What problem does it solve?
In a bank or insurer the knowledge that staff need to do their job correctly is scattered across credit and risk policies, operating procedures, product manuals, compliance guidance and internal research: many documents in several systems, each with versions. Keyword search returns a list of PDFs. Staff ask a colleague, use an outdated copy, or give a customer a wrong answer.
Retrieval augmented assistants change the interaction: ask a question, get an answer with the paragraphs it came from. The hard problems are not the model. They are permissions (an answer must never come from a document the person may not see), currency (the answer must come from the version in force), and trust (people must be able to check the source quickly). Done well, one governed retrieval layer can serve the service desk, HR, frontline and specialist assistants instead of each building its own.
How does it work?
- Ingest and index approved sources. Policies, procedures and research are ingested from the document management system, intranet and knowledge base, with owner, version and access rights kept as metadata.
- Retrieve with permissions. When an employee asks, retrieval runs only over documents their role and entitlements allow, combining semantic and keyword search.
- Answer with citations. The model answers only from the retrieved passages and cites each document and section, so the employee can open the source.
- Refuse when unsure. If retrieval finds nothing relevant or sources conflict, the assistant says so and points to the owner, rather than guessing.
- Learn from gaps. Unanswered and badly rated questions go to content owners, who fix or write the missing document.
- Audience
- Employee facing
- Autonomy
- Assist
- Adoption
- Mainstream
- Channels
- Internal tools, Microsoft Teams, Agent desktop
What is it worth?
Benchmarks are computed from the public deployments below: one data point per organization per KPI, with who made each claim.
| KPI | Median | Reported range | Data points | Claimed by |
|---|---|---|---|---|
| Employee adoption | Too few to pool | 80% to 90% | 2 | 1 organization, 1 vendor |
| Search time reduction | Too few to pool | 90% to 95% | 2 | 2 organization |
| Interactions handled | Not pooled | at least 23 million | 1 | 1 organization |
| Productivity gain | Too few to pool | 15% | 1 | 1 organization |
Value drivers: Employee productivity, Speed and cycle time, Compliance quality, Customer experience.
Indicative value
A bank with 5,000 employees who regularly look up policies and procedures
USD 2.3 million to USD 18 million
Employee time released per year
How this is calculated
Formula: employees * searchHoursPerWeek * workingWeeks * timeSaved * hourlyCost. The low scenario uses every low input, the high scenario every high input.
| Input | Low | High | Basis |
|---|---|---|---|
| Employees who search internal knowledge regularly employees, employees | 5,000 | 5,000 | The reference organization. |
| Hours per week each spends finding and reading internal information searchHoursPerWeek, hours per employee per week | 2 | 4 | Editorial assumption. Replace with a time study of your own staff. |
| Share of that time saved timeSaved, fraction of search time | 0.1 | 0.25 | Editorial assumption, replace with a time study of your own. No source on this page reports a measured share of search time saved. For comparison, Google Cloud reports that information searches by less experienced SIGNAL IDUNA agents are 30% faster (read as speed, that is about 23% less time per search, since 1 / 1.3 is about 0.77), and that Wells Fargo's tool reduced the workflow for query resolution by about 20%, without saying whether that is time. |
| Working weeks per year workingWeeks, weeks | 45 | 45 | Editorial assumption. |
| Fully loaded cost per hour hourlyCost, USD per hour | 50 | 80 | Editorial assumption. Replace with your own blended cost. |
What it leaves out: Released time, not cash. It leaves out the value of fewer wrong answers to customers and fewer policy breaches, which is often larger, and the cost of cleaning up and maintaining the content.
Who already uses it?
13 public deployments, strongest evidence first. Grades: A regulator or audit, B the organization itself, C vendor case study, D anonymous or estimate.
Cathay Financial Holdings
Taiwan · Banking · 2026
Cathay Financial Holdings (Cathay FHC) announced a long term strategic collaboration with OpenAI to deploy ChatGPT Enterprise across its subsidiaries and to build an enterprise grade AI workstation that integrates OpenAI's large language models with Cathay FHC's own internal knowledge bases and proprietary intelligent agents. The workstation gives employees an AI assistant for document summarization and rewriting, data synthesis, presentation generation, operational inquiries and natural language code generation, connected to existing enterprise systems. Cathay FHC says it is the first financial institution in Taiwan to work with OpenAI at this scale, and is separately building agentic AI systems for corporate banking and insurance with OpenAI's API.
No outcome disclosed.
