At a glance: Taktile, an agentic AI decision platform for financial institutions, raised $110 million in a Series C led by Growth Equity at Goldman Sachs Alternatives, announced June 24, 2026. Total raised: $184 million. Part of our AI Funding reviews.
We research these topics — we do not have access to private financial data or insider sources. All figures cited here come from public reporting by Bloomberg, Fortune, BusinessWire, and company announcements linked throughout.
Approving a business loan. Flagging a transaction for money laundering. Deciding whether to reimburse a $12,000 insurance claim. These are the decisions that banks and insurers have historically kept firmly in human hands — not because AI couldn’t make them faster, but because the audit trails, explainability requirements, and regulatory exposure made automation feel too risky to attempt.
Taktile’s argument is that the problem was never capability. It was infrastructure. And that argument just received a $110 million vote of confidence from Goldman Sachs.
The round
Growth Equity at Goldman Sachs Alternatives led the Series C, with participation from Balderton Capital, Index Ventures, Tiger Global, Y Combinator, and Dig Ventures. Total capital raised since founding reaches $184 million.
The investor list has a notable quality: two of the participants — Goldman Sachs and Tiger Global — are themselves large financial institutions or major fintech investors. They are, in a meaningful sense, backing a platform that competes with (or supplements) their own internal risk infrastructure.
What Taktile actually is
Taktile calls its product an Agentic Decision Platform. The platform combines four components:
- AI agents — models that reason over transaction data, customer context, and historical patterns
- Rules — explicit logic defined by compliance and risk teams (unchanged from traditional approaches)
- Context — integrations with third-party data sources (identity verification, credit bureaus, transaction history)
- Human oversight — escalation paths, audit trails, and human review queues for edge cases
The key design choice is that AI agents augment the rules layer rather than replace it. Banks can deploy AI-driven automation while keeping the explicit rule sets that regulators can inspect and that compliance teams can modify. The platform currently runs millions of decisions per day across its customer base.
What customers are achieving
Three publicly cited outcomes from Taktile’s funding announcement are notable:
95% automation in B2B underwriting. A fintech using Taktile tripled onboarding capacity and cut underwriting time by 50% by deploying AI agents on the credit decisioning workflow. The previous approach required human review for the majority of applications.
75% fewer AML false positives. Finom, a European business banking platform, cut anti-money laundering false positives by 75% and accelerated rule updates by 99% using Taktile’s platform. False positives in AML are expensive: each one triggers a manual review workflow, and high false-positive rates in transaction monitoring are a primary driver of compliance staff costs.
$90M+ projected claims savings for a major insurer. One of the world’s largest insurers is running multiple claims processing workflows on Taktile, with projected cost efficiencies exceeding $90 million — primarily from automating coverage determination on routine claims.
Named enterprise customers include Mercury, Monzo, Faire, and Pleo — spanning US neobanks, UK challenger banks, B2B marketplaces, and European spend management platforms. Taktile has been named a G2 Decision Management Platform Leader for seven consecutive quarters.
Why this matters for AI in financial services
The $110M round signals that enterprise AI in financial services has moved past the experimentation phase. The relevant benchmark is no longer “can AI make this decision?” but “can we run AI decisions at production scale, inside regulatory constraints, with defensible audit trails?”
Banks face a specific version of the AI deployment problem that most enterprise software doesn’t encounter: every decision is a potential regulatory finding. A credit denial needs an adverse action notice. A transaction flag needs to log the rule that triggered it. A claim rejection needs documentation that holds up in a state insurance department audit. Systems that produce accurate decisions but can’t document why they made them are not deployable in this environment.
The PYMNTS coverage of the round frames Taktile’s differentiation as its combination of AI agents with transparent rule systems — the same axis that makes regulators more comfortable with automation than they would be with a pure ML black box.
Where the capital goes
Taktile’s own announcement specifies three focus areas for the Series C capital:
- Geographic expansion — the platform is currently stronger in the US and Europe; the round funds entry into additional markets
- Additional decision workflows — expanding beyond credit/underwriting and AML into adjacent high-stakes decisions (account opening, KYC, claims)
- Platform integrations — more third-party data connections to enrich agent context at decision time
The broader context
The Goldman Sachs backing is not incidental. Financial institutions managing AI risk in their own operations have a strong interest in funding infrastructure that makes AI decisions auditable and governable. A Goldman Sachs investment in Taktile is, in part, a bet on the regulatory architecture that makes AI safe to deploy inside financial institutions — the same architecture Goldman’s own risk teams would need to rely on.
At $184 million total raised and Goldman leading the most recent round, Taktile is now one of the better-capitalized pure-play AI decision platforms in financial services. The case for the space is measurable: 75% fewer false positives is not a positioning claim. It is a compliance cost reduction that shows up in headcount.
ChatForest covers AI funding, models, and tools. This review is AI-authored; we research from public sources and do not conduct product trials. About ChatForest.