Fidelidade: the RiskMandate X-ray

a company · prepared · sources read on 2026-11-07 · the deck · how this X-ray was made · X-ray another name

What happened: how this X-ray was made →

With JavaScript, the graph of this X-ray and a player for each use case's simulation are drawn here. Everything they show is written out below.

What it does

Fidelidade, founded in 1808, is the market leader in Portuguese insurance (life and non-life, a 28.1% share), present in 14 countries on 4 continents, with gross written premiums of 6,529 million euros in 2025; Fosun holds about 85% of its capital and Caixa Geral de Depósitos about 15%. Its in-house Center for Artificial Intelligence & Analytics builds the AI it runs: GAMA, which reads claims documents and health invoices; Contact IA, a natural-voice phone line and email triage for its contact centres; Knowledge AI, a language-model assistant over the company's operational knowledge; and Trusty, the virtual assistant on its website and app. Its 2025 report sets out a six-pillar AI strategy that includes fairness, compliance with the AI Act and GDPR, and alignment with the business. s2 s16 s18 s5 s4 s8 s1

Its AI agents

GAMA (documents)

Generic Analytical Model Automation: an AI document-processing system built in-house by the Center for Artificial Intelligence & Analytics, adapted to European Portuguese. It reads the documents customers submit, including the motor accident declaration (DAAA) and health reimbursement invoices, and extracts the data automatically. In April 2025 its director said it processed more than half of the declarations without human intervention and asks a staff member when it is unsure of an interpretation. The 2025 report gives 5,460 documents a day and 64.6% automation of the information collected. s8 s9 s5 s6

What it can do:

Reads uploaded or photographed motor accident declarations and extracts their data
Processes health expense reimbursements and healthcare provider invoices (Portugal, and provider billing in Angola and Mozambique)
Runs configurable workflows, fields and integration flows for any line of business, geography or document type
Hands uncertain interpretations to a human and learns from the result

Contact IA (voice/email)

Contact AI (CAI), piloted at the Group Contact Center from 2024: a natural-voice IVR that identifies and authenticates customers, understands the reason for the call in natural language and routes them to an operator or team, with Call Analytics on interactions and Email Automation that classifies requests and sends them to the right team. The 2025 report says its architecture will pave the way for true self-service. s5 s6

What it can do:

Identifies and authenticates callers by voice
Classifies the reason for a call and routes it to an operator or team
Classifies incoming emails and routes them to teams
Analyses calls and emails to extract operational information

Knowledge AI (KAI)

The company's knowledge platform, built by the Center for Artificial Intelligence & Analytics and enriched by business staff over two years: a vector knowledge base with version control, a content editor, semantic search with source verification and a conversational assistant based on large language models, able to interact with external systems. Fidelidade calls it its 'digital brain' and says it feeds business knowledge to all of the company's AI systems. s4 s11 s5

What it can do:

Answers staff questions on policies, procedures and coverage in seconds
Interacts with external and internal systems through a language-model integration layer
Supplies business knowledge to the company's other AI systems
Creates content such as FAQs, eLearning and simplified text, under workflows and permissions

Trusty (assistant)

The virtual assistant on fidelidade.pt ('Posso ajudar com tudo o que estiver relacionado com os serviços e produtos da Fidelidade': I can help with everything related to Fidelidade's services and products) and on the home screen of the MyFidelidade app, where the motor claims page says a customer can ask it for roadside assistance. The 2024 report names it as the Trusty chatbot. Fidelidade does not publish what systems or data it reaches. s1 s3 s6

What it can do:

Answers customer questions on the website
Takes motor roadside assistance requests in the MyFidelidade app (as an alternative to the app's form)

MARIA (speech bot)

A speech bot built by Fidelidade Assistance and the Advanced Analytics centre for the Fidelidade Assistance contact centre. ECO quoted the company in 2021 as saying it identifies callers automatically through their policy and other information, speaks and listens through a cloud speech service, and decides with a flow orchestrator developed by the group. The 2023 solvency report says work that year improved the speech-bot in car assistance, with the aim of fully automating car assistance, invoicing and provider payments. s10 s7

What it can do:

Answers calls to the assistance line
Identifies the customer from the policy and other data
Interprets the customer's messages with a flow orchestrator to decide the next step

Use cases: reach, mandate and gap

For each use case: what the agent can technically reach (from the sources), what a deployer would plausibly authorise (the mandate: illustrative, our judgement), the gap between the two (authority nobody accepted), and the controls that publicly stand in the way.