All Nippon Airways
Japan · Travel and hospitality · 2025
All Nippon Airways rolled out neoAI Chat, a generative AI platform built on its secure private cloud, to every ANA Group company and department involved in aircraft operations, including airport, maintenance, cabin and flight operations, from October 2025. The platform analyses internal documents such as regulations and meeting minutes so staff can retrieve information in plain language instead of searching manually, and the rollout followed a trial at Narita Airport that started in August 2024. ANA says the goal is to speed up access to complex regulatory and operational knowledge and free staff time for higher value work.
- Search time reduction: about 90%
"Rapid Knowledge Retrieval: Retrieve critical information from documents like regulations and meeting minutes, reducing access time by roughly 90% compared to conventional methods."
Claimed by: organization
DZ BANK AG
Germany · Banking · 2025
DZ BANK built a GenAI platform that lets any department build and run its own chatbot, grounded only in documents the department uploads. Staff submitted more than 200 AI ideas; some run on the bank's general assistant, DZ Chat, while most need the new platform's department specific chatbots, which can use up to 100 documents each, read tables and images, and are provisioned through a self service ordering flow within days. DZ Chat will also use the new platform in future.
No outcome disclosed.
Bank of America
United States · Wealth and asset management · 2024
Merrill and Bank of America Private Bank teams use ask MERRILL and ask PRIVATE BANK, built on the technology behind Erica, to curate the information they need for clients. For more complex requests, the chat can connect teams with experts at the bank. The bank reports more than 23 million interactions with the two tools in 2024.
- Interactions handled: at least 23 million, calendar year 2024
"In 2024, there were more than 23 million interactions with ask MERRILL and ask PRIVATE BANK, an increase of 1 million over 2023, helping employees more proactively connect with clients about timely and relevant opportunities."
Claimed by: organization
Zain Jordan
Jordan · Telecommunications · 2024
Zain Jordan built a chatbot on OpenAI's large language model that lets employees query the company's contracts, legal agreements and business policies in natural language, with optical character recognition added so scanned paper documents are searchable too. The project began in the first quarter of 2024 with a pilot in the legal department, then rolled out to three more business units by the third quarter of 2024. Zain Jordan reported large drops in document search time and external legal spend, plus adoption above nine in ten users in the teams it reached.
- Search time reduction: about 95%, 2024, document related operational workload
"Task Automation: Reduced manual document searches and external legal consultations for routine queries resulted in a 95% decrease in document-related operational workload."
Claimed by: organization - Cost savings: about USD 3630, per year, external legal consultations for routine queries
"Legal Consultation Costs: An estimated 25% reduction in reliance on external legal consultations for routine queries, leading to annual savings of approximately $3,630."
Claimed by: organization - Productivity gain: 15%, 2024, Legal and Compliance teams
"The platform enables employees to focus on higher-value tasks, contributing to a 15% increase in productivity metrics for the Legal and Compliance teams."
Claimed by: organization - Employee adoption: at least 90%, 2024, teams the platform reached
"The platform scaled from the Legal Department to three additional business units, with over 90% adoption across relevant teams."
Claimed by: organization
Morgan Stanley
United States · Wealth and asset management · 2023
Morgan Stanley Wealth Management fully rolled out the AI @ Morgan Stanley Assistant in September 2023, a generative AI chatbot that gives Financial Advisors quick access to the firm's intellectual capital. The rollout followed the firm's March 2023 announcement of OpenAI as its strategic partner. In its June 2024 release the firm said that 98% of Financial Advisor teams had adopted it. It was followed by AI @ Morgan Stanley Debrief, which drafts meeting notes and follow up emails with client consent.
No outcome disclosed.
PLDT
Philippines · Telecommunications · 2026
PLDT, the Philippines' largest digital services provider, built KAI (Knowledge, Automation, Intelligence) on the UiPath Platform to replace a manual research process that could take up to five days with an agentic assistant that answers employees in one to three seconds, grounded only in approved sources. Teams use KAI to cut repetitive research for proposal creation and customer support, freeing staff for higher value, consultative work instead of information gathering. PLDT deployed KAI alongside two related UiPath backed systems, Ellie for broader workflow productivity and ERICA for risk reviews, as part of the same AI transformation.
- Response time reduction: 99%, since deployment, reported September 2026
"This enables a 99% faster response rate, increases organizational capacity without additional headcount, and allows teams to shift from repetitive information gathering toward higher-value, consultative work."
Claimed by: vendor - Hours saved: up to 30,000 hours, per year
"Powered by an agentic AI assistant and grounded exclusively in approved sources, KAI eliminates repetitive research for proposal creation and customer support, creating approximately 12 FTE-equivalent annual productivity capacity and reducing an estimated 25,000 to 30,000 hours of manual effort annually."
