OnePay is reportedly adding its own branded wireless plan as it works to become an American super app, in partnership with mobile services startup Gigs. The plan costs $35 a month for unlimited 5G data, talk and text on the AT&T network, Gigs said. The plans are activated in-app with a few clicks and don’t require credit checks or activation fees. OnePay services include credit and debit cards, high-yield savings accounts, buy now, pay later loans, and a digital wallet with peer-to-peer payments. Gigs CEO Hermann Frank said that embedding wireless plans into fintech services can lower AT&T’s customer acquisition costs — savings which can be shared with end users. “The average consumer largely overpays for their phone bill,” Frank said. “We can now offer a product at a price point that is about half what the typical consumer pays right now, with all the modern features that you require.”
Bluwhale helps Gen Z bridge Web2 and Web3 info with unified dashboard connecting stablecoins, crypto activity, and bank accounts, delivering personalized AI agents for automated asset management and scoring
Bluwhale, the decentralized intelligence network powering AI agents across blockchains, has launched its digital financial health dashboard aimed at Gen Z navigating a hybrid world of digital assets and traditional finance. Bluwhale bridges Web2 and Web3, creating a unified dashboard that connects bank accounts, stablecoins, crypto wallets, staking activity, and DeFi flows. With the introduction of the Whale Score, Bluwhale is rewriting the rules for financial health by combining and analyzing crypto, stablecoins, staking income, digital assets, cash, and traditional investments in real time. The app lets users seamlessly track, benchmark, and gamify their financial standing, while AI agents deliver personalized products and services—all without manual input. It combines digital asset transparency with AI-powered analytics to deliver a dynamic financial health score, reflecting the broader trend among younger users who view wealth not just as an asset—but as a social signal often called “flexing.” Bluwhale aggregates real-time data from wallets, bank accounts and credit cards while encouraging large financial institutions and web3 enterprises to meet Gen Z’s digital expectations. Unlike traditional tools like RocketMoney, which require manual data input or tedious accounting practices, the Whale Score (0–1000) automatically unifies traditional banking data and crypto activity into a gamified financial health dashboard while AI agents manage assets in the background. The system produces a real-time, percentile-ranked financial health score—similar to Apple Health or Whoop—that measures physical health, so consumers can compare their performance with their peers.
CFOs adopt “generative finance” to shift from backward‑looking data compilation to forward‑thinking strategic modeling; creating tailored forecasts that simulate infinite business evolution possibilities
Generative AI is introducing a whole new paradigm in the form of Generative Finance. With it you have a forward-looking capability that takes you from reporting on what happened to a model that more strategically considers what could happen next. It helps you refocus on the windshield, simulating the infinite possible ways your business can evolve. It means you can finally begin looking ahead, testing strategies for what happens next, and making assumptions and decisions based on a more complete picture of everything ahead of you. The shift from hindsight to foresight equips your organization with an edge in strategic thinking. It doesn’t just organize the numbers you have; it intelligently creates new data to show you what’s possible. It learns the unique patterns and rhythms of your specific business from your own data. It lets you explore hundreds of potential outcomes for any decision you’re considering. It presents a range of probable futures, not just a single, rigid prediction. It continually refines its understanding, getting smarter and more accurate over time. With generative finance, you can get detailed, data-supported answers in minutes, not months. When you adopt generative finance, your FP&A professionals are freed from the tedious work of manual data entry and report building. They can stop spending their time compiling spreadsheets and start investing their time in analysis and strategy. They become true business partners who can interpret complex scenarios, stress-test business plans, and provide the insightful guidance your leadership team needs to make better decisions. The real power of this technology is unlocked when it is explicitly trained on your company’s own data. By learning from your unique financial history, sales cycles, and operational details, the AI becomes a true expert on your business. This deep customization is what makes its forecasts and scenarios so reliable. The insights you get from a tailored generative finance model are directly relevant and immediately actionable because they are rooted in your reality.
Figure Technology embeds AI and Provenance blockchain into consumer lending to cut HELOC approval to 10 days, enabling on‑chain origination, AI underwriting and smart‑contract loan trading.
Figure Technology’s recent filing to go public spotlights the growth of AI and blockchain into loan origination, underwriting and secondary market trading. The company’s platform is built on the Provenance blockchain, which it describes as a “record of truth” for assets. Every loan originated through its system is recorded on the blockchain, providing an immutable record of ownership and performance. Figure combines that with automated valuation models, AI-powered underwriting, and smart contracts that govern loan sales and transfers. This approach has allowed the company to shorten approval times for home equity lines of credit (HELOCs) to a median of 10 days from an industry average of 42 days. Loan applications can be completed in five minutes, with funding available in as little as five days. Figure estimates its addressable market across lending and capital markets at approximately $185 billion in annual revenue potential, based on consumer credit originations and marketplace trading. In addition to lending, management is targeting tokenization and stablecoins as growth opportunities. The filing contends that the company has achieved profitability and scaled it in a capital-efficient way. Revenue models are built on fees from originations, servicing, gain on loan sales, and technology usage. Partner-branded lending, where banks and mortgage originators use Figure’s platform under their own brand, accounts for 77% of total originations. Figure had 168 active partners as of mid-2025. The company has also built regulatory infrastructure to support its ambitions. It holds more than 180 lending and servicing licenses, 48 money transmitter licenses, and SEC registration as a broker-dealer with authority to operate an alternative trading system. Internationally, it has crypto licenses in the Cayman Islands and Ireland. Management argues that this licensing framework differentiates it from competitors and will support scaling of new products.
