Mercury launches conversational AI Command across platform

Mercury launched Mercury Command, a conversational AI interface for its financial platform, enabling users to perform tasks like checking cash positions and sending invoices via natural language. The feature requires explicit customer confirmation before executing actions and only accesses authorized data, rolling out to over 300,000 business and personal customers.

Mercury launches conversational AI Command across platform PYMNTS reports that Mercury has added a conversational AI interface called Mercury Command to its financial platform, per a company release. Mercury Command can perform tasks including checking a user's cash position, changing auto-transfer rules, categorizing transactions and sending invoices, according to the release cited by PYMNTS. The feature requires explicit customer confirmation before executing any action and can only access data each user is authorized to see, per the release. PYMNTS also reports that Mercury began rolling out the feature to business and personal customers on Tuesday and that the platform serves more than 300,000 customers. Mercury Co-Founder and CEO Immad Akhund has described Mercury's goal as building financial software that helps businesses run rather than simply holding deposits. What happened PYMNTS reports that Mercury added a conversational AI interface called Mercury Command to its financial platform, per a company release. The release, cited by PYMNTS, says Command lets users interact in natural language and see the platform complete tasks with their approval. Mercury Co-Founder and CEO Immad Akhund has described the company's goal as building financial software that helps businesses run rather than simply holding deposits, according to PYMNTS reporting. Product capabilities reported Per the release reported by PYMNTS, Mercury Command can: - •check a user's cash position - •change auto-transfer rules - •categorize transactions - •send an invoice Access and safety controls reported The release noted that Command can only see and act on what each user is authorized to access and that it requires the customer's explicit confirmation before executing actions, according to PYMNTS. PYMNTS also reports that Mercury began rolling out the feature to business and personal customers on Tuesday and that the company's platform serves more than 300,000 customers. Editorial analysis - technical context Conversational, action-capable interfaces like Command typically sit atop a large language model or ensemble of models combined with deterministic business-logic layers that map intent to platform actions. Industry-pattern observations include use of retrieval-augmented generation for grounding answers in account data, intent classification models to disambiguate commands, explicit confirmation flows to convert intent into state-changing API calls, and audit logs to capture what the assistant recommended and what the user approved. Practitioners will watch how providers chain LLM outputs to deterministic validators to reduce risk of incorrect actions. Industry context Industry observers note fintechs and banks increasingly experiment with natural-language front ends to reduce UI friction for routine workflows. For operations teams and ML engineers, these interfaces raise the same integration questions as other embedded LLM features: data access policies, latency trade-offs when calling backing systems, explanation and provenance for automated recommendations, and guardrails against hallucinations. Mercury's requirement for user confirmation before execution reflects a cautious approach to agentic financial workflows. What to watch Indicators to follow include adoption rates among Mercury's customer segments, metrics on confirmation versus suggestion how often users approve suggested actions , auditability and logs exposed for compliance, latency and error rates when Command invokes platform APIs, and vendor disclosures about model choice and data retention. These signals will determine how usable and auditable action-capable conversational layers are in regulated financial settings. Scoring Rationale Notable product launch for fintech practitioners: action-capable conversational interfaces change integration and compliance requirements for teams building safe, auditable LLM-driven workflows in regulated financial settings. Mercury's 300,000+ customer base and cautious human-confirmation approach make this a practical reference deployment. Practice with real FinTech & Trading data 90 SQL & Python problems · 15 industry datasets Active Verified Users by Income TierEasy /problems/sql/active-verified-users-by-income Technology Stocks with High BetaMedium /problems/sql/technology-stocks-with-high-beta Portfolio Performance ScorecardHard /problems/sql/portfolio-performance-scorecard 250 free problems · No credit card See all FinTech & Trading problems /problems/datasets/fintech