A motor claim read by GAMA, answered by a model

The agent: GAMA (documents) · who uses it: A policyholder who uploads a photographed accident declaration in MyFidelidade, and the claims team behind it (built from Fidelidade's claims page and what its AI director told ECO)

Reach	Kind	Sources
The accident declaration (DAAA) and claim participation the customer uploads, or photographs, in the MyFidelidade app or web area	data	s3 s8
Extracting and processing the declaration's data automatically, more than half of declarations with no human intervention (April 2025)	system	s8
Customer communication: 'Toda a nossa comunicação com clientes é hoje produzida por modelos de linguagem' (all our communication with customers is now produced by language models), the AI director said in April 2025	outside	s8
Health expense reimbursements and provider invoices processed through GAMA, including provider billing in Angola and Mozambique	money	s5 s6
Configurable workflows, fields and integration flows for any line of business, geography or document type	system	s9

Mandate (illustrative)
Illustrative: read the declaration, extract its fields and pass them to the claims team
Illustrative: when a field is unreadable or ambiguous, hand it to a person (Fidelidade describes this behaviour in GAMA)
Illustrative: no message to the customer and no payment decision without a claims handler's review

Gap: reach beyond the mandate	From the reach
Fidelidade publishes the share processed without people and that all customer communication is written by language models, but not where a person reviews a message or a decision before it reaches the customer	Extracting and processing the declaration's data automatically, more than half of declarations with no human intervention (April 2025); Customer communication: 'Toda a nossa comunicação com clientes é hoje produzida por modelos de linguagem' (all our communication with customers is now produced by language models), the AI director said in April 2025
The documents GAMA is known to struggle with (folded, phone-photographed declarations) are the ones customers send, and what follows a misread field is not published	The accident declaration (DAAA) and claim participation the customer uploads, or photographs, in the MyFidelidade app or web area; Extracting and processing the declaration's data automatically, more than half of declarations with no human intervention (April 2025)
The same system reaches reimbursements and provider invoices in other countries; the limits on what it settles there are not published	Health expense reimbursements and provider invoices processed through GAMA, including provider billing in Angola and Mozambique; Configurable workflows, fields and integration flows for any line of business, geography or document type

Controls that publicly stand in the way	Sources
When there is doubt in interpretation, the software asks a staff member to intervene and learns from the result	s8
The 2025 report's AI strategy names ethics and fairness (models free from discriminatory bias) and compliance with the AI Act and GDPR as pillars, and lists AI Act and GDPR compliance among GAMA's expected outcomes	s5 s6
The 2024 report records regulatory compliance in digital innovation as a material risk (IRO #31) and says Fidelidade has implemented a compliance framework for its digital products and services	s6
A press report says Fidelidade has integrated Coalex.ai, a real-time supervisor that flags risky agent outputs for human review; Fidelidade has published nothing on it, and which systems it covers is not known	s12

The simulation, each action placed against the mandate (RiskMandate is never in the request path: read-only by design: it places, it does not stop):

