Agentic artificial intelligence for customer experience startup Parloa GmbH announced that it has raised $120 million in new funding on a $1 billion valuation to accelerate its expansion across North America and Europe, enhance its Agent Management Platform and hire international talent. Parloa offers the AI Agent Management Platform, a platform that allows enterprises to design, deploy and manage AI-powered customer service agents across various communication channels, including voice, chat and messaging. The platform works via a low-code interface that gives businesses the ability to create AI agents equipped with prebuilt or custom skills for tasks such as routing, authentication and handling frequently asked questions. The agents are flexible by design and can be tailored to specific business needs, including integration with existing systems such as Salesforce Inc., ServiceNow Inc. and Zendesk Inc. for real-time access to relevant data. To complement its core offering, Parloa provides tools, such as large-scale testing and behavior evaluation, to fine-tune responses before deployment. Once live, a customer analytics dashboard delivers the ability to monitor key performance metrics for the continuous improvement of customer interactions. Parloa’s platform also assists human agents with features like real-time translation and suggested responses. The augmentation enhances agent productivity and ensures consistent customer service quality across different languages and regions.
Mistral’s platform enables enterprises to build AI agents tailored to their operations and gain full control over the AI stack—from infrastructure and platform features to model-level customization and user interfaces without vendor lock-in
AI startup Mistral unveiled Le Chat Enterprise, a unified AI assistant platform designed for enterprise-scale productivity and privacy, powered by its new Medium 3 model that outperforms larger ones at a fraction of the cost (here, “larger” refers to the number of parameters, or internal model settings, which typically denote more complexity and more powerful capabilities, but also take more compute resources such as GPUs to run). Available on the web and via mobile apps, Le Chat Enterprise is like a ChatGPT competitor, but one built specifically for enterprises and their employees, taking into account the fact that they’ll likely be working across a suite of different applications and data sources. It’s designed to consolidate AI functionality into a single, privacy-first environment that enables deep customization, cross-functional workflows, and rapid deployment. Among its key features that will be of interest to business owners and technical decision makers are: Enterprise search across private data sources; Document libraries with auto-summary and citation capabilities; Custom connectors and agent builders for no-code task automation; Custom model integrations and memory-based personalization; Hybrid deployment options with support for public cloud, private VPCs, and on-prem hosting. Le Chat Enterprise supports seamless integration into existing tools and workflows. Companies can build AI agents tailored to their operations and maintain full sovereignty over deployment and data—without vendor lock-in. The platform’s privacy architecture adheres to strict access controls and supports full audit logging, ensuring data governance for regulated industries. Enterprises also gain full control over the AI stack—from infrastructure and platform features to model-level customization and user interfaces. Mistral’s new Le Chat Enterprise offering could be appealing to many enterprises with stricter security and data storage policies (especially medium-to-large and legacy businesses). Mistral Medium 3 introduces a new performance tier in the company’s model lineup, positioned between lightweight and large-scale models. Designed for enterprise use, the model delivers more than 90% of the benchmark performance of Claude 3.7 Sonnet, but at one-eighth the cost—$0.40 per million input tokens and $20.80 per million output tokens, compared to Sonnet’s $3/$15 for input/output. Benchmarks show that Mistral