Infosys’ business process management arm, Infosys BPM, has added AI agents for invoice processing to its Infosys Accounts Payable on Cloud solution. This solution makes invoice processing largely autonomous to enhance efficiency and accuracy. The solution includes the capabilities of the suite of generative AI services, solutions and platforms called Infosys Topaz, as well as Microsoft’s AI stack, including Azure AI Foundry and other LLMs. By integrating Infosys Topaz with a purpose-built multi-agent framework, along with Microsoft’s AI stack, we’ve developed a solution that is autonomous by design, responsive to change, and built to evolve, Anantha Radhakrishnan, CEO and managing director of Infosys BPM, said. Infosys developed the AI agents in collaboration with Americana Restaurants, which operates more than 2,600 restaurants across the Middle East, North Africa and Kazakhstan. “With AI-powered Infosys Accounts Payable on Cloud, we have made invoice processing faster, enhanced accuracy and improved efficiency,” Harsh Bansal, chief financial officer and chief growth officer at Americana Restaurants, said. “The addition of Agentic AI takes this a step further, reducing manual dependencies and bringing more intelligence and autonomy into our invoice processing.”
CloudZero’s platform breaks down cloud spending data by source to display the costs incurred by each of a company’s cloud environments, workloads and business units and provides granular data associated with specific application features
Startup CloudZero has raised $56 million from investors to enhance its namesake software platform, which helps enterprises lower their cloud expenses. Companies search for opportunities to lower cloud costs by analyzing their infrastructure- and software-as-service bills. Each cloud provider formats spending data differently, which means that the data has to be organized into a single, consistent format before it can be reviewed. CloudZero says that its platform automates the task to save time for finance teams. After normalizing spending data, the company’s software breaks it down by source. CloudZero can display the costs incurred by each of a company’s cloud environments, workloads and business units. It also provides more granular information. The platform can calculate, among others, the cloud expenses associated with specific application features. CloudZero says that its platform can provide visibility into Kubernetes spending regardless of label quality. CloudZero can measure how effectively a company uses long-term purchase agreements. It also identifies spending spikes. Built-in artificial intelligence models measure out the average hourly cost of running a cloud workload and detect sudden increases. CloudZero now manages more than $14 billion in cloud spending for customers. It claims that those customers take an average of three months to make back what they spend on its software.
Thomson Reuters tool for tax and audit operations draws on >20 billion proprietary and public documents and 4,500 subject matter experts and connects internal knowledge, regulatory materials and firm-specific data into a single workspace
Thomson Reuters has launched an AI platform designed to help professionals automate complex workflows. The platform — Agentic Intelligence — integrates with professional applications to plan, reason and act across tasks while maintaining audit trails and data controls. The launch also includes a new tool, CoCounsel for Tax, developed for tax, audit and accounting professionals. It connects internal knowledge, regulatory materials and firm-specific data into a single workspace. Early users report gains in speed and accuracy. Accounting firm BLISS 1041 used the system to reduce residency and filing code reviews from several days to under an hour. CoCounsel for Tax is now available in the U.S., with expanded features and geographies expected by year-end. The company said Agentic Intelligence draws on more than 20 billion proprietary and public documents and is supported by 4,500 subject matter experts. It leverages recent investments in AI infrastructure and partnerships with OpenAI, Anthropic, Google Cloud and AWS. The next product, Ready to Review, will focus on tax return preparation. Thomson Reuters plans to expand the platform into legal, compliance and risk sectors in 2025.
Google is aiming to control the distributed AI network and win data privacy war through its experimental Android app that enables running gen AI models entirely on the edge
Google has quietly released an experimental Android application that enables users to run sophisticated AI models directly on their smartphones without requiring an internet connection. The app, called AI Edge Gallery, allows users to download and execute AI models from the popular Hugging Face platform entirely on their devices, enabling tasks such as image analysis, text generation, coding assistance, and multi-turn conversations while keeping all data processing local. The application, released under an open-source Apache 2.0 license and available through GitHub rather than official app stores, represents Google’s latest effort to democratize access to advanced AI capabilities while addressing growing privacy concerns about cloud-based artificial intelligence services. “The Google AI Edge Gallery is an experimental app that puts the power of cutting-edge Generative AI models directly into your hands, running entirely on your Android devices.” At the heart of the offering is Google’s Gemma 3 model, a compact 529-megabyte language model that can process up to 2,585 tokens per second during prefill inference on mobile GPUs. This performance enables sub-second response times for tasks like text generation and image analysis, making the experience comparable to cloud-based alternatives. The app includes three core capabilities: AI Chat for multi-turn conversations, Ask Image for visual question-answering, and Prompt Lab for single-turn tasks such as text summarization, code generation, and content rewriting. Users can switch between different models to compare performance and capabilities, with real-time benchmarks showing metrics like time-to-first-token and decode speed. The local processing approach addresses growing concerns about data privacy in AI applications, particularly in industries handling sensitive information. By keeping data on-device, organizations can maintain compliance with privacy regulations while leveraging AI capabilities. Qualcomm’s AI Engine, built into Snapdragon chips, drives voice recognition and smart assistants in Android smartphones, while Samsung uses embedded neural processing units in Galaxy devices. By open-sourcing the technology and making it widely available, Google ensures broad adoption while maintaining control over the underlying infrastructure that powers the entire ecosystem. Google open-sources its tools and makes on-device AI widely available because it believes controlling tomorrow’s AI infrastructure matters more than owning today’s data centers. If the strategy works, every smartphone becomes part of Google’s distributed AI network. That possibility makes this quiet app launch far more important than its experimental label suggests.
