The AI operating system for media companies, brands, and research teams

Strategic intelligence,
grounded in your data and your market.

ChatDatak is a private decision-intelligence platform for media companies, brands, broadcasters, and research teams. Trained on your own knowledge, validated MediaDatak market research, your audience, and your competitive environment — not a generic AI chatbot.

No hallucinations · Source-grounded · Built for strategic decision-making · No credit card required for the public chat

Public free access

Ask the NYC Audience Twin

An experimental public intelligence lab exploring audience behavior, positioning, and competitive strategy across the New York radio market. Three free questions — no account required. Built to demonstrate what ChatDatak does for your own market.

  • Strategic depth — not generic AI summaries
  • Format psychology, audience drift, talent dynamics
  • Free account unlocks persistent sessions and deeper threads
Sample answer · NYC Audience Twin

“NYC CHR listeners aged 24–34 increasingly tolerate repetitive music more than repetitive personality patterns. The bigger source of fatigue is perceived emotional scripting. Listeners describe many morning shows as sounding professionally energeticrather than genuinely alive — creating a measurable drift toward passive streaming during high-cognitive-load moments like commuting.”

Proprietary · Psychologically precise · Difficult to fake
Free downloadable reports

Strategic research from our NYC lab

Drop your email and we’ll send you the full report. No spam — we publish a few of these per quarter and you can unsubscribe anytime.

📻
Coming soon

The State of NYC Music Radio 2026

A strategic snapshot of New York's music-radio landscape — format competition, audience drift, and emerging tensions across CHR, Hot AC, and Urban.

☀️
Coming soon

The Future of Morning Shows

Why morning shows are losing emotional bandwidth — and the structural shifts that will reshape day-part strategy over the next 24 months.

🎧
Coming soon

Streaming vs Radio: Emotional Value

Listeners increasingly use radio for emotional companionship, not discovery. A fresh look at the streaming/radio divide through audience-psychology research.

How it works

Built for serious enterprise teams, not consumer chatbots

Workspace isolation

Every brand, department, and client team gets its own ChatDatak instance with isolated retrieval, documents, conversations, and permissions. Programming, marketing, brand, and GM teams each see only what they’re authorized to see.

Source-grounded only

ChatDatak answers only from validated documents, MediaDatak studies, and workspace-verified facts. If the answer isn’t in the knowledge base, the platform says so — no fabricated conclusions, ever.

MediaDatak intelligence

Market studies, audience research, brand perception, programming recommendations, and competitor analysis — embedded as a trusted intelligence layer that improves with every study.

Why this isn’t just another AI tool

ChatDatak understands more than uploaded files

Most AI tools only understand uploaded documents and disconnected knowledge bases. ChatDatak understands the full strategic landscape your team operates in.

Your company
Your market
Your audience
Your competitors
Your strategic studies
Customer & listener behavior
Content & brand positioning
Programming & marketing strategy
Market evolution over time

For media companies

Programming directors, marketing leads, and GMs each get their own ChatDatak — connected to MediaDatak audience studies, ratings panels, and competitive positioning intel.

For brands & research teams

Strategy, brand, and insights teams operationalize their proprietary research, customer studies, and competitive intelligence — without it ever leaving their workspace.

Pricing

Strategic infrastructure, not a productivity tool

Essential
€3,500/month

For brands and single-station teams operationalizing their internal knowledge and MediaDatak intelligence layer.

  • ·1 ChatDatak workspace · 10 users
  • ·Internal knowledge AI
  • ·MediaDatak intelligence layer
  • ·Study uploads & competitor awareness
  • ·Strategic search/chat & dashboard
  • ·Basic analytics
  • ·Email support
Talk to sales
Most popular
Professional
€7,500/month

Multi-department, multi-market intelligence for broadcaster groups operating across formats, brands, and markets.

  • ·Multiple ChatDatak instances with department bots
  • ·25–50 users
  • ·Advanced permissions
  • ·Cross-study & multi-market intelligence
  • ·Strategic reporting & executive dashboards
  • ·Slack/Teams integrations
  • ·Priority support
Talk to sales
Enterprise
€15k–50k+/month

Dedicated infrastructure for multi-country broadcasters and media groups, with proprietary audience-modeling systems.