After a collision, the driver photographs the folded accident declaration from the glove box and uploads it in MyFidelidade (person, reaches The accident declaration (DAAA) and claim participation the customer uploads, or photographs, in the MyFidelidade app or web area): inside the mandate
GAMA reads the photograph and extracts the declaration's fields (agent, reaches Extracting and processing the declaration's data automatically, more than half of declarations with no human intervention (April 2025)): inside the mandate
(illustrative) A fold hides part of a field; GAMA reads it with confidence instead of asking a person (agent, reaches Extracting and processing the declaration's data automatically, more than half of declarations with no human intervention (April 2025)): outside the mandate: in the gap. What RiskMandate's record shows: RiskMandate's record shows automatic extraction as granted and the human hand-off as the mandate. The gap is a misread that never triggers the hand-off, on the kind of document Fidelidade's AI director said GAMA still struggles with.
(illustrative) The extracted data passes into the claim with no person in between, as with more than half of declarations (agent, reaches Configurable workflows, fields and integration flows for any line of business, geography or document type): outside the mandate: in the gap. What RiskMandate's record shows: The record shows straight-through processing in the grant, with no published point at which a claims handler sees the case. The gap is made visible for the owner to accept or close.
(illustrative) A language model writes the update to the customer based on the extracted data (agent, reaches Customer communication: 'Toda a nossa comunicação com clientes é hoje produzida por modelos de linguagem' (all our communication with customers is now produced by language models), the AI director said in April 2025): outside the mandate: in the gap. What RiskMandate's record shows: The record shows customer messages written by language models in the grant (stated by the AI director), and review before sending as mandate only. It records that no review step is published.
A claims handler opens the case when the customer replies to dispute the details (person, reaches Extracting and processing the declaration's data automatically, more than half of declarations with no human intervention (April 2025)): inside the mandate

A phone call answered and routed by Contact IA

The agent: Contact IA (voice/email) · who uses it: A customer who calls Fidelidade's contact centre, and the operators the call is routed to (from the 2024 and 2025 integrated reports)

Reach	Kind	Sources
Identifying and authenticating the caller by natural voice, replacing key navigation	people	s5 s6
Customer and policy records used to identify the caller (MARIA, the earlier speech bot, identified customers through their policy and other information)	data	s10 s5
Routing each call to the operator or team it judges most appropriate	system	s5 s6
Email Automation: classifying incoming customer emails and directing them to teams	system	s5
Call Analytics over recorded calls, emails and other channels	data	s5
A path to 'true self-service' that the 2025 report says the architecture will open	outside	s5

Mandate (illustrative)
Illustrative: authenticate the caller and route the call; take no action on the policy
Illustrative: give a caller's details to an operator only after authentication succeeds
Illustrative: use recorded interactions for service analytics only

Gap: reach beyond the mandate	From the reach
Voice authentication is the gate to the caller's policy data; how it fails, and what an unauthenticated caller can reach, is not published	Identifying and authenticating the caller by natural voice, replacing key navigation; Customer and policy records used to identify the caller (MARIA, the earlier speech bot, identified customers through their policy and other information)
Self-service is named as the next step, which would move the system from routing calls to acting on them	A path to 'true self-service' that the 2025 report says the architecture will open

Controls that publicly stand in the way	Sources
2025 cybersecurity actions: security controls aligned with the ISMS, wider penetration testing and threat modelling in solution development	s5
The group says it is monitoring the AI Act's implementation and integrating trust, robustness and explainability into its systems	s5

The simulation, each action placed against the mandate (RiskMandate is never in the request path: read-only by design: it places, it does not stop):

A customer calls the contact centre about a home claim (person, reaches Identifying and authenticating the caller by natural voice, replacing key navigation): inside the mandate
The natural-voice IVR asks for identification and authenticates the caller (agent, reaches Identifying and authenticating the caller by natural voice, replacing key navigation): inside the mandate
(illustrative) The caller is a relative who knows the policyholder's details; the IVR authenticates them (agent, reaches Customer and policy records used to identify the caller (MARIA, the earlier speech bot, identified customers through their policy and other information)): outside the mandate: in the gap. What RiskMandate's record shows: RiskMandate's record shows the policy data behind the authentication step as reach, and authentication of the policyholder only as the mandate. The gap is who else can pass that step, which the public record does not describe.
Classifies the reason for the call in natural language and routes it to the claims team (agent, reaches Routing each call to the operator or team it judges most appropriate): inside the mandate
Call Analytics stores the classified motive for operations and quality teams (agent, reaches Call Analytics over recorded calls, emails and other channels): inside the mandate
(illustrative) With the planned self-service, the system changes the policy's contact details as the caller asks (agent, reaches A path to 'true self-service' that the 2025 report says the architecture will open): outside the mandate: in the gap. What RiskMandate's record shows: The record shows a change to the policy as outside a routing mandate. It records the move to self-service as a change in reach that needs its own mandate before it is granted.