Google has a new voice input waveform for AI Mode; the transcription appears in real-time below it
Ahead of Search Live, Google is giving AI Mode a straightforward voice input feature that has a particularly delightful animation. The main input screen (from the Search bar shortcut) now has a microphone icon next to the submit button. It joins the gallery to add existing images and Google Lens on the left side. Upon tapping you get an arc-shaped speech-to-text indicator that alternates between the AI Mode colors as you talk. The transcription appears in real-time below it. This replaces a more generic rectangular version that was available at launch on the AI Mode homepage. Search Live will be using this same animation for the immersive conversation experience, and it’s nice that we’re getting it ahead of time. Google has long used the four bouncing dots that morph into a waveform for voice input in Search and Assistant. This new one makes for a nice modernization, and contributes to how AI Mode is one of the nicest interfaces out of Google Search in quite some time
New Klarna Card is based on Visa Flexible Credential allowing choice of paying from stored cash or Pay in 4 and pay later plans at any of the over 150 million merchants
Klarna is piloting a card that spends like debit but can flip into pay later mode, extending the buy now, pay later (BNPL) model from the checkout screen to the in-store experience. Unveiled Tuesday (June 3) at Money20/20 Europe, the Klarna Card is built on Visa’s Flexible Credential, a network capability that lets a single piece of plastic surface multiple funding sources, Klarna said in a Tuesday press release provided to PYMNTS. Issued by WebBank and housed in a Federal Deposit Insurance Corp.-insured wallet, the card lets testers in the United States pay from stored cash or activate Klarna’s Pay in 4 and pay later plans at any of the over 150 million merchants that already accept Visa. A wider launch in the U.S. and Europe is slated for later this year, the release said. More than 5 million consumers have joined the waitlist. Visa’s Flexible Credential works behind the scenes. Consumers can pay up front in debit mode or activate Klarna’s installments. Eventually, cardholders will be able to select paid tiers that layer in cash back and merchant discounts, per the release. “We consistently hear from consumers that they want the freedom to choose how and when to pay — whether that’s paying now with debit or spreading the cost over time,” Klarna Chief Marketing Officer David Sandstrom said in the release. “They want simplicity, flexibility and transparency — all in one place. That’s exactly what has made Klarna’s payment methods so popular online, and now we’re bringing that same experience to a physical card.”
Citi’s CMO sets three priorities- accelerating creation of written and visual content, AI for personalization and validation with brand guidelines
Under Chief Marketing and Content Officer Alex Craddock, the bank has restructured its marketing department to better support its broader vision, and hired for several new marketing roles, including a head of wealth marketing, a head of banking and markets marketing, a head of sponsorships and partnerships, and a head of marketing innovation, he said. Citi has doubled down on growth in its wealth segment, which had struggled in the past but has become a key piece to improving the bank’s business mix by adding more fee-based revenue. Craddock pointed to wealth as an area ripe for opportunity when it comes to reinvigorating the bank’s brand. With its marketing efforts, Citi aims to think about its brand more holistically and make sure business marketing is aligned with the broader brand strategy, Craddock told. Content also falls under Craddock’s purview, and the approach there has changed, too. “We unified a fragmented marketing team: Business marketing was embedded in the businesses, enterprise marketing was centralized. We see opportunities to reenergize the brand, and wealth is a great example. We recognized, as we were bringing all the individual parts of the wealth business together, we needed to develop a new proposition and a new brand platform to launch that proposition to market. So we reposition wealth in the minds of our clients and ensure that it meets the business that we are today, and not some perception of who we were in the past. We did a lot of work when I came in through last year, to really understand our clients’ needs, understand our landscape and identify a client audience that is the right kind for Citi to go after, and build a proposition and a platform to reach them. The world change-maker audience is focused on wealth creation, and often multigenerational wealth creation. They’re often international in the way they live their lives, but also in business. We’re focused on three key priorities. One is around content creation, and how we can start to accelerate creation of written and visual content. The other is around personalization: how can we use AI to personalize content, whether that’s for specific client segments, or how we personalize, for example, our credit card marketing real-time? That can be accelerated with [artificial intelligence]. The third is around validation. We are producing a vast amount of content across the firm. It’s hard for humans to make sure it is all on-brand. We can use AI to do that validation and make recommendations as to how we might want to augment certain aspects of our content to make sure that it is consistent with brand guidelines. The other part we’re exploring is compliance approvals: how do we accelerate the speed at which we are getting compliance approval for content, by using AI to do a lot of that heavy lifting for us? We’re building the expertise and knowledge in-house, but we’re complementing that with third-party AI solutions where it makes sense, because they bring something distinctive that we couldn’t build ourselves, or it would take too long to build.”
