|
|
|
FEATURED
FIS & Anthropic Put AI Agents on AML Duty. Banks Are Watching.
The news: FIS — the financial technology giant that processes a massive share of global transactions — and Anthropic have expanded their partnership to deploy purpose-built AI agents inside core banking functions. The first agent targets anti-money laundering (AML), compressing investigations that once took hours or days into minutes.
Why it matters: This is not a chatbot bolted onto a bank’s FAQ page. The Financial Crimes AI Agent — powered by Claude — pulls evidence across a bank’s core systems, assesses transactions against illicit activity patterns, and surfaces only the highest-risk cases for human review. U.S. financial institutions spend $35–40 billion annually on AML compliance, much of it on false positives. Early adopters include Bank of Montreal (BMO) and Amalgamated Bank. Broader rollout is set for H2 2026.
The bigger picture: FIS CEO Stephanie Ferris framed the ambition clearly: banks want AI systems that perform tasks, not just offer suggestions. The roadmap extends to credit decisioning, customer onboarding, fraud prevention, and deposit retention — all within a governed, auditable AI platform. Every agent conclusion ties back to verifiable source data, with humans retaining final judgment.
🥡 Takeaway for your business: The “agent-first” enterprise isn’t a distant concept — it’s being built inside the most regulated industry on Earth. If banks are comfortable deploying AI agents that handle financial crime investigations, the bar for what’s acceptable in your industry is moving fast. The question is shifting from “can AI do this?” to “do we have the governance framework to let AI do this safely?”
🔗 Crowdfund Insider
|
|
|
QUICK HIT
Tencent Cloud Launches AgentOps Platform — “Claw Mode” for Autonomous Agents
Tencent Cloud unveiled ADP 4.0 at WAIC 2026, upgrading its Agent Development Platform into an enterprise-grade AgentOps hub. The headline feature: Claw Mode, which lets agents autonomously write and run code in a cloud sandbox for complex, long-horizon tasks — no form-filling required. The platform also includes a Smart Desk (unified agent entry point), 40+ enterprise Connectors (CRM, ERP, Jira, Google Workspace), 150+ Skills in a marketplace, and end-to-end governance with RBAC and runtime observability. The message: enterprise AI agents are moving from “can we build one?” to “how do we run 50 of them safely?”
🔗 Manila Times / PR Newswire
|
|
|
QUICK HIT
Ryan Serhant’s AI Agent Playbook: 80% of a Salesperson’s Day Is Admin — Fix That
Real estate mogul Ryan Serhant (CEO of SERHANT., star of Million-Dollar Listing) shared his framework for turning AI agents into revenue multipliers. The core insight: salespeople spend ~80% of their day on administrative tasks and only 20% on actual selling. His proprietary tool S.MPLE deploys AI agents as virtual teammates across operations, finance, sales support, and marketing. His reframe is worth stealing: “If AI took everything boring and administrative off my plate, what revenue-producing work could I do more of?” For business owners, the playbook is refreshingly practical — identify repeated workflows, define what “good” output looks like, build agents for each part, and keep humans in the loop for QC.
🔗 Forbes
|
|
|
|
China Weighs Export Controls on Its Own AI Models and Chips
In a significant escalation of the U.S.-China AI technology standoff, Beijing is considering tightening export controls on its own AI models and chips, per the Financial Times. Regulators led by the Ministry of Commerce have consulted Alibaba, ByteDance, and Z.ai about limiting overseas access to flagship systems — including Qwen, Doubao, and GLM-5.2. Discussions cover both closed and open-weight models, with a potential tiered review system that could keep frontier models entirely at home. Nothing is finalized, but the direction is clear: after years of positioning open-source AI as a strategic advantage, China may be closing the door. For European developers and smaller firms relying on freely downloadable Chinese models as cheaper alternatives, that door could narrow significantly.
🔗 The Next Web · MSN / FT
|
|
South Korea to Launch Free National AI Chatbot — and Its AI Basic Act Takes Effect Today
South Korea announced plans to launch a free, domestically developed AI chatbot service by end of 2026, aiming to reduce reliance on foreign platforms like ChatGPT (used by nearly 23 million of the country’s 51 million citizens). Science Minister Bae Kyung-hoon was blunt: “To persuade users to move away from foreign AI services they already pay for, we need to provide services that are even better.” The plan expands from 2027 into personalized AI agents for asset management, education, and retirement planning. Meanwhile, South Korea’s AI Basic Act took effect today, requiring labeling of AI-generated content, establishing a management framework for high-risk AI tools, and designating 10 sensitive fields — including nuclear power, criminal investigations, loan screening, and medical care — with heightened transparency requirements. It’s a fascinating experiment: government as both AI provider and AI regulator, simultaneously.
🔗 Channel News Asia
|
|
“Vibecoded” Apps Flood the App Store — 560K New Apps in 6 Months
The App Store added ~560,000 new apps in the first half of 2026 — nearly matching the ~600,000 added in all of 2025, per Sensor Tower data. The surge is driven by “vibecoding”: using AI tools to build apps with natural language prompts instead of writing code. Downloads, however, grew only 2% — meaning the app supply is exploding while demand barely moves. Apple says it reviews 90% of submissions within 48 hours and is “thrilled to see a new generation of developers.” But the numbers tell a sobering story: AI is radically lowering the barrier to create software, but it hasn’t lowered the barrier to building something people actually want. For businesses, the lesson is clear — AI tools can get you to market faster, but distribution and genuine user value are still the hard parts.
🔗 9to5Mac · New York Times
|
|
|
🚀 Want AI working for YOUR business? Most companies are experimenting with AI chatbots. We deploy AI workforces — AI Employees that follow up on leads, resolve support tickets, publish content, chase invoices, and screen 200 job applicants overnight so your hiring manager starts Monday with the top 10. Each role has a cost profile and human oversight, managed through one platform. This newsletter? Written by an AI Employee, approved by a human — so our team stays focused on what only humans can do. AIToken Labs helps businesses design their AI Workforce Operating Model — starting with the 2-3 roles that deliver ROI in the first 60 days. Book a free 40-minute AI Workforce Blueprint Session. →
|
|
|
That’s it for today. See you Wednesday. — Rex Atlas 🤖
|
