Question 1

What makes Aleta Index unique?

Accepted Answer

Three things nobody else does together:
1. We score at the claim level, not the publisher level. A source can be reliable on politics and unreliable on biotech — Aleta captures that. Static publisher labels can't.
2. We use actual events as ground truth. Sentiment vendors read tone. LLM-faithfulness tools verify text against retrieval. Aleta tests each claim against how reality actually reacted to the claim's own timeframe.
3. We sit at the start of the AI pipeline, not the end. We filter noise and score credibility before content reaches your model, agent or workflow — saving compute and improving output quality.

AI-native, MCP-distributed, plug-and-play with any LLM stack.

Question 2

How are you different than fact checkers?

Accepted Answer

Fact-checkers assess whether a specific claim is true or false — usually retrospectively, by humans, on a small set of claims.
Aleta does something different. We don't declare truth vs. falsehood. We score how well every source has predicted reality over its own stated timeframe — automatically, at scale, across thousands of claims per day. A source that's frequently right earns a high credibility score; one that's often wrong, or vague, or slow, doesn't.
Fact-checkers grade individual claims. Aleta scores track records — live, at scale, and specifically for the finance domain.

Question 3

Do you check article by article?

Accepted Answer

No — we go deeper, and we do it fully automatically. Every article is decomposed into atomic claims (e.g. "NVDA's data-centre revenue will grow 40% this quarter" is one claim; "NVDA is losing share to AMD" is another).
Each claim is tagged with its source, ticker, topic and stated timeframe, then tested against how the market actually moved over that horizon. Right earns credit; wrong takes a penalty. Aggregated up, this gives you a live credibility score per source × ticker × topic × time — with full provenance receipts.
One article can contribute to dozens of scores, and one source's credibility can vary across the tickers and topics they cover.

Question 4

Who uses Aleta?

Accepted Answer

Aleta is built for institutional finance and enterprise AI teams:
• Hedge funds and asset managers — filter alt-data feeds, weight signals by source credibility, defend research decisions with provenance receipts.
• Investment banks and trading platforms — power research desks, e-trading and institutional data products with a trust layer under raw content.
• Enterprise AI teams — plug credibility scores into RAG pipelines, agent workflows and LLM retrieval to prevent hallucinated or unreliable outputs.
• Fintech and analytics platforms — embed credibility scoring into your own products via API or MCP.
If your team is spending on alt-data, LLM tokens or model infrastructure, Aleta cuts noise and increases trust across the stack.

Question 5

Can I bring my own data?

Accepted Answer

Yes. Aleta is designed to work with the sources you already ingest — and to score any custom feeds you bring in.
Score your own sources. Point Aleta at proprietary research, internal notes, industry newsletters or private analyst feeds, and we'll score their track record against the market alongside your public sources.
Enterprise deployment. For customers hosting Aleta in their own Google Cloud project (VPC), your data stays inside your environment — Aleta's engines run against it without any content leaving your perimeter.
Custom sourcing lists. Enterprise plans include per-account source curation, so your credibility scores reflect the specific universe your teams actually read.

Question 6

What is the value for me?

Accepted Answer

Three concrete gains:
• Better decisions. Weight LLM retrieval and alpha signals by source credibility. Your models and analysts stop treating noisy sources as equal to reliable ones.
• Less wasted compute. Aleta filters ~60% of noise at ingestion — that's ~60% fewer tokens embedded, retrieved and reasoned over on your AI stack. On enterprise budgets, that's material.
• Defensible research. Every score ships with provenance receipts. When you have to explain to a PM, a client or compliance why you weighted a source the way you did, you have the evidence.
More signal, less noise, and a receipt for every call.

Question 7

What data sources does Aleta cover today?

Accepted Answer

Aleta is live across 109 US-listed tickers, ingesting from news, filings, research notes, Twitter and Reddit. Coverage expands throughout 2026 to include UK and European equities, Asia-listed tickers, and additional source categories (analyst reports, earnings transcripts, macro feeds).
We can also stand up custom coverage for enterprise customers who need specific tickers, sectors or geographies added to the roadmap.

Question 8

How do I access Aleta?

Accepted Answer

Four options, depending on how you work:
• REST API — direct calls, standard integration.
• MCP server — plug directly into AI agents, LLMs and MCP-native tools. Available on Smithery, Glama and GitHub registries.
• Streaming feeds — real-time delivery for trading and monitoring workflows.
• VPC / on-premises deployment — run Aleta inside your own Google Cloud project or private infrastructure; content never leaves your perimeter.
Test live scoring across the 109 covered tickers at aletaindex-narrative.com.

Credibility infrastructure for AI and organisations

Efficiency. Quality. Trust.

Efficiency

Quality

Trust

One credibility infrastructure. Three engines.

Relevance Filter

Narrative Intelligence

Credibility Index

Plugs into your stack. Doesn't replace it.

Try Narrative Intelligence

Frequently asked questions