AI infrastructure funding shifts toward specific enterprise workflows
AI infrastructure funding describes investment rounds backing technical platforms that supply core data, billing, security, and workflow capabilities other products depend on, instead of building standalone consumer applications or general-purpose models. In the latest wave, four startups together raised USD 109.5 million (approx. RM503.7 million) across payments infrastructure, AI billing, identity verification AI, and litigation-focused tools. Their products share a pattern: they target precise bottlenecks in how enterprises run and monetize AI, from payment routing to fraud checks and case analysis. Rather than chase headline-grabbing chatbots, investors are betting on infrastructure that quietly powers usage-based pricing, risk decisions, and complex legal processes. This signals a maturing market where value lies in dependable data layers, programmable workflows, and domain-specific automation that can slot into existing stacks and scale globally.
Primer turns unified payments infrastructure into an AI decision layer
Primer, a unified payments infrastructure startup, secured a USD 100 million (approx. RM460 million) Series C to build what it calls an AI-enabled operating layer for payments and finance. The company sits across the full payments lifecycle, from checkout to payout, capturing over 400 data points per transaction and handling more than 95% of customer payment volume on average. That unified, contextual dataset underpins Primer Companion, an AI agent evolving from an insights tool into one that can execute payment decisions autonomously within rules merchants define. According to Primer, processing billions of transactions annually for brands such as GetYourGuide and Dialpad shows how fragmented processors and acquirers limit reliable AI decisions. Its expansion plans include lifting US revenue from about one-fifth to more than one-third of the business, highlighting how payments infrastructure startups now compete on data completeness as much as on routing or uptime.
Flexprice builds an AI billing platform for usage-based monetization
Flexprice is positioning itself as an AI billing platform for companies whose pricing depends on tokens, API calls, GPU usage, and other real-time workloads. The startup raised USD 1.5 million (approx. RM6.9 million) in seed funding to grow its open-source billing infrastructure for AI-native and API-first enterprises. Its system already processes over 20 billion events per month and has seen 6x revenue growth in the last quarter, with event volumes increasing twentyfold over the past year. Flexprice argues that billing is the most error-sensitive layer of the AI stack, and that usage-based and outcome-based models now outpace flat subscriptions. The new capital will fund expansion into the US and Europe and support new finance products for metering, revenue recognition, and financial reporting workflows, showing how investors see billing infrastructure as a foundation for full revenue automation in the AI economy.
Didit and Crimson show demand for identity and litigation-native AI
Identity verification AI and legal infrastructure are also drawing investor interest. Didit raised an additional USD 6 million (approx. RM27.6 million) in seed funding, bringing its total to USD 7.5 million (approx. RM34.5 million), to expand programmable identity and fraud infrastructure. Its API-first tools connect to government data sources and use AI to analyze more than 200 signals, including document checks, biometric liveness, injection and deepfake detection, and behavioral patterns, serving over 1,500 customers across 220+ territories. In parallel, litigation-focused startup Crimson closed an oversubscribed USD 2.5 million (approx. RM11.5 million) seed round and opened a New York office. Crimson’s platform ingests entire case files and structures people, events, arguments, and deadlines into a queryable layer, betting that complex disputes cannot be handled well by generic legal AI assistants. Together, these raises underline demand for enterprise AI tools tuned to high-risk, high-context domains.
Why investors prefer domain-specific AI infrastructure over general models
Across these deals, a clear pattern emerges: investors now prioritize AI infrastructure that embeds into specific enterprise workflows instead of funding broad, general-purpose models. Primer’s focus on complete payments data, Flexprice’s event-level billing, Didit’s programmable identity stack, and Crimson’s litigation-native case layer all reflect this shift. Each company builds a high-fidelity data foundation and then layers AI decision-making on top, anchoring automation in concrete outcomes such as higher payment approval, cleaner revenue reporting, safer onboarding, or faster case strategy. This approach reduces the risk of AI making unreliable decisions on fragmented or shallow inputs. For enterprises, it means AI arrives not as a vague assistant, but as a set of reliable tools wired directly into payments, billing, identity verification, and legal workflows—an architecture that is likely to define the next phase of AI infrastructure funding.
