Bloomberg: Initiation of Coverage

GAII Intelligence defines the AI Research & Intelligence market as software that applies generative and agentic AI to the research process of professional investors: locating candidate securities, assembling and interrogating evidence, producing written analysis with traceable sources, and monitoring positions once held. Buyers are hedge funds, long-only managers, multi-manager platforms, family offices and allocators; the economic buyer is typically the chief technology officer, the head of research, or a portfolio manager with a technology budget.

Capabilities we treat as mandatory

• Evidence retrieval at scale: the ability to search and reason across regulatory filings, transcripts and news, and across documents the client supplies.
• Source traceability: every material claim in an output must be reachable back to the passage that supports it.
• Institutional data handling: a stated position on where client data goes, how long it is held, and whether it is used for model training.

Capabilities we treat as differentiating

• Persistent state: the system remembers a thesis, a coverage list or a position between sessions and acts on it without being re-prompted.
• Workflow surfaces: delivery inside the applications analysts already use — the spreadsheet, the research management system, email and messaging.
• Owned or exclusive content: material that a competitor cannot obtain by signing the same third-party contract.
• Published evaluation: a measured, disclosed accuracy methodology rather than an assertion of quality.

This note rates Bloomberg's AI offering for investment professionals — ASKB and the Terminal's AI features, Enterprise MCP and the data those features can reach — not the Terminal or the company as a whole. Bloomberg competes with FactSet, LSEG Workspace and S&P Capital IQ Pro among incumbents, and with AlphaSense, Hebbia and Rogo for AI-led research. Its strategy is to keep the agent inside the Terminal; peers are distributing data into the assistants clients already use.

The following are the capabilities and characteristics that, in our assessment, most distinguish Bloomberg within the AI Research & Intelligence market. Each is drawn from publicly documented product capability, published interviews with management, or independent hands-on review.

• Content depth: Bloomberg News, 400m+ documents, 800+ entitled research providers, Third Bridge transcripts, alternative data and 100m+ securities in one agent.
• Citation discipline: Page- and paragraph-level citations, BQL shown for data points and deterministic validators on summaries.
• Engineering scale: 400+ AI engineers, published research since BloombergGPT (2023) and leadership in MCP standards work.
• Client context: ASKB reads a client's IB chats, Bloomberg Drive files and worksheets (October 2026).
• Reach: ASKB is included in the Terminal seat, so its AI reaches users without a separate purchase.
• Commitment: A 17-year AI programme with quarterly generative AI releases since 2024.

The following are factors that prospective clients should weigh, and areas where our public review did not find sufficient evidence to reach a settled view. Several are common across vendors at this stage of the market and are not specific to Bloomberg. We encourage prospective clients to raise each of them directly with the company; we have done the same, and will update this note as answers arrive.

• Closed ecosystem: No Bloomberg connector for Terminal users in Claude, ChatGPT or Copilot; Enterprise MCP requires a Data License Plus contract.
• Disclosure: No published accuracy metrics, AI-specific certifications, trust centre or model-provider list.
• MNPI questions: Ingestion of IB chats and uploads raises information-barrier questions that published controls do not yet answer.
• Status: ASKB's general availability is described differently by Bloomberg and press (October 2026).
• Deliverables: No native deck or memo generation; Excel output through BQL only.
• Commercial terms: Unpublished seat pricing with two-year auto-renewal and regular price rises.

Bloomberg has applied machine learning to news and data since 2009 sentiment models, and in 2023 published BloombergGPT, a 50-billion-parameter finance model that is not used in products today. Its AI work is led from the CTO office, with 400+ AI engineers and a research group publishing at NAACL and FAccT.

Generative AI reached the Terminal through earnings-call summaries (2024), news summaries and document insights (2025), Document Search & Analysis (beta from mid-2025) and ASKB (beta from February 2026), with Enterprise MCP for data-licence clients in September 2026. ASKB reads Bloomberg News, filings, entitled broker research, Third Bridge transcripts and, since October 2026, a client's IB chats, files and worksheets.

Company-reported figures are shown as such and are not independently verified. Verification is carried into the vendor questionnaire.

Bloomberg delivers AI inside the Terminal and, for data clients, through Enterprise MCP. Figure 2 sets out our assessment of how it maps onto the investment workflow.

ASKB

Agentic assistant across news, research, documents, data and client context, with Workflows for briefs and thesis checks (beta from February 2026).