Claimed by: vendor
Zillow
United States · Real estate · 2026
Zillow's permanent shift to remote work early in the pandemic left knowledge scattered across video calls, online documents and cloud data shares, with no single place to find an answer. It deployed Glean to connect more than 30 data sources and 138 million documents into one search and AI platform, giving employees a single source of truth for people, projects and company operations, and it now underpins specialized agents for HR, UX research, performance reviews and incident documentation. Zillow's own senior vice president of engineering credits the platform with helping teams work closer together, and the vendor reports fast, wide staff adoption alongside weekly search time saved per employee.
- Employee adoption: 80%, as of the case study
"80%adoption across 7,000 employees"
Claimed by: vendor
Booking.com
Netherlands · Travel and hospitality · 2025
Booking.com juggled numerous workplace applications with stale data and inefficient information access, while meeting strict security and GDPR requirements. It deployed Glean as a company wide search and AI platform, letting employees query their workplace applications in natural language instead of hunting across tools by hand. IT technicians who used to spend up to ten minutes per ticket finding the right information now resolve tickets in little to no time. Glean became the first AI platform adopted company wide at Booking.com, which the vendor says scaled AI to 14,000 employees.
No outcome disclosed.
Raiffeisen Bank International AG
Austria · Banking · 2025
Raiffeisen Bank International (RBI), headquartered in Vienna, built its own internal assistant, RBI ChatGPT, on Azure OpenAI Service and Azure AI Search within Azure AI Foundry, moderated with Azure AI Content Safety. Employees use it to search and access internal information faster, draft customer summaries and emails, answer regulatory questions, and automate repetitive tasks such as line levelling in reconciliation reporting and summarizing public news articles and financial reports. A feature called "yellowGPTs" lets employees contribute their own knowledge bases and solutions on top of the platform.
- Users served: at least 20,000
"The RBI ChatGPT was first deployed to 2,000 users and now has an active user base of over 20,000."
Claimed by: vendor
SIGNAL IDUNA
Germany · Insurance · 2025
SIGNAL IDUNA, a German insurer, built Co SI with Google Cloud, BCG and Deloitte: a knowledge assistant that helps customer service agents answer complex health insurance questions. Google Cloud reports that for less experienced agents, information searches are 30% faster and inquiries that previously needed further escalation dropped from 27% to 3%.
No outcome disclosed.
Wells Fargo
United States · Banking · 2025
Wells Fargo deployed a retrieval augmented tool for branch bankers that finds the relevant policies and procedures during customer interactions. Google Cloud reports that it reduced the workflow for query resolution by about 20%, without saying whether that means time, steps or effort. The bank uses reusable APIs on Apigee to scale generative AI across teams.
No outcome disclosed.
Insurance Australia Group
Australia · Insurance · 2024
IAG's CGU brand built CASI (Claims Assistant Supporting Intermediated), a generative AI tool trained on IAG's content so claims handlers in the intermediated (broker) business can ask it questions in natural language about coverage terms in up to 1,000 different product disclosure statements, and get an answer with references to trace where the information came from. CGU began scaling CASI during the second half of 2024 and it is now used by around 500 front end claims staff, with IAG extending it to other divisions and claims teams.
- Users served: about 500
"That's rolled out to 500 or so front end claims people in intermediated and it's proven to be so powerful that now it's actually going out across other divisions as well and other claims teams,"
Claimed by: organization
How do you implement it?
A model agnostic playbook: what to prepare, the order to build in, and what goes wrong.
Data you need
- An inventory of authoritative sources with an owner and review date per document
- Access rights per document or collection that can be carried into the index
- A set of real questions per domain with expected answers, for evaluation
Systems to integrate
- Document management and intranet (for example SharePoint or Confluence)
- Identity provider and entitlement data for permission aware retrieval
- The channels where employees work (Teams, the agent desktop, the intranet)
- Feedback routing to content owners
Complexity: Medium
A prototype over a folder of PDFs is quick to build. Production takes longer: connecting several document systems, carrying access rights into the index, handling versions and retirement, and building evaluation sets per domain so answer quality can be measured.
- 1
Start with one domain and its owners
Pick a domain with heavy lookup volume and willing owners, such as operations procedures or product terms. Clean its documents before indexing anything.
- 2
Carry permissions into retrieval
Index access rights with every chunk and filter at query time. Test with accounts of different roles that restricted content never appears.
- 3
Build the evaluation set
Collect a few hundred real questions with expected answers and sources, and run them on every change to content, retrieval settings or model.
- 4
Make citations the product
Show the source passage next to the answer with a link. Staff trust and adopt tools whose answers they can check in seconds.
- 5
Close the loop with content owners
Send unanswered questions and negative feedback to owners weekly, and retire documents that are superseded.