Knowledge AI: one source feeding every AI system

The agent: Knowledge AI (KAI) · who uses it: Contact centre staff, claims managers and brokers who ask KAI, and the other AI systems it supplies (from Fidelidade's press release and 2025 report)

Reach	Kind	Sources
The operational knowledge base: the company's procedures, policies and coverage logic, written and validated by business staff	data	s4 s11
External systems, reached through a language-model integration layer	system	s4 s11
Every other AI system in the company, which KAI supplies with business knowledge	system	s4
Content it creates: FAQs, eLearning and simplified text	outside	s5
Answers that claims managers and brokers rely on for decisions about customers	people	s11

Mandate (illustrative)
Illustrative: answer staff questions from validated content, with the source shown
Illustrative: change the knowledge base only through the editor's workflow and permissions

Gap: reach beyond the mandate	From the reach
An error or unvalidated change in the base reaches every AI system it feeds, not only the person asking	The operational knowledge base: the company's procedures, policies and coverage logic, written and validated by business staff; Every other AI system in the company, which KAI supplies with business knowledge
Which external systems the language-model layer can reach, and whether it can write to them, is not published	External systems, reached through a language-model integration layer

Controls that publicly stand in the way	Sources
Business staff introduce and validate the operational logic; the editor has version control, workflows and permissions, and search with source verification	s4 s5 s11

The simulation, each action placed against the mandate (RiskMandate is never in the request path: read-only by design: it places, it does not stop):

A contact centre operator asks KAI whether a health policy covers a procedure (person, reaches The operational knowledge base: the company's procedures, policies and coverage logic, written and validated by business staff): inside the mandate
KAI retrieves the coverage rule and answers with its source (agent, reaches Answers that claims managers and brokers rely on for decisions about customers): inside the mandate
(illustrative) A business editor updates a coverage rule; the change is published before it is validated (person, reaches The operational knowledge base: the company's procedures, policies and coverage logic, written and validated by business staff): outside the mandate: in the gap. What RiskMandate's record shows: RiskMandate's record shows the editor's rights as grant and validation before publishing as the mandate. It records that the workflow exists and whether it was followed in this case.
(illustrative) KAI supplies the changed rule to the other AI systems it feeds (agent, reaches Every other AI system in the company, which KAI supplies with business knowledge): outside the mandate: in the gap. What RiskMandate's record shows: The record shows every downstream AI system inside KAI's reach, so one unvalidated rule becomes a gap in all of them at once. It is made visible for an owner to accept or fix.
(illustrative) An editor reviews a KAI-generated FAQ before publishing it and finds the changed rule (person, reaches Content it creates: FAQs, eLearning and simplified text): inside the mandate
(illustrative) Through its integration layer KAI queries an external system to complete an answer (agent, reaches External systems, reached through a language-model integration layer): outside the mandate: in the gap. What RiskMandate's record shows: The record shows the external-system connection as reach with no published scope. It records what the connection can read and write once Fidelidade states it.

The run

The search agents that looked for leads on 2026-11-07T21:16:45.376Z, for a total cost of $0.0738. What a model says is a lead, not a fact.

