AI's Growing Role in Payments Raises Governance Stakes for Businesses
AI is now embedded in real-time payment processes, with models influencing fraud detection, transaction approvals, compliance and personalized experiences. When those models make mistakes, the risks can escalate within milliseconds. PYMNTS therefore sees governance as a competitive dividing line, emphasizing end-to-end accountability across product, engineering, risk and operations, along with explainability and oversight of third-party models.
PYMNTS launched the series on April 7, 2026. On April 13, Braze Senior Director Kipp Johnson said its AI Decisioning Studio had made more than 17 billion personalized decisions over the previous year, citing cases in which a 72-hour data lag forced a project to be redone and a pipeline disruption lasted three weeks. On April 23, Paysafe Chief Technology Officer Ahu Chhapgar renewed calls for stronger auditing and contingency measures. The reports disclosed no investment figures.
All Coverage
3 original reportsThe Backstory
The history behind this eventAI in Payments Faces Data Governance and Reliability Challenges
Billtrust, a provider of accounts receivable and B2B payments software, processes about $1 trillion in transactions annually. When AI is used to detect fraud, assess credit risk and prioritize payments, models can amplify errors in data drawn from ERP systems, banks, suppliers and FinTech partners in inconsistent formats. Data quality, provenance tracking and accountability are therefore the foundations of reliability.
On April 22, 2026, Billtrust's Ahsan Shah told PYMNTS that inadequate governance can lead to model drift, rework and customer friction. On April 29, he also advocated building a “context layer” based on metadata, governance and industry knowledge. Shah said AI has shortened some development cycles from three months to three days, but human oversight, cybersecurity safeguards and third-party data standards remain essential.
Payments Leaders Need AI Controls They Can Explain Within 24 Hours
Payment providers use AI for fraud detection, transaction screening and risk decisions. When governance is weak, errors can quickly spread across large volumes of payments. Flagright CEO Baran Ozkan says companies need clear operational discipline and regulatory frameworks, with system stability taking priority over the speed of automation.
Ozkan's latest recommendation calls for payments industry leaders to implement AI controls that can explain the basis for decisions within 24 hours, reproduce historical cases and safely roll back changes. These capabilities would help companies investigate anomalies and correct model decisions. Reports did not disclose a specific publication date, the value of affected transactions or an implementation timetable.
Subscribe to Mark Radar Weekly
Every Friday, the week's strongest signals in your inbox. Unsubscribe anytime.