Document Search & Analysis

Search and comparison across 400m+ documents with re-runnable tables.

Summaries and insights

Earnings-call, news and document summaries with source links (2024–25).

BQuant Assistant and Enterprise MCP

Coding agent for quants; MCP access to 50,000+ data fields for Data License Plus clients (September 2026).

Scope of the product today

We did not locate a connector for Terminal users in third-party AI assistants, native deck generation or published AI accuracy metrics.

ASKB is built on several commercial and open-weight models plus smaller in-house models, routed by cost and reliability; Anthropic is the one provider Bloomberg has named. Specialised agents for news, filings and data run in parallel, and retrieval weighs sources by type, recency and perspective.

Bloomberg publishes peer-reviewed research on finance models and AI risk but no accuracy figures for ASKB.

Content spans Bloomberg News (about 5,000 stories a day), 400m+ company documents, research from 800+ providers subject to entitlement, Bloomberg Intelligence, Third Bridge transcripts, alternative data and estimates across 100m+ securities.

Since October 2026 ASKB can also use a client's IB chats, uploaded files and worksheets. No content-licensing deal with an AI platform was located.

Bloomberg fits the full analyst day inside the Terminal: screening, research, monitoring and quant work.

Delivery outside the Terminal is the gap: no Office-side AI and no connector into the assistants clients increasingly use.

Bloomberg states that client content is not used to train models without consent, that model providers are contractually barred from training on client data, and that AI features follow existing entitlements. Figure 3 sets out what our public review located.

We would encourage buyers to request AI-specific security documentation, model providers, audit-log capability and MNPI controls for IB chat ingestion.

The offsetting factors are entitlement enforcement and the operational record of the Terminal.

Bloomberg does not publish prices; third parties report about $32,000 per seat a year with two-year terms. No separate AI price has been announced.

For existing users, the AI is effectively included in the seat.

Bloomberg sits in the Emerging quadrant of the matrix, within 0.45 of both boundaries (X 7.07, Y 7.23); open distribution or published AI documentation would move it toward Leaders.

GAII Research & Intelligence coverage plus the Established Platforms group (Bloomberg, FactSet, LSEG, S&P Global MI, MSCI), 21 October 2026. Established Platforms are rated only on their AI offering for investment professionals and the data it can reach — not on company size or non-AI products.

11.1 Where Bloomberg is strongest relative to peers

• Against AI-native vendors, on proprietary news and data.
• Against other incumbents, on citation discipline and engineering depth.
• Against document tools, on breadth of entitled research.

11.2 Areas of narrower functional focus

Bloomberg has built depth inside the Terminal rather than openness or disclosure.

• Openness: no Terminal connector to third-party AI.
• Disclosure: no AI metrics or certifications.
• Deliverables: no decks or memos.
• Commercial: unpublished pricing.

Scores are on a 0–10 scale. Each dimension score is the mean of its sub-criteria, and the GAII Overall Score is the weighted mean of the six dimensions. The detailed sub-criteria scores that build each dimension are set out in Appendix A.

These are the areas where our public review did not find sufficient evidence to form a settled view. We have put them to the company and will update this note when they are answered. Prospective clients may find them a useful agenda for their own evaluation call.

• ASKB general-availability status and coverage of user types.
• Model providers used by ASKB.
• Plans for a Terminal-user connector in Claude, ChatGPT or Copilot.
• AI-specific security certifications and trust documentation.
• Audit-log capability for AI use.
• MNPI controls for IB chat ingestion.
• Use of opted-in client content for evaluation.
• Accuracy evaluation for ASKB.
• Pricing of AI features in future.
• Native deck and memo generation plans.
• Broker-research provider count (800+ or 1,200+).
• Successor to BloombergGPT.

This note is an assessment of a vendor against a defined framework, not a recommendation to buy or to decline. The following are the considerations we would put at the centre of an evaluation of Bloomberg specifically.

If you are a fundamental equity manager

• Test ASKB on a live thesis and compare citations.
• Check which broker research your seat entitles.
• Use Workflows for monitoring.

If you are responsible for technology or vendor risk

• Open with Figure 3; request AI security documentation.
• Review IB chat ingestion with compliance.
• Plan for AI tools outside the Terminal separately.

If you are assessing the category rather than the company

• Incumbents are building agents over proprietary data.
• Openness to third-party assistants now differentiates incumbents.
• Citation discipline is becoming table stakes.

Each dimension score in Section 12 is the mean of the sub-criteria below.