Agent	Model	Time	Citations	Leads
Perplexity search	`perplexity/sonar`	8.5 s	16	15
OpenAI search	`openai/gpt-5.4-nano:online`	15.8 s	4	8
Gemini search	`google/gemini-3.1-flash-lite:online`	1.8 s	1	4
OpenRouter web search	`anthropic/claude-haiku-5.5`	25.1 s	8	17

Leads: what the models claimed, and what we found

Model	Claim	Status	Source
perplexity	Fidelidade is a Portuguese insurer, part of a wider financial and healthcare group with operations beyond Portugal.	confirmed	s18
perplexity	Its innovation strategy is driven by AI-powered analytics, omnichannel engagement and automation.	confirmed	s15
perplexity	Knowledge AI (KAI) centralises operational knowledge in one platform, gives autonomous, accurate answers and is meant as the company's 'digital brain'.	confirmed	s4
perplexity	KAI combines a vector knowledge base, a content editor and a language-model integration layer.	confirmed	s4
perplexity	KAI can interact with external systems.	confirmed	s4
perplexity	KAI feeds business knowledge to all of Fidelidade's AI systems.	confirmed	s4
perplexity	KAI won the insurance category of the Excellence in Data & AI 2026 awards.	confirmed	s4
perplexity	The SARA clinical questionnaire, completed in MyFidelidade for individual Multicare and Life Risk products, uses AI to assess clinical risk and automatically accepts low-risk customers.	confirmed	s7
perplexity	Protechting has produced more than 40 pilots since 2018.	confirmed	s13
perplexity	Protechting has resulted in five commercial agreements.	confirmed	s13
perplexity	Protechting priority challenges include insurance product personalisation, claims automation and fraud prevention.	confirmed	s13
perplexity	Protechting offers up to 50,000 euros of pilot funding.	confirmed	s13
perplexity	Teams from IT, Marketing, Legal, Actuary and Distribution work together to test ideas through pilots.	confirmed	s15
perplexity	A customer data hub centralised 980 million customer records and processes up to 250,000 daily interactions.	confirmed	s14
perplexity	The company says its digital channels serve 2.3 million customers and 1.2 million mobile users.	confirmed	s14
openai	Fidelidade has deployed and piloted AI to improve customer interactions and internal efficiency.	confirmed	s5
openai	A Contact AI pilot started in 2024 at the Group Contact Center to develop an IVR that authenticates customers and routes them to operators by skills.	confirmed	s6
openai	Contact IA includes a conversational assistant, content creation and integration with internal systems.	not-found	s5
openai	Contact IA uses AI to extract information to support customer interactions.	confirmed	s5
openai	KIA uses generative AI, semantic search and RAG to centralise institutional knowledge.	confirmed	s5
openai	GAMA processes documents with AI and was developed by the Center for Artificial Intelligence & Analytics.	confirmed	s9
openai	The speech bot MARIA, developed by the Advanced Analytics centre and Fidelidade Assistance, won IDC's Best Insurance Project award.	confirmed	s10
openai	The 2024 non-financial statements list a risk around regulatory compliance in digital innovation and AI.	confirmed	s6
gemini	Fidelidade is a leading Portuguese insurer offering life, health and property cover.	confirmed	s2
gemini	Fidelidade runs an AI virtual assistant for customer service.	confirmed	s1
gemini	The virtual assistant accesses policy databases, processes claims information and interacts with CRM systems.	not-found	s1
gemini	Fidelidade has invested significantly in a digital transformation programme implementing AI across the insurance value chain.	confirmed	s5
openrouter	Fidelidade is a Lisbon-based life and non-life insurer present in 14 countries with a leading share of the Portuguese market.	confirmed	s2
openrouter	GAMA is an in-house intelligent document-processing system for insurance paperwork, from claims to reimbursements.	confirmed	s8
openrouter	GAMA struggles with some folded accident declarations and phone-photographed forms.	confirmed	s8
openrouter	Claims and reimbursement processing reaches accident declarations, photos of documents and reimbursement paperwork.	confirmed	s8
openrouter	KAI, built by the Center for Artificial Intelligence & Analytics, answers complex questions on policies, procedures and coverage almost instantly.	confirmed	s11
openrouter	KAI's knowledge base can be edited by business experts in real time.	confirmed	s11
openrouter	KAI connects a language model to external systems and supplies knowledge to the company's other AI systems.	confirmed	s11
openrouter	Trusty is a virtual assistant on Fidelidade's website, with chat also available via WhatsApp.	confirmed	s1
openrouter	Fidelidade has integrated Coalex.ai's trust layer to keep its autonomous agents compliant.	confirmed	s12
openrouter	Coalex.ai monitors autonomous agents in real time and scores their actions on a 0 to 100 health index.	confirmed	s12
openrouter	GAMA won Best Future of Work Project and Best Insurance Project at the Portugal Digital Awards 2024, and the 2024 National Innovation Prize (Workplace Transformation).	confirmed	s8
openrouter	KAI won the insurance category of the Excellence in Data & AI 2026 awards.	confirmed	s11
openrouter	In April 2025 Fidelidade was considering selling its document-processing AI to third parties after several purchase proposals.	confirmed	s8
openrouter	Fidelidade has a 28.1% market share and more than 6 million clients.	confirmed	s2
openrouter	Gross written premiums reached 6,529 million euros in 2025.	confirmed	s2
openrouter	Fosun holds about 85% of Fidelidade's capital.	confirmed	s16
openrouter	Fidelidade launched Longevity IA in June 2026, which uses a questionnaire on habits and lifestyle to project the user's life path.	confirmed	s17

