What is the difference between ProjectVIEW AI and a generic AI copilot? A generic AI copilot is a general-purpose assistant for writing, summarising and answering questions across many tasks and industries. DANAOS ProjectVIEW AI is the AI add-on to ProjectVIEW ERP: it works on live project data structured by BoQ ↔ WBS ↔ Cost Codes, respects ERP roles and approvals, and can act in the ERP with human approval. The two are complementary: through the ProjectVIEW MCP connector, assistants such as Claude and ChatGPT can work on ProjectVIEW data under the same controls. Choosing an AI copilot for construction depends on the job. This page helps directors in construction, shipbuilding and repair, marine and offshore, mining and quarrying, and project-based manufacturing decide which tool fits which job. What does a generic AI copilot do well? Generic copilots have become strong, everyday tools. For many tasks they are the right choice. Drafting letters, reports, method statements and minutes. Summarising long documents and email threads. Translating correspondence and technical text. Working inside office documents, spreadsheets and presentations. General research and brainstorming. They are designed to be broad. That breadth is also their limit in project controls: they do not know your cost structure, your commitments or your approval rules unless that context is supplied and governed. How do they compare on project work? Criterion Generic AI copilot ProjectVIEW AI Data context Documents, email and files the user can open; no live link to project cost data unless connected Live ERP data across estimating, procurement, subcontractors, materials, machinery and finance Project cost structure None by default; must infer from what it is given Every record tied to BoQ ↔ WBS ↔ Cost Codes Permissions Based on the office or file environment Inherits ERP identity, role and permissions Actions in the ERP None, unless connected through an integration Creates and updates ERP records after human approval Audit Chat history Each step logged: detection, recommendation, approver, action Industry logic General knowledge Cost recipes, committed vs actual cost, variations, retention, certifications, EVM Setup effort Low: available out of the box Configured on ProjectVIEW ERP or on top of existing systems Best for Writing, summarising, translating, personal productivity Cost control, procurement checks, forecasting, early warnings, governed actions When is a generic copilot enough? Use a generic copilot when the task is about language, not about project numbers, and when a person checks the output before it matters. Writing a response to a client letter. Summarising a subcontract before a meeting. Preparing a presentation from notes. Translating site instructions. If the answer does not depend on current committed cost, budget rates, progress or approvals, you do not need an ERP-aware AI for it. When do you need AI that knows the project? The question changes when the answer depends on live project data and leads to a decision. “Which projects are most likely to exceed budget?” “What is the current committed cost for concrete on Project X?” “Which subcontractors represent the highest commercial risk?” “Why is Project Alpha over budget?” “Show me my pending approvals.” A generic copilot can only answer these if someone exports and explains the data first, and even then the answer is a snapshot. ProjectVIEW AI answers on the live data set, within the user’s permissions, and its agents follow the cycle Detect → Investigate → Explain → Recommend → Request Approval → Execute → Audit. See the ProjectVIEW AI assistant and ProjectVIEW AI agents. How does ProjectVIEW work with Claude and ChatGPT? This is not an either-or choice. MCP (Model Context Protocol) is an open standard for connecting AI assistants to business systems. The ProjectVIEW MCP connector is available for Claude and ChatGPT and lets users: Run headless operations without opening ERP screens. Ask questions of their own data in plain language. Create the dashboards and KPIs they need. Build their own AI agents and assistants. The governance chain stays the same: Identity → Role → Permission → Data → AI → Action. The assistant sees what the user’s role allows, and consequential actions still require approval in the ERP. Teams can keep the copilot they already use and give it governed access to project data. Details are on the ProjectVIEW MCP server page. What about data security? With ProjectVIEW AI, project data is not sent to public consumer AI services. It runs in a private environment on Microsoft Azure or on-premises. Azure options include prompts not used for model training, data residency options, encryption at rest and customer-managed keys. DANAOS Projects holds ISO 27001, ISO 9001, ISO 14001 and ISO 20000-1. For generic copilots, check the terms of your specific licence: consumer and enterprise plans often differ on data use and retention. See security and governance. How do you decide? List the questions your directors and project managers ask every week. Mark which ones depend on live cost, commitment, progress or approval data. Use a generic copilot for the rest. For the data-dependent ones, use ProjectVIEW AI directly or through the MCP connector in Claude or ChatGPT. Keep human approval on every action that affects cost, payment or contract position. For the underlying ERP, see construction ERP software. Frequently asked questions Is ProjectVIEW AI a replacement for Claude or ChatGPT? No. Generic assistants remain useful for writing and general work. ProjectVIEW AI adds project-aware answers and governed actions on ERP data, and the MCP connector lets Claude and ChatGPT use that data under the same permissions. Can a generic copilot read our ERP data? Only if it is connected to it. Without a governed connection, users export data into the copilot, which creates copies outside ERP controls and answers based on a snapshot. Which models does ProjectVIEW AI use? It is model-flexible. Domain-specific small language models handle pattern recognition and numerical analysis, and leading LLMs handle reasoning, language and multi-step planning. ERP context is retrieved at inference time, so models are not retrained on client data. Does ProjectVIEW AI take actions without approval? No. Consequential actions require human approval and are logged for audit. How is ProjectVIEW AI priced? Pricing depends on the modules, the number of users and the deployment option. AI usage can be metered with DANAOS AI Credits. Want to see it on your own data? See ProjectVIEW AI answer questions on live project data, and how it connects to Claude and ChatGPT through MCP. Book a Demo