regimes in DIFC and ADGM. A compliance platform that maps one set of controls to several frameworks saves real effort for companies selling across the region.
Governing AI agents inside the compliance programme
The scenario Comp AI describes, a new agent with access to customer data deployed after the audit, is common. Add agents to your asset inventory, require a risk review before an agent gets access to sensitive data or permission-changing tools, and monitor their permissions as controls. The OWASP Top 10 for LLM Applications 2025 and the NIST AI RMF provide useful control ideas, such as least privilege, human approval for high-impact actions and evaluation before release.
UX for compliance tools
Compliance software is used by engineers, managers and executives with very different needs. Engineers want precise, actionable failures with links to the resource; managers want status by team; executives want a short summary of risk and readiness. A “what changed since the last attestation” view, in plain language, is often the most valuable screen of all.
Evidence that enterprise buyers in the Gulf ask for
Vendors selling to Gulf banks, telcos and government entities are often asked for more than a certificate. Typical requests include current penetration test summaries, data flow diagrams showing where data is hosted, lists of sub-processors, incident response procedures, business continuity plans and confirmation of how AI features handle customer data. A compliance platform that keeps these documents current and shareable through a trust page or secure data room shortens sales cycles noticeably.
Getting started with a small team
Startups do not need a large compliance team to begin. Start by connecting identity, cloud and code systems to an evidence tool, write a short set of core policies, assign an owner for each control and schedule a quarterly review. Add frameworks and automation as customers ask for them, rather than trying to cover everything at once.
Common mistakes
- Treating the audit report as proof of current security.
- Letting AI-generated policies go live without review.
- Leaving new AI agents and integrations out of the asset inventory.
- Collecting evidence manually in spreadsheets that nobody updates.
- Mapping controls to one framework only, then redoing the work for the next customer.
Continuous compliance is ultimately a habit supported by tools: small, regular checks that keep evidence fresh, so that the next customer questionnaire or audit is a review rather than a scramble.
For Gulf scale-ups selling to large enterprises, that readiness often decides whether a deal closes this quarter or next year.
Frequently asked questions
Does continuous compliance replace audits?
No. Independent auditors still assess and attest; continuous compliance keeps you ready and shows what changed between audits.
Can AI write our security policies?
It can draft them, but people in your organisation must review, adapt and approve them.
Which framework should a Gulf startup start with?
Start with what your customers and regulators require. Many begin with SOC 2 or ISO/IEC 27001, mapped to local requirements.
How do AI agents affect our compliance scope?
Any agent with access to customer data or systems becomes part of your control environment and should be inventoried, reviewed and monitored.
Related: Saudi PDPL compliance checklist · Governing agents like services. Building compliant SaaS for the Gulf? See AI development for Saudi Arabia and AI development for the UAE.