  • ·Unlimited workspaces · SSO · APIs
  • ·Dedicated infrastructure
  • ·Synthetic audience systems & digital twins
  • ·Forecasting models & executive simulations
  • ·Custom AI workflows
  • ·Dedicated strategist & onsite workshops
  • ·Multi-country intelligence
Talk to us

Want to see it before talking? Ask the public NYC Audience Twin — 3 free questions, no account required.

Security & Trust

Built for confidential strategic work.

ChatDatak handles broadcaster studies, internal strategy memos, and competitive intelligence. Below is what we actually do — not what we wish we did.

Your files are encrypted

AES-256 at rest across Postgres and Storage. TLS 1.2+ for everything in transit. HSTS enforced site-wide.

Your workspace is private

Tenant isolation is enforced inside the database (Postgres Row Level Security), not just the application. Embeddings live in per-workspace Pinecone namespaces.

LLMs don't train on your data

OpenAI and Anthropic API contracts forbid training on API content. We don't enable any data-sharing flags. The full sub-processor list is public.

You own your data

Delete a document and the row, chunks, raw file, parsed cache, and Pinecone vectors are all removed in the same operation. In-flight parsing is cancelled.

Security & Privacy FAQ

Straight answers.

Are my uploaded files secure?+
Yes. Files are stored in private Supabase Storage buckets in EU/Frankfurt, encrypted at rest with AES-256, and accessible only via short-lived signed URLs. Access is gated by Row Level Security on the database side — even a developer mistake in application code cannot leak files between workspaces.
Are my files encrypted?+
Yes. AES-256 at rest in Supabase Storage. Server-to-server traffic and browser-to-server traffic are TLS 1.2+ encrypted. The MediaDatak integration token is additionally encrypted at the application layer with pgcrypto.
Is my data encrypted at rest?+
Yes. Both Postgres and Storage are encrypted at rest with AES-256, managed by Supabase. This covers uploaded files, extracted markdown, embeddings metadata, prompts, responses, and conversation history.
Is my data encrypted in transit?+
Yes. TLS 1.2+ is enforced everywhere — browser to ChatDatak, ChatDatak to Supabase, ChatDatak to AI providers. HSTS with a two-year max-age is enabled site-wide.
Can LLMs train on my content?+
No. ChatDatak uses the OpenAI and Anthropic APIs, both of which contractually exclude API content from model training. We do not enable any data-sharing or store flags.
Is my data shared with third parties?+
Only with the sub-processors listed on the security page, each only for the specific purpose described there (hosting, parsing, embeddings, AI inference, email). ChatDatak ships no third-party analytics or trackers.
Can other users access my documents?+
No. Workspace membership is the access boundary. A user with no membership in your workspace cannot read your documents, conversations, or embeddings — enforced at the database via Row Level Security. Document chunks cannot be read directly via the database API at all; retrieval happens through server-side RPCs scoped to the caller.
What happens when I delete a file?+
In the same operation we remove: the Postgres row, every chunk row (cascade), the raw file in Storage, the parsed markdown cache, and every Pinecone vector for that document. In-flight LlamaParse jobs are cancelled. The deletion is written to activity_logs.
Are embeddings and extracted text also protected?+
Yes — they are treated as the same sensitivity tier as the source file. Extracted markdown lives in a private Storage bucket with the same RLS rules. Embeddings live in a per-workspace Pinecone namespace; queries are always namespace-scoped, so no cross-tenant similarity match is possible.
Is the platform suitable for confidential business documents?+
Yes — that is the design center. ChatDatak is built for broadcaster studies, audience research, strategy memos, and competitive intelligence, where a confidentiality incident is existential. Customers with stricter requirements (e.g. zero third-party processors for parsing) should ask about the Enterprise tier.

How does this know radio this well?

Find out by asking. Three free questions on the NYC Audience Twin — no account required.