Wells Fargo to refocus on new branch openings and renovations in NYC, Chicago and Nashville; branches to feature a “Welcome Touchpoint” desk and a giant, interactive replica of the bank’s mobile app, for demos
Wells Fargo is back in the consumer banking game, and as it steps things up, it will be relying heavily on branches. This will include new openings and renovations, including heavily contested turf in the New York metropolitan area, the Chicago area, Nashville and other areas. Branch optimization is mostly complete. Now Wells Fargo will pivot to grow its number of branches again, to serve new markets and to increase density in areas where service needs allow. CEO Charles Schraf said, a renewed focus is primary checking account growth. Beyond being part of the vanguard of this push, the uptown Manhattan location epitomizes what Wells is going for as it leans into branches. The new Wells Fargo branch (1) in Manhattan at Broadway and 82nd Street demonstrates concepts going into new branches and renovations. First stop after the ATM lobby is the “Welcome Touchpoint” desk (2), where a staffer can handle basics and let bankers beyond that point know that a customer wants to see them. (3) The customers can wait in an area patterned more after a living room than a bank branch. Nearby is a giant, interactive replica of the bank’s mobile app, for demos. (4) Just around the corner is a small teller station. (5) Finally, beyond those areas, bankers and financial advisors await, with fully walled offices to accomodate customers’ expressed desire for privacy. A key change in approach is the branch’s “Welcome Touchpoint.” This is a station just past the ATM lobby where a Wells staffer sits at a tall freestanding podium to check people in for appointments and to answer questions. The banker at the station can help out with cash withdrawals or deposits or bring the customer to the teller station. Barstools for customers are also available. The station is also close to a tall live replica of a smartphone screen to demo Wells digital features to interested customers. This giant screen can help the banker show the customer how to handle simple or even more complicated transactions, such as setting up a wire transfer, on their own, from anywhere. The bank is also working to pull the customer into processes like account enrollment. Instead of simply having customers sit passively, answering a banker’s questions as they input data on a screen, the banker does the setup and then presents the customer with a QR code that pulls their phone or tablet into the process. They can then answer questions directly, with prompts from the banker. Some branches are designated as “Wells Fargo Premier” locations, which means that Premier bankers, trained to help wealthier customers, are based there. Other locations may have such bankers present on a rotation or by appointment. A key lesson for all banks: Customers want privacy when discussing their affairs. Wells branches are going back to enclosed offices, where bankers and customers can converse out of earshot.
KeyBank to use vendor Personetics’ Engage, a client experience that delivers timely insights and recommendations based on each client’s spending and savings habits
KeyBank is further advancing its mission to empower clients to thrive by utilizing Personetics’ Cognitive Banking platform, which fosters deep personal relationships and assists consumers in achieving their financial goals. KeyBank’s 2025 annual Financial Mobility Survey found increased stress levels among Americans trying to balance economic pressures and financial goals, with more than half (51%) of Gen Z respondents indicating they are taking proactive steps to improve their financial future. Similarly, Personetics’ Global Consumer Banking Survey released in February found that most consumers (70%) want their financial institutions to provide timely insights on spending and saving habits to improve their financial wellness. To address this growing demand, particularly among younger consumers, KeyBank will use Personetics’ Engage, a client experience that delivers timely insights and recommendations based on each client’s spending and savings habits. Emily Gessner, Head of Consumer Digital at KeyBank. “By leveraging Personetics’ platform and experience, we will address the financial burden and stress consumers face by empowering our clients with real-time insights and guidance to help them effectively manage their financial futures.”
JPMorgan is reportedly hiring Citigroup dealmaker Theodoros Giatrakos as it seeks to bolster its business advising private equity firms
JPMorgan Chase is hiring Citigroup dealmaker Theodoros Giatrakos as it seeks to bolster its business advising private equity firms. The London-based banker is leaving Citigroup after more than 14 years and is set to join JPMorgan’s financial sponsors group, the people said, asking not to be identified because the information is private. Giatrakos, who joined Citigroup in 2011, was previously head of the firm’s alternative assets group for Europe, the Middle East and Africa. He had earlier led its investment banking team in central and southeast Europe. Late last year, Citigroup hired JPMorgan banker Sidharth Punshi as the new head of its EMEA alternative assets group, with Giatrakos stepping down from the leadership position to dedicate more time to client relationships.