Frequently asked questions about EWE
The questions to ask before choosing an AI agent for your company, and EWE’s answers.
Understanding EWE
What is EWE?
EWE is an autonomous artificial intelligence agent for businesses, installed on your server or on a virtual machine in Europe. In practice, it acts on your software, remembers every decision and turns every successful task into a reusable method, under human control.
What is the difference between an AI agent and a chatbot?
A chatbot answers a question; an AI agent achieves a goal. In practice, EWE plans the steps, acts on your tools itself, checks the result and reports back, then remembers it for next time. This is known as agentic AI.
What are the benefits of a model-agnostic AI orchestrator?
A model-agnostic orchestrator gives each task to the best-suited model, instead of locking you into a single vendor. In practice, EWE chooses among more than 300 models, switches to another provider on its own if one goes down and adopts a new model the day it comes out, without rebuilding anything.
How does EWE learn from its tasks?
EWE runs a closed learning loop: after a successful complex task, it writes down the procedure that worked, reuses it and improves it. In practice, its memory is split in three: the conversation history (episodic), facts and preferences (semantic) and skills (procedural).
Which messaging apps does EWE support?
EWE answers on 25 channels, including Slack, Microsoft Teams, WhatsApp, Telegram, Signal, email and SMS. In practice, it is one agent with one memory: a conversation started in Teams can continue on another channel.
Which AIs can it use?
EWE uses more than 300 models: those from OpenAI, Anthropic, Google, Mistral, DeepSeek, Qwen, Llama and many more, or your own AI installed on your servers. You switch model with one command, without rebuilding anything.
Is EWE open source?
No: EWE is proprietary software. Factually, it is built on one of the most-followed open-source technology foundations on GitHub, the developers’ reference platform: more than 250,000 stars and 53,000 forks (as of 30 September 2026). A proven base, and a finished product designed for businesses.
Data and security
How does EWE protect data compared with ChatGPT?
EWE installs on your premises: memory, files and history stay on your infrastructure, and only the request goes to the chosen AI. By contrast, ChatGPT processes and stores data in OpenAI’s cloud; storage in Europe is only offered on the Enterprise and Edu plans. (sources: 3)
Where is our data stored?
Your data stays on your server, or on a virtual machine hosted in Europe with the hosting provider of your choice. In practice, only the request goes to the AI you chose; with an AI installed on your premises, nothing leaves the company.
Can EWE be used without an internet connection?
Yes: with an AI model installed on your servers, EWE works entirely offline. Factually, no request then leaves the company, and there is no more pay-per-use bill; the cost is your hardware, equipped with graphics cards.
Is it GDPR compliant?
EWE gives you the means to comply with the GDPR, but compliance always depends on how it is used. In practice: data on your premises or in Europe, choice of AI provider, human approval of sensitive actions and full traceability with EWE Panel. Your data protection officer knows where every piece of information is.
Can it make mistakes, or take a dangerous action?
Like any AI, EWE can make mistakes: that is why sensitive actions wait for a human green light. In practice, its commands can run in isolated spaces (Docker containers or sandboxes), and EWE Panel logs every action and monitors every budget.
What is a sovereign alternative to ChatGPT Enterprise or Microsoft 365 Copilot?
EWE is an alternative installed on your premises: the agent, its memory and its files run on your infrastructure, with the AI of your choice. By contrast, ChatGPT and Microsoft 365 Copilot remain services hosted by their vendors; among the assistants compared, only Mistral’s Enterprise plan can also be deployed on-premise, with Mistral models only. (sources: 3, 18, 40)
Does EWE work with European AI models?
Yes: EWE uses Mistral’s models, hosted in the European Union by default, and can install open models on your premises. In practice, you choose the provider according to your confidentiality requirements, agent by agent. (sources: 42)
Deployment and costs
How much does EWE cost?
EWE is priced on quotation: the price depends on the number of agents, the chosen hosting and the AIs used. Factually, with an AI installed on your premises, there is no pay-per-use bill: the cost is your hardware.
What if an AI provider retires a model or changes its terms?
You switch to another model with one command, memory and skills intact. Factually, OpenAI retired four models from ChatGPT on 13 February 2026; an AI installed on your premises, by contrast, cannot be taken away from you. (sources: 1, 2)
Do we need a technical team?
Not to use it: you talk to EWE in Slack, Teams, WhatsApp or by email, like a colleague. For installation, EWE Panel deploys with one script, and its updates and rollbacks take one click.
See EWE on your own use cases
A demo of what EWE would do for you, on your tools: installed on your server or in Europe, with the best AIs on the market or your own, and sensitive actions always approved by a human.