AI Governance & Security Vendor Wiki

6 min read

SteelEye

One-liner — A London-based regtech that bundles trade surveillance, communications surveillance, recordkeeping/archiving and regulatory reporting onto a single cloud platform, pitched as one place to catch market abuse and meet records rules.

What it does

SteelEye is an integrated compliance platform for regulated financial firms. Its core jobs:

  • Communications surveillance — captures and monitors email, chat, voice, meetings and attachments, scanning for market abuse, MNPI/insider risk and conduct breaches.
  • Trade surveillance — monitors orders and executions for market-manipulation patterns (spoofing, layering, wash trades, front-running) against the EU/UK Market Abuse Regulation (MAR) and equivalent regimes.
  • Recordkeeping & archiving — WORM-style retention of communications and trade data, with eDiscovery and trade reconstruction (rebuilding the full comms + trade timeline around an event).
  • Best execution / transaction cost analysis and regulatory reporting (e.g. MiFIR).

The pitch is that holding comms, trade and reference data in one place lets surveillance correlate what was said with what was traded — the combination that actually surfaces market abuse and MNPI misuse, rather than two siloed tools.

Its AI layer, Compliance CoPilot (launched 2023-11), uses LLMs to triage and explain surveillance alerts across both trade and comms — ranking alerts, assigning a risk score, suggesting a resolution category and writing a rationale, with a vendor-claimed reduction in review time (marketing). It analyzes communications in 12 languages and returns insights in English. Note this is AI-assisted review of alerts, not (per available primary sources) capture of employees’ own generative-AI prompt/response traffic as a surveilled channel — see Open questions.

Where it sits in the stack

  • Category: comms-surveillance, in the governance layer.
  • This is a detective control, not a preventive one — it does “none-detection” (catching abuse/MNPI/conduct after the fact in logged comms and trades). It is a green-zone backstop: it watches the trusted internal environment for misconduct rather than blocking untrusted input or egress.
  • It is not an inline prompt/egress control; its value is regulatory and conduct oversight, not data-exfiltration prevention.

Deployment & architecture

  • Cloud / SaaS data platform; data-ingestion- and API-led. Consolidates fragmented sources (comms channels, order/execution feeds, reference data) into a single store for surveillance, reporting and reconstruction.
  • Integrates with communications-capture vendors (email, Teams/Slack/Zoom, mobile/voice, chat) and ingests trade/order data to align comms with trades.
  • SOC 2 and ISO certified (vendor-stated). UK entity registered in England and Wales (#10581067); offices in London, New York, Singapore, Bangalore and Braga.

Positioning & differentiators

SteelEye’s distinguishing claim is the single integrated platform spanning trade surveillance + comms surveillance + recordkeeping + reporting, versus point tools you have to stitch together. How it differs from neighbors:

  • vs behavox and shield — both are comms-surveillance / conduct specialists leaning heavily on AI/NLP; SteelEye spans the wider compliance stack (trade surveillance, reporting, recordkeeping) rather than being comms-first.
  • vs nice-actimize — NICE Actimize is the incumbent enterprise surveillance/financial-crime suite (tier-1 banks); SteelEye positions as a more consolidated, lower-friction option aimed at mid-size firms.
  • vs theta-lake — Theta Lake focuses on capture/compliance for modern collaboration tools (video, voice, UC); SteelEye covers comms surveillance but as part of a broader trade+records platform.
  • vs relativity-trace — Relativity Trace is comms-surveillance built on Relativity’s eDiscovery lineage; SteelEye couples comms with native trade surveillance and reporting.

Ownership, funding & M&A

  • Independent, VC-backed. Founded 2017 (London); CEO and co-founder Matt Smith (ex-Bloomberg, ex-Noble Group CIO). Other co-founders reported: Matthew/Matt Storey (CPO), David Haines (CTO), Shankar Vasudevan, Chris Young.
  • Series B: $21M, announced 2022-09, led by Ten Coves Capital, with Fidelity International Strategic Ventures, Illuminate Financial, Beacon Equity Partners and a large family office; total raised ~$43M. Eight Roads Ventures also cited as an investor (aggregator data).
  • No acquisition found as of 2026-06-28 — ownership remains independent. Confidence: medium (corroborated press + vendor; exact cap table not public).

CTO / hedge-fund lens

  • Day-2, but hedge-fund-critical. Comms + trade surveillance and recordkeeping are regulatory obligations under MAR (EU/UK), and SEC / FINRA / CFTC / NFA records and supervision rules in the US — including the SEC/CFTC off-channel communications (“WhatsApp”) enforcement wave that has hit asset managers and advisers. A fund of any size handling MNPI needs some surveillance + recordkeeping answer; this is the layer.
  • SteelEye is frequently pitched to mid-size firms that want one consolidated platform rather than a tier-1 stack (NICE Actimize) plus separate capture/archiving tools. That consolidation can be attractive to a lean compliance team.
  • The AI angle (Compliance CoPilot) is mainly an efficiency play — fewer analyst hours per alert — relevant if alert volume is the pain point. Treat the “90%” figure as vendor marketing.
  • Diligence points for a CTO: which comms channels you actually capture upstream (SteelEye surveils what it ingests), data residency, and whether you need trade surveillance at all (a pure long-only fund’s MAR exposure differs from a multi-strategy/active trader).

Competitors / alternatives

behavox · shield · nice-actimize · theta-lake · relativity-trace

Open questions / to verify

  • Does SteelEye capture employees’ generative-AI tool usage (prompt/response) as a monitored surveillance channel? Compliance CoPilot is AI assisting alert review; no primary source confirms AI-prompt capture as a channel. Flag — likely conflation in the brief.
  • Exact post-Series B cap table / control; whether any funding has occurred since 2022.
  • Precise list of native trade-surveillance model coverage and which capture vendors are pre-integrated.

Sources

  • SteelEye — About us — fetched 2026-06-28 — supports: HQ London + offices, platform capabilities (trade/comms surveillance, recordkeeping, MAR, best execution, reporting, trade reconstruction), cloud platform, leadership, SOC2/ISO; confidence: med (vendor/marketing).
  • SteelEye raises $21 million in Series B funding — bobsguide — fetched 2026-06-28 — supports: Series B $21M (2022-09), lead Ten Coves Capital, investors, ~$43M total, “London-based SaaS regtech”; confidence: high.
  • SteelEye introduces AI-driven Compliance CoPilot — fetched 2026-06-28 — supports: Compliance CoPilot launch 2023-11, LLM alert triage across trade+comms, 12 languages, Matt Smith quote; confidence: med (vendor/marketing).
  • Corroborating (WebSearch, not separately cached): Tracxn / Crunchbase — founded 2017, founders, Eight Roads Ventures as investor; confidence: med (aggregators).

History

  • [2026-06-28] Stub created from seed registry.
  • [2026-06-28] Researched; established founded 2017 (London), CEO/co-founder Matt Smith, independent VC-backed (Series B $21M 2022-09 led by Ten Coves Capital, ~$43M total — no acquisition), SaaS integrated trade+comms surveillance + recordkeeping + reporting platform with MAR coverage and AI “Compliance CoPilot” alert triage. Set ownership_confidence medium, status researched, hedge_fund_fit medium. Flagged that AI-prompt/response capture (vs AI-assisted alert review) is unconfirmed.

Graph View

Backlinks