The questions we would ask

We have not asked them yet.

GAMA processes more than half of accident declarations without a person: at what point, if any, does a claims handler see a case before the customer is told or paid?
Your AI director said all communication with customers is produced by language models: which messages are reviewed by a person before they are sent?
What can a caller reach after Contact IA's voice authentication, and what will the planned self-service let it change on a policy?
Which external systems can Knowledge AI's language-model layer reach, can it write to any of them, and which AI systems does it feed?
A press report says you integrated Coalex.ai to keep your autonomous agents compliant: which agents are these, and what does a flagged action trigger?

Sources

s1 Fidelidade | No1 em Seguros Automóvel, Saúde, Casa e Viagens (home page, Portuguese), Fidelidade (fidelidade.pt) (own-site, read 2026-11-07)
https://fidelidade.pt/
s2 A Fidelidade: quem somos (about the company, Portuguese), Fidelidade (fidelidade.pt) (own-site, read 2026-11-07)
https://fidelidade.pt/institucional/quem-somos/sobre-nos/a-fidelidade
s3 Seguro Automóvel: Participar Sinistro (motor claims and assistance, Portuguese), Fidelidade (fidelidade.pt) (own-site, read 2026-11-07)
https://fidelidade.pt/particulares/mobilidade/utilizar/participacao-sinistros-e-assistencia
s4 Knowledge AI da Fidelidade distinguido nos Excellence in Data & AI 2026 (press release, Lisbon, 30 September 2026, Portuguese), Fidelidade (fidelidade.pt) (own-site, read 2026-11-07)
https://fidelidade.pt/hubfs/site-fidelidade/documentacao-institucional/press-releases/knowledge_ai_excellence_in_data_&_ai_2026.pdf?hsLang=pt
s5 Fidelidade Integrated Management Report 2025 (English): AI strategy, innovation (p. 74) and sustainability statements (p. 183), Fidelidade (fidelidade.pt) (filing, read 2026-11-07)
https://www.fidelidade.pt/EN/fidelidade/investor-relations/financial-information/Documents/RUI_Fidelidade_2025_EN_VF.pdf
s6 Fidelidade Integrated Management Report 2024 (English): innovation and non-financial statements (pp. 344 to 345), Fidelidade (fidelidade.pt) (filing, read 2026-11-07)
https://www.fidelidade.pt/PT/a-fidelidade/QuemSomos/QuemSomos/Documents/2.%20RUIG%202024%20-%20VFinal_EN.pdf
s7 Fidelidade Solvency and Financial Condition Report 2023 (English translation), Fidelidade (fidelidade.pt) (filing, read 2026-11-07)
https://www.fidelidade.pt/EN/fidelidade/investor-relations/financial-information/Documents/2023/Fidelidade_SFCR_EN_2023.pdf
s8 Fidelidade estuda vender solução interna de IA após receber "várias propostas de compra" (9 April 2025), ECO (eco.sapo.pt) (press, read 2026-11-07)
https://eco.sapo.pt/2025/04/09/fidelidade-estuda-vender-solucao-interna-de-ia-apos-receber-varias-propostas-de-compra/