Cognigy conversational and agentic AI will be enriched with NiCE’s purpose-built CX AI models to make agents smarter, deployments faster, and outcomes more impactful
NiCE announced the successful closing of its acquisition of Cognigy, following receipt of all required regulatory approvals. This brings together two AI industry leaders, each with a proven track record of market leadership, innovation, and customer impact to accelerate AI adoption in customer experience across the front and back office. By bringing together Cognigy’s AI and NiCE’s award-winning CXone Mpower CX AI platform, organizations of all sizes will transform how they deliver AI-powered customer experience. Cognigy conversational and agentic AI will be enriched with NiCE’s purpose-built CX AI models leveraging decades of CX intelligence, making agents smarter, deployments faster and outcomes more impactful. With this unified power, organizations can scale agentic and conversational AI at speed across CXone Mpower, delivering seamless, intelligent experiences across every customer touchpoint, from the contact center to enterprise-deep workflows. Scott Russell, CEO, NiCE, said, “With the completion of this acquisition, we are bringing together two AI market leaders to redefine the future of customer experience. Together, we are accelerating AI adoption and value realization for global enterprises by delivering one of the industry’s most powerful and comprehensive customer experience platforms leveraging CX models and agentic and human agents powered by decades of CX purpose-built data and insights.
Visa enables agentic commerce with tokenized credentials, device-specific authentication and intent-matching “payment instructions,” that verify agent purchases against consumer requests to mitigate hallucinations and fraud
Visa has released new developer tools that allow AI agents to connect directly to Visa’s payment infrastructure, enabling what the company calls “agentic commerce” — a system where AI bots handle everything from product discovery to checkout completion based on consumer preferences and spending limits. Rather than browsing websites and manually completing purchases, consumers would set parameters for AI agents that then autonomously find, evaluate, and buy products across multiple merchants. Rubail Birwadker, Visa’s Global Head of Growth said “These agents will need to be trusted with payments, not only by users, but by banks and sellers as well.” Visa’s new offering centers on two key products: a Model Context Protocol (MCP) Server that provides secure access to Visa’s payment APIs, and the Visa Acceptance Agent Toolkit, which allows both technical and non-technical users to deploy AI-powered payment workflows using plain language commands. The MCP Server represents a significant technical breakthrough, providing AI agents with a standardized way to communicate with Visa’s trusted network without requiring custom integrations for each application. Developers can now move “from idea to functional prototype in hours instead of days or weeks,” according to the company. Visa has implemented multiple layers of protection, including immediate tokenization of card credentials, device-specific authentication, and what Birwadker calls “payment signals” and “payment instructions” that verify AI agent actions align with original consumer intent. “Your PII or your PAN is never going to be exposed,” Birwadker said, referring to personally identifiable information and primary account numbers. “We almost immediately take that pan, we convert it into a token, and we authenticate that token and tie it to a specific device for a specific application.” The company has also developed a matching process that prevents transaction completion until it confirms an AI agent’s intended purchase matches what the consumer originally requested. This addresses concerns about AI “hallucinations” — instances where language models generate incorrect or nonsensical outputs.
Walmart partners with OpenAI to deliver free, customized AI training certifications via Walmart Academy for all U.S. frontline and office associates starting in 2026
Beginning next year, all U.S. front-line and office-based associates at Walmart will have access to AI training through a new collaboration with OpenAI, the company behind ChatGPT. As indicated by Chief People Officer Donna Morris, OpenAI is launching a new OpenAI Certifications program, and Walmart is working with them to create a customized experience for its own associates. Through Walmart Academy, the largest private training program in the world, with more than 3.5 million participants, associates will have free access to a tailored version of this certification. The training is designed to help associates succeed at work, grow their careers and thrive in an increasingly digital world. John Furner, president and CEO, Walmart U.S said, “By bringing AI training directly to our associates, we’ll enable our people to maximize the benefit of AI-powered technology – giving them the skills they need to rewrite the playbook and shape the future of retail.” While this certification will launch in 2026, associates can currently access training opportunities, including AI training, through the company’s Live Better U education benefit. The collaboration with OpenAI builds on Walmart’s nearly $1 billion commitment to skills training through 2026.
