Every week, we get the same question: We already have Microsoft CoPilot, why would we not just configure our requirements extraction and review in there?
The short answer: No. Not if you care about accuracy, auditability, and data sovereignty.
Of course in a real meeting we give more context, that longer answer is what this article is about. Horizontal AI tools — general-purpose assistants designed to help everyone write emails, summarize meetings, and debug code — are fundamentally mismatched with the demands of Systems Engineering (SE).
Even with custom prompts, plugins, and careful configuration, they hit a structural ceiling that vertical solutions like Basewise were specifically architected to break through.
General assistants require an exceptionally broad dataset to work, which is why the models which power CoPilot are trained on data of the entire internet. As such, they know a little about everything and are an expert in none.
All frontier foundation models are trained as generalists. Research from DeepMind and EPFL demonstrates that multi-task fine-tuning inevitably creates constrained capacity and negative task interference — the same weights optimized for legal reasoning, creative writing, and open-ended conversation cannot simultaneously achieve the depth required for safety-critical engineering standards [1].
We do not pretend to escape this foundation. We use the same base model family as Copilot — but that is where the similarity ends. Where Copilot is a one size fits all consumer endpoint governed by Microsoft’s backend routing, capacity management, and opaque parameter schedules. This stack is subject to change in context windows, models, token usage and other parameters which change the outputs without warning.
Basewise owns the full stack. We control the endpoints, the model variants, and the underlying inference parameters directly. That control means the behavior of our system is deterministic and tuneable, not subject to the quality fluctuations that occur when a shared cloud backend throttles capacity or silently swaps model versions.
Inside that controlled environment, we cage the generalist: deterministic INCOSE guardrails execute before any token reaches the model; project-specific retrieval grounds every response in your contract, standards, and verification history; confidence scoring forces the system to declare uncertainty rather than hallucinations; and a narrow operational scope strips away everything except the reasoning required for the engineering task at hand.
So where CopIlot gives you a raw generalist with variable backend, Basewise gives you a specialized expert which operates in a controlled environment.
When it comes to Systems Engineering, precision is a non-negotiable. A requirement like "the system shall respond quickly to overcurrent conditions" is poorly written, unverifiable, untestable, and potentially contractually dangerous. Generic AI might flag it as "vague," but only a system trained on INCOSE and IEEE 29148 standards can explain why it's unverifiable, cite the specific rule (missing measurable threshold), and propose a compliant rewrite. Moreover, only an engineered system will ensure it delivers that analysis consistently and warns when it encounters ambiguity.
Analyst firm Gartner predicts that this vertical specialization will become more important in using AI effective in Enterprise solutions, as they predict that 40% of Enterprise Apps Will Feature Task-Specific AI Agents by this year [2].
In regulated industries — healthcare, finance, insurance, and yes, large-scale infrastructure — every decision must be traceable years into the future. A system that cannot explain its determinations is not just unhelpful either, the lack of explainability constitutes a compliance gap.
Microsoft Copilot's audit logs capture prompts and final responses, delivering some of the required trail. However the log is not engineered to include the model’s reasoning steps and intermediate data, which is necessary context if an audit investigates why a requirement was approved.
For Systems Engineering, this is a dealbreaker. INCOSE standards, ISO/IEC/IEEE 15288, and project-specific contractual frameworks demand traceability. When Basewise's Requirements Quality Analyser (RQA) flags a requirement as ambiguous, it links to a specific rule code, shows the exact text triggering the flag, and generates an audit-ready report. Every finding is explainable. and decisions are traceable.
Audit trails cannot be an afterthought — they are part of the architecture. Our output is designed to withstand scrutiny from clients, regulators, and courts.
Here's something most enterprise AI buyers miss: running Copilot in Europe doesn't guarantee your data stays in Europe. Microsoft's own documentation acknowledges that Copilot may route requests to the primary tenant location — meaning a European user's prompt could be processed in the United States if that's where the tenant home region is set. For infrastructure projects governed by GDPR, the EU AI Act, and national procurement laws, this is unacceptable.
