Our product
Moviwa: your AI, fully under control
Blocks confidential data from leaking into ChatGPT, Claude or Gemini in real time. Productivity without exposure, compliance without friction.
Discover MoviwaAI boutique for companies
AI that works. And that you can defend.
We put your company's AI to work, protect your data, govern it and train your team to use it. With the AI Act and the GDPR as the starting point, not an annex.
Behind every project, a team of experts in IT, cybersecurity and compliance (ISO 27001 · GDPR · NIS2). Solutions that adapt, never improvise.
Why trust us
Our pillars
These four principles are not a slogan: they are the standard we apply before accepting any project.
We build before we teach
The same team that deploys AI to production in regulated sectors is the one that trains and certifies: there is no gap between what we teach and what we do.
Compliance as an entry condition
AI Act, NIS2, ISO 27001 and GDPR are not an annex at the end of the project: they are the filter every decision goes through from day one.
Honesty that sometimes says no
We turn down projects that make no sense and flag limits before signing, even if it costs us the sale. We never promise shortcuts that don't exist.
Real sovereignty, not just the word
We work with ALIA-40B, on-prem architectures and processing that never leaves the client's browser: sovereignty shows in the architecture, not in the pitch.
Four pillars
Where we act
Not a closed catalogue: this is how we usually intervene in each area. Each pillar groups several services; see the Services page for full details.
Data IntelligenceData auditing, governance and architecture: the foundation any AI project stands on.View services →AI SecurityAuditing, continuous monitoring and secure architecture for AI systems and agents in production.View services →Compliance and GovernanceAI Act and NIS2 compliance, Shadow AI assessment and auditable evidence.View services →TrainingWe teach what we build: executive sessions, packs and modular programs so your team uses AI with real judgment.View services →
Our own products
Two products, one goal: AI under control
Tools of our own, born out of the projects we take to production. One holds the technical controls; the other, verifiable evidence.
noshadow screenshot
Our product
noshadow: certified AI literacy
The training evidence Article 4 of the AI Act asks for: short courses and a certificate with a verifiable serial number.
Discover noshadowFrom uncertainty to efficiency
Clients & Success Stories
Proven trust in leading institutions and companies. Time savings, cost reduction, improved security.
// Open source
We build in the open
We share the tools, agents and experiments we build with the community. Open collaboration makes our AI safer — and yours too.
montevive/penpot-mcp
Penpot MCP server
#ai#api-integration#cursor
Python23428updated 11mo ago
montevive/autocache
🚀 Autocache - Intelligent Anthropic API Cache Proxy Automatically inject cache-control fields into Claude API requests to reduce costs by up to 90% and latency by up to 85%. Works as a transparent drop-in replacement for popular AI platforms like n8n, Flowise, Make.com, LangChain, and LlamaIndex—no code changes required
#agent#agentic-ai#ai
Go14624updated 8mo ago
montevive/go-name-detector
High-performance Go library for PII name detection. 10x faster than Python with embedded 727K names dataset. Production-ready out-of-the-box functionality for privacy compliance and data security.
#data-security#dlp#go
Go60updated 2mo ago
montevive/prompt-injection-demo
Prompt Injection Demo with n8n
#ai-agents#ai-security#cybersecurity
Python20updated 6mo ago
Models on Hugging Face
Shieldstral-1.0-3B-ONNX
text-classificationtransformers.js
#onnx#ministral3#text-generation
100apache-2.0
ALIA-40b-fc-2605-NVFP4
text-generationvllm
#safetensors#llama#nvfp4
210apache-2.0
ALIA-40b-fc-2605-NVFP4-GGUF
text-generationgguf
#llama.cpp#quantization#nvfp4
240apache-2.0
ALIA-40b-instruct-2601-NVFP4
text-generationvllm
#safetensors#llama#nvfp4
240apache-2.0
// Our own R&D, not talk
We build the technology before we sell it
We don't buy third-party models to resell them: we train, evaluate and publish our own research in the open. At labs.montevive.ai you can try live what we build (safety classifiers, privacy filters, model comparisons) without your data leaving your browser.
We publish them out of conviction and out of method: working in the open is how a claim gets verified, and the same team that audits AI security needs to build it. Every demo came out of a real project.
Explore labs.montevive.aiShieldstral Policy ProbeWatch live how Mistral's safety classifier decides which prompts to block and why, across languages.See Labs →OpenAI Privacy FilterDetect and anonymize personal data before it reaches an external LLM.See Labs →ALIA Tokenizer ComparisonCompare how ALIA and other reference models tokenize Spanish.See Labs →
// Ecosystem
Awards & collaborations
Incubators, mentorships and programs that have shaped our path — and where we now give back.
08institutions · programs · alliances
Frequently asked questions
The questions we answer before we start
What companies ask us most about secure AI, compliance and where to begin. If your question is not here, write to us.
It is Montevive AI's certification attesting that an organization uses AI in a safe, documented and legal way. It verifies four criteria: a completed Shadow AI assessment, a documented AI usage policy with training delivered, active technical controls with traceability, and an EU AI Act and GDPR compliance review. The first certified organizations are Educa Edtech Group and Fundación PTS Granada.
The EU AI Act is already in force and applies in phases: since February 2025 it requires AI literacy for the teams using AI and bans certain practices. It obliges you to classify your AI systems by risk and document them, with fines of up to 7% of global turnover. Our assessment tells you which obligations apply to you and how to cover them.
It is the use of AI tools by employees without company oversight: personal ChatGPT accounts, extensions or bots connected to internal data. It is the main channel through which confidential information leaks into generative AI. We measure it with a dedicated assessment and control it with policy, training and tools such as Moviwa.
We work on four pillars: Data Intelligence (data audit and governance for AI), AI Security (security for AI systems and agents), Compliance & Governance (EU AI Act and NIS2 compliance, Shadow AI assessment) and Training (executive and technical). The same team that deploys AI in production is the one that trains and certifies.
We are based in Granada, but we work with organizations across Spain, remotely or on site. We have supported companies in healthcare, industry, education, tourism and the public sector.
With an assessment: we measure your real AI usage (Shadow AI included), the state of your data and the regulatory obligations that apply to you, and we hand you a roadmap prioritized by risk and impact. Request it from the contact page.
Let's talk
The risk you don't see, and the opportunity you're not seizing yet
We reply in less than 24 hours.
ISO 27001GDPRNIS2Reply in <24h
Request your diagnostic
We'll reply within 24h.