- 6
Offer it as a shared layer
Expose the same governed retrieval to the other assistants (service desk, HR, contact centre) so permissions, residency and versions are enforced in one place.
Guardrails
- Permission aware retrieval, tested with role based test accounts
- Answers only from retrieved passages, with citations, and refusal when nothing relevant is found
- Only the version in force is indexed; superseded documents are removed
- Prompt injection defences for content from shared or external sources
- Query logs protected and retained according to policy
KPIs to instrument
- Answer accuracy and citation correctness on the evaluation set, per domain
- Share of questions answered versus refused
- Weekly active users among target employees
- Time to find information in a time study, before and after
- Negative feedback and content gaps closed per month
Human in the loop
Content owners are accountable for their documents and review flagged answers. Employees remain responsible for decisions they take on the basis of an answer, and high impact decisions (credit, compliance, customer remediation) still follow their documented approval steps.
Common failure modes
- Oversharing through search
- The assistant surfaces documents that were technically accessible but never meant to be widely read. Review permissions before indexing, not after an incident.
- Confident answers from old versions
- Superseded policies stay in the index. Index only the version in force and track effective dates.
- Answers without sources
- Staff cannot verify and either distrust the tool or trust it blindly. Always show the cited passage.
- Many point solutions
- Every department builds its own index with its own permissions. Build one governed layer and reuse it.
What are the risks and rules?
EU AI Act
Limited risk (transparency)
Article 50(1) requires that people who interact directly with an AI system are informed of it, unless this is obvious from the context, as it usually is for an internal assistant. The system would be high risk only if it were intended for an Annex III purpose, such as assessing the creditworthiness of natural persons (point 5(b)) or making decisions on or evaluating workers (point 4(b)).
Rules that apply
Guidance
- Artificial Intelligence (AI) Model Risk Management, information paper (Monetary Authority of Singapore, Asia Pacific). Good practices for AI and generative AI model risk management observed in a thematic review of banks in mid 2024, covering governance and oversight, risk management systems and processes, and development and deployment.
- OWASP Top 10 for LLM Applications (OWASP Gen AI Security Project, Global). The 2025 list covers prompt injection (including through retrieved content), sensitive information disclosure and vector and embedding weaknesses, the main security risks of retrieval assistants.
- Article 50, transparency obligations for providers and deployers of certain AI systems (European Union, Europe). Providers must design AI systems that interact directly with people so that those people are informed they are interacting with AI, unless this is obvious from the context. Applies from 2 August 2026.
Controls to put in place
- Inventory entry with an accountable owner and the list of indexed sources
- Permission tests per role before each new source is added
- Evaluation set runs on every change to content, retrieval or model
- Document ownership and review dates enforced for indexed content
- Monitoring of refusals, negative feedback and unusual query patterns
When it went wrong elsewhere
- CVE-2025-32711: AI command injection in Microsoft 365 Copilot. A vulnerability recorded by NVD in June 2025, not a reported breach: AI command injection in Microsoft 365 Copilot allowed an unauthorized attacker to disclose information over a network. It shows that an enterprise assistant can be made to disclose information through injected instructions.
Frequently asked questions
- How is this different from the search we already have?
Search returns documents; the assistant returns an answer with the paragraphs it came from. Google Cloud reports that Wells Fargo's retrieval tool for branch bankers reduced the workflow for query resolution by about 20%, and that information searches by less experienced SIGNAL IDUNA service agents are 30% faster.
- Will employees actually use it?
Two wealth managers have published usage figures. Morgan Stanley said in June 2024 that 98% of its Financial Advisor teams had adopted its AI @ Morgan Stanley Assistant. Bank of America reports more than 23 million interactions in 2024 with ask MERRILL and ask PRIVATE BANK, a volume figure that does not say what share of employees use the tools.
- How do we stop it from showing confidential documents?
Carry each document's access rights into the index and filter at query time, then test with accounts of different roles. Review what is technically accessible before you index it, because an assistant makes forgotten oversharing easy to find.
How to cite this page
Blits.ai AI Use Case Library, "AI enterprise knowledge search for employees", last verified 27 September 2026, https://www.blits.ai/ai-use-cases/enterprise-knowledge-search. Licensed under CC BY 4.0. Method: how we verify use cases.
Changelog
- 1 October 2026: Added deployments: PLDT, All Nippon Airways, Cathay Financial Holdings, Insurance Australia Group.
- 30 September 2026: Added deployments: DZ BANK AG, Raiffeisen Bank International AG.
- 30 September 2026: Added deployments: Zain Jordan.
- 29 September 2026: Added deployments: Booking.com, Zillow.
- 27 September 2026: First published