Even with multi-geo configurations, Copilot does not guarantee data processing in the same region as the user's mailbox or SharePoint site. And fundamentally, you're still dependent on a US tech giant's infrastructure, subject to the CLOUD Act and opaque subprocessors. This is not down to poor performance by Microsoft, it’s simply the legislation its subject to.
Basewise made a deliberate, strategic decision: we are moving off the Microsoft stack entirely. We are in the midst of achieving this transition, as we work to move our inference off hyperscaler stacks and onto a sovereign EU provider.
Soon we operate on a fully data-sovereign EU stack with in-house technology and zero dependencies on large tech providers. Your data never leaves European jurisdiction. In many projects, this is a legal requirement.
The most dangerous of all failures is AI’s equal confidence whether it is right or wrong.
In Systems Engineering, a hallucinated regulatory claim or a fabricated verification standard can derail a €500 million infrastructure project, create legal claims and even lead to dangerous situations.
Copilot has no built-in validation for regulatory accuracy. It doesn't know which version of a building code applies to your Dutch rail project. It can't tell whether a verification method satisfies your specific EMVI contract. And when it's uncertain, it does not naturally flag that uncertainty — it guesses. Due to the underlying technology, even adding this context to a knowledge store does not guarantee the correct application.
Basewise addresses this with specialized retrieval engineering, structured reasoning, source citation, and confidence scoring. When our Requirements Evidence Finder (REF) can't find sufficient proof for a requirement, it says so. If our apps are not sure about what they find, they flag it to you and loop humans into the review.
The outputs are not obscure, they are an accurate reflection of the analysis and will escalate to a human when needed.
You can't prompt your way to INCOSE compliance. The INCOSE Guide to Writing Requirements, ISO 29148, and the SE Handbook represent a domain-specific ontology — a structured way of thinking about necessity, verifiability, consistency, and traceability that took decades and many experts to formalize.
Research comparing AI-assisted requirement evaluation to human expert assessment found that while AI tools "can provide consistent and rapid preliminary assessments, particularly for syntactic and structural quality attributes... expert judgment remains essential for contextual interpretation, ambiguity resolution, and trade-off reasoning" [5].
The key is embedding that expert judgment into a multi-agent architecture — not asking a single general model to pretend it understands systems engineering. Basewise has engineered the review process as a specialized framework that assigns tasks to specific models and trains agents for discrete functions:
Each agent operates within a narrow, governed scope with task-specific parameters, creating a system instead of a single agent performing a review.
Research into specialized industries supports the view that any gap cannot be bridged by improved prompts. The structural mismatch between the operational design of general-purpose AI frameworks and the actual requirements of regulated industries results in a mismatch by design.
The shift from horizontal to vertical for niche application is not new or a theory, there are quite few examples in the market. Gartner predicts the agentic AI market will reach $450 billion by 2035, with domain-specific implementations outpacing general-purpose deployments [2].
Systems Engineering is following the same trajectory. The question isn't whether you'll adopt AI for requirements management — it's whether you'll adopt the tool built for your domain or try to force a generalist to do a specialist's job.
When we started Basewise we did not set out to build another chatbot. We wanted to deliver a high quality, reliable and purpose built system which takes systems engineering to the next level. Basewise needs to save time of engineers, drive the quality of projects up and embed systems engineering best practices in more projects.
Our current solutions target key areas of the SE journey:
We've processed 50,000+ requirements and performed 10,000+ verifications. Our customers report time savings and quality improvements in every project they run.
Keeping all our points in mind, here’s the outcome:
But by design, generic horizontal assistants are not designed to deliver the right outcome. Outputs will vary, change over time and may be hard to explain. In the end, that approach puts the SE process and projects at risk.
And that is why you need AI built for SE from the ground up. That's Basewise.