Microsoft-backed Sola Security’s no-code AI cybersecurity platform enable teams to build custom threat detection apps with graph research capabilities
Sola Security offers an artificial intelligence platform that enables cybersecurity teams to build threat detection apps using prompts. To use the platform, workers must specify the systems they wish to monitor and the kind of vulnerabilities that they’re looking to track. Sola Security then automatically generates the software components necessary to perform the task. The platform first connects to the systems that it’s instructed to monitor and collects cybersecurity data. Companies can create apps that detect vulnerable assets such as Amazon S3 buckets without encryption enabled. According to Sola Security, its platform also spots issues that affect employee accounts. Customers can build apps that identify accounts with access to more data or systems than strictly necessary. Sola Security says that user-created apps can visualize their findings in graphs to ease analysis. According to the company, its platform regularly checks for new vulnerabilities and generates an alert when one is found. A chatbot embedded in custom apps allows customers to analyze the data with natural language questions. The platform uses a feature dubbed graph research to answer some user questions. According to the company, the technology helps cybersecurity teams determine whether an issue in one system might affect other assets. For example, it could point out if a vulnerability in a tool that a company uses to manage administrator accounts might expose its cloud environment to cyberattacks. Users can modify the dashboards that Sola generates using a no-code customization tool. An administrator might create one version of a database monitoring dashboard for the cybersecurity team and another for the business workers who use the database. For customers that don’t require extensive customization, Sola offers prepackaged cybersecurity apps built by its engineers.
Apiiro report finds AI-assisted coding increases developer speed fourfold but raises security risks tenfold amid rising architectural flaws and secret exposures
A new report from application security posture management company Apiiro Ltd. details a tenfold increase in security findings among Copilot users, peaking in mid-2025. Two primary factors were found to be driving the surge: open-source dependencies and secure coding issues. AI-assisted developers were found to be more prone to design-level flaws versus conventional developers, who were more likely to introduce logic mistakes. The architectural weaknesses are more costly to remediate and harder to catch later on, creating a structural challenge for organizations trying to balance speed with security. Secrets exposure was also found to diverge between developers. Developers working with Copilot leaked higher volumes of cloud credentials, while non-Copilot users were more likely to expose generic application programming interface tokens. The key takeaway is that AI assistance may inadvertently amplify risks related to cloud identity and credential management. The findings in the report include that developers using AI tools on average generate three to four times more commits on average, but the contributions were consolidated into fewer, larger pull requests, or proposed code changes. The increased throughput was found to accelerate delivery but also add complexity for application security teams — since traditional review processes are now insufficient to keep up with the scale and intricacy of AI-assisted code. The report also details how average pull request sizes and commit volumes have sharply increased as AI coding assistance has been adopted. AI-assisted developers were found to produce more code but open fewer pull requests. Larger, more complex code submissions are noted as elevating the risk of shallow reviews and missed vulnerabilities. Apiiro’s researchers warn that though AI code assistants can drive dramatic improvements in developer productivity, they also introduce new categories of risk that organizations must address
Google’s EmbeddingGemma small model optimizes 308M-parameter multilingual embeddings for phones and laptops; enabling offline, private semantic search and retrieval in enterprise apps
Google’s open-source Gemma is already a small model designed to run on devices like smartphones. However, Google continues to expand the Gemma family of models and optimize these for local usage on phones and laptops. Its newest model, EmbeddingGemma, will take on embedding models already used by enterprises, touting a larger parameter than most and strong benchmark performance. EmbeddingGemma is a 300 million token parameter, open-source model that is best optimized for devices like laptops, desktops and mobile devices. Min Choi, product manager, and Sahil Dua, lead research engineer at Google DeepMind, wrote in a blog post that EmbeddingGemma “offers customizable output dimensions” and will work with its open-source Gemma 3n model. “Designed specifically for on-device AI, its highly efficient 308 million parameter design enables you to build applications using techniques such as RAG and semantic search that run directly on your hardware,” Choi and Dua said. “It delivers private, high-quality embeddings that work anywhere, even without an internet connection.” The model performed well on the Massive Text Embedding Benchmark (MTEB) multilingual v2, which measures the capabilities of embedding models. It is the highest-ranked model under 500M parameters. A significant use case for EmbeddingGemma involves developing mobile RAG pipelines and implementing semantic search. RAG relies on embedding models, which create numerical representations of data that models or agents can reference to answer queries. Building a mobile RAG pipeline enables information gathering and answering queries more directly on local devices. Employees can ask their questions or direct agents through their phones or other devices to find the information they need. Choi and Dua said that EmbeddingGemma works to create high-quality embeddings. To do this, EmbeddingGemma introduced a method called Matryoshka Representation Learning. This gives the model flexibility, as it can provide multiple embedding sizes within a single model.