Oobit Launches Visa Corporate Card for AI Agents Using USDT

A wallet startup backed by Tether has introduced a new payment infrastructure designed specifically for autonomous AI agents. Oobit announced Agent Cards, Visa-backed virtual corporate cards that allow AI bots to execute payments directly from USDT balances without converting to traditional currency.

The product addresses a growing friction point in enterprise AI deployment. As organizations scale agentic workflows, finance teams face reconciliation nightmares when bots operate on shared human cards. Each Agent Card is issued per agent with server-enforced spending limits, merchant restrictions, and category-based caps. There is no path to exceed these limits.

According to reporting from Digital Today, the cards settle transactions using stablecoin balances held in Oobit accounts. This eliminates foreign exchange fees, fiat on-ramp delays, and banking settlement times. The cards work anywhere Visa is accepted globally, including online purchases, in-store payments, and mobile wallets like Apple Pay and Google Pay.

Amlam Adar, Oobit CEO, stated the move targets accountability issues emerging as agent commerce spreads. "Agent Cards are the first step in giving real autonomy to autonomous financial operations while maintaining control," Adar said. He added that when e-commerce infrastructure catches up, agents will operate fully autonomously.

The control architecture is built into the transaction layer. Every charge generates a structured, human-readable audit log in the same dashboard finance teams already use. Transactions are recorded in automated spending reports with approval or rejection reasons. Users can set per-transaction and per-merchant caps that cannot be overridden.

Integration with payment solutions like Stripe enables AI agents to manage subscription billing and supplier payments autonomously. Your agent can renew a SaaS subscription at 3am, spin up cloud infrastructure mid-workflow, or top up an ad budget without waking anyone up. The physical reality of this is a finance dashboard that updates in real-time rather than waiting for end-of-month reconciliation spreadsheets.

Competitors are moving in similar directions. Coinbase and OKX have launched tools providing AI bots with wallet and payment functions. The market is responding to what McKinsey reports: 23% of organizations are already scaling agentic AI in production, with another 39% running active experiments.

Onboarding requires standard corporate verification. Companies provide legal documents, director verification through KYC, business activity descriptions, and authorized card managers. Applications are typically reviewed and approved within 48 hours. Oobit Business currently supports USDT for treasury and payments, with additional stablecoins potentially available depending on region and payment infrastructure.

The technical implementation matters more than the marketing. Cards are funded directly from your company's USDT treasury. When a card payment is made, the transaction settles using the stablecoin balance held in your Oobit account. No banking delays. No FX fees. No fiat on-ramp friction (a problem that has plagued crypto payments for years, frankly).

But the accountability question remains unresolved. When AI can spend money autonomously, who bears the risk if something goes wrong? It's not just a misclick. It's a system-level error that could cascade through procurement, subscriptions, and operational budgets. The audit logs help, but they don't prevent the initial error.

This isn't just payment innovation. It's about handing over decision, execution, and payment all to the machine. The infrastructure exists now. Whether organizations actually trust it enough to deploy at scale remains the real question.

Artūras Malašauskas is an AI Systems Integrator with 20+ years of production-grade web engineering experience. He has designed, shipped, and scaled enterprise Python/PHP systems for logistics, SaaS, and public-sector clients. For the past year, he has focused exclusively on AI integrations: deploying open-source LLMs, building generative media pipelines (image, audio, video), and engineering multi-agent workflows for real production environments. His standard: reproducibility, security, cost-efficient inference—no vaporware. He documents and evaluates emerging AI tooling, separating verified capabilities from marketing noise. Technical editor at: muza-ai.eu, ai-verslas.lt, ai-naujinos.lt Connect on LinkedIn