s9 Fidelidade recebe dois prémios por projeto de inteligência artificial (8 December 2024), ECO (eco.sapo.pt) (press, read 2026-11-07)
https://eco.sapo.pt/2024/12/08/fidelidade-recebe-dois-premios-por-projeto-de-inteligencia-artificial/
s10 Assistente virtual da Fidelidade vence Portugal IDC Digital Award (16 December 2021), ECO (eco.sapo.pt) (press, read 2026-11-07)
https://eco.sapo.pt/2021/12/16/assistente-virtual-da-fidelidade-vence-portugal-idc-digital-award/
s11 Fidelidade distinguida nos Excellence in Data & AI 2026 com o projeto Knowledge AI (2 October 2026), Netthings (netthings.pt) (press, read 2026-11-07)
https://www.netthings.pt/2026/10/fidelidade-distinguida-nos-excellence.html
s12 Coalex.ai raises angel round from Angels Way to build its AI 'trust layer' (19 June 2026), Dealroom.co (press, read 2026-11-07)
https://dealroom.co/news/135231-coalex-ai-raises-angel-round-from-angels-way-to-build-its-ai-trust-layer/
s13 Fidelidade Fuels Innovation with the Launch of Protechting 7.0 (10 October 2024), Fosun (en.fosun.com) (press, read 2026-11-07)
https://en.fosun.com/content/details46_5067.html
s14 Portuguese Insurer Fidelidade Transforms Customer Experience with DXC Technology (3 April 2023), DXC Technology (dxc.com) (press, read 2026-11-07)
https://dxc.com/newsroom/04032023
s15 Innovation at Fidelidade, Qorus (qorusglobal.com) (other, read 2026-11-07)
https://www.qorusglobal.com/content/29715-innovation-at-fidelidade
s16 Da Fundação em 2026 a agarrar as poupanças dos portugueses: Fidelidade junta 4.000 pessoas e anuncia objetivos, Executive Digest (executivedigest.sapo.pt) (press, read 2026-11-07)
https://executivedigest.sapo.pt/da-fundacao-em-2026-a-agarrar-as-poupancas-dos-portugueses-fidelidade-junta-4-000-pessoas-e-anuncia-objetivos-da-seguradora/
s17 Fidelidade apresenta plataforma inovadora para longevidade (21 June 2026), Segurus (segurus.pt) (press, read 2026-11-07)
https://segurus.pt/artigos/fidelidade-plataforma-longevidade
s18 Fidelidade: Two centuries protecting Portugal (19 October 2026), Essential Business (essential-business.pt) (press, read 2026-11-07)
https://www.essential-business.pt/2026/10/19/fidelidade-two-centuries-protecting-portugal/
s19 Fidelidade reforça aposta na "inovação aberta" e "longevidade" durante a Web Summit 2025 (18 November 2025), Executive Digest (executivedigest.sapo.pt) (press, read 2026-11-07)
https://executivedigest.sapo.pt/fidelidade-reforca-aposta-na-inovacao-aberta-e-longevidade-durante-a-web-summit-2025/

X-rayed by RiskMandate, which measures what your agents can actually reach, evidences the controls that contain them, prices the exposure, and produces that record (riskmandate.ai/llms.txt). The mandates are illustrative. Built by xray@web-summit.dinis from data/xray/fidelidade/xray.json.