OpenAI Telecommunications Lisbon, Portugal Company newsroom

MEO Portugal

Document Processing · Enterprise Knowledge Search

MEO's legal analysts had to handle a steady stream of regulatory requests that required cross-referencing corporate legal records and manually processing large volumes of legal documents, consuming significant time and effort for compliance work.

Reported impact

Business outcome

The assistant helps MEO reduce the hours and effort required for compliance work, respond to regulatory requests with greater speed and efficiency, generate executive-ready summaries, and free legal analysts to focus on more complex work.

  • Productivity Improvement
  • Risk Reduction

Solution

DXC and MEO built a proof-of-concept AI assistant with ChatGPT capabilities to support regulatory compliance work for MEO's legal team. The implementation centers on a controlled, private document repository that holds legal documents and acts as the governed source for answers, helping preserve confidentiality. Legal analysts interact with the assistant through conversational prompts, and the system automatically sources relevant information from the document set, structures it into a standardized response format, and produces summaries and next-step guidance. During development, MEO's legal analysts provided recurring feedback on responses to improve accuracy and usefulness, creating a human-in-the-loop tuning process. The assistant serves both frontline analysts, who use it to answer regulatory requests faster, and MEO's executive board, which receives regulatory response summaries for quicker decision-making.

Data flow

MEO's legal documents are kept in a controlled, private document repository. Legal analysts submit conversational prompts to the AI assistant, which uses ChatGPT capabilities to source relevant information from those documents, format responses in a standardized way, and return summaries, insights, and recommended next steps to analysts and the executive board.

Solution architecture

3 components · 3 layers
  1. Compute
    • ChatGPT Supplies the generative AI capabilities used to analyze legal content and generate responses, summaries, and next steps.
  2. Storage
    • controlled, private document repository Stores the legal documents that the assistant uses as its secure source of regulatory response information.
  3. Serving
    • AI assistant Provides a conversational interface for legal analysts to retrieve regulatory response information and actionable guidance from MEO's legal documents.

Architecture clues

  • Conversational prompt interface for regulatory query handling
  • Human feedback loop from legal analysts was used to improve response effectiveness and accuracy
  • Proof-of-concept implementation that became a proof of value
  • Uses a controlled, private document repository for security and confidentiality

Evidence from the source

"DXC developed an AI assistant with ChatGPT capabilities."
"During development, MEO’s legal analysts provided regular feedback on the AI assistant’s responses to improve its effectiveness and accuracy."
"It's a tool that allows us to free up resources and speed up the work carried out by our team of legal analysts."
"The generative AI solution, first created as a proof-of-concept, simplifies the process of regulatory compliance by automatically sourcing regulatory response information from legal documents into a standardized format."
"Using a controlled, private document repository, the tool ensures maximum data security and confidentiality."
"Using conversational prompts, legal analysts can quickly get the data they need to respond to regulatory requests with greater speed and efficiency."
Transparency Evidence & methodology 75/100 reference value · 92% confidence · 1 source

Review how this reference was scored and trace each normalized claim back to its documented source.

01

Scoring methodology

How scores are calculated

Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.

Customer reference value75/100
  • Problem clarity20/20
    Measures whether the customer's operational problem is specific enough to compare.
  • Implementation depth25/25
    Rewards named components, data flow, integrations, and architecture detail.
  • Outcome specificity20/20
    Measures whether the reported benefit is concrete rather than a generic claim.
  • Quantified impact0/25
    0 metrics describe measurable impact.
  • Deployment scale10/10
    Rewards evidence of a live or scaled deployment and meaningful integration scope.
Evidence confidence92%
  • Source authority25/25
    The deployment is described by the adopting organization or a government source.
  • Claim traceability21/25
    10 of 12 key fields link to a named source.
  • Direct evidence20/20
    6 direct source quotes retained.
  • Date provenance10/10
    The publication date comes from the source page or publisher.
  • Source transparency8/10
    1 source reference document publisher, URL, authority, retrieval, and supported fields.
  • Implementation specificity8/10
    Rewards source-backed components, data flow, quantified impact, and a specific outcome.
02

Traceability

Sources & provenance

Field coverage shows which facts each source supports. Dates are shown only when documented by the source.

  1. MEO Portugal

    MEO Portugal customer story

    First-party
    https://dxc.com/insights/customer-stories/ai-legal-assistant-meo-portugal
    Source type
    Customer-published story
    Published
    Mar 21, 2025
    Supports
    • Customer
    • Business problem
    • Solution
    • Products
    • Solution components
    • Data flow
    • Integrations
    • technical area
    • use case category
    • Business outcome
    • outcome category
    • Metrics
    • Architecture
    • evidence quotes
    • Publication date

Date evidence

  • Published on source page Mar 21, 2025 · DXC Technology · First-party
    Open date source
Not separately sourced Industry, Region, Company size. Treat these normalized attributes as lower-confidence until a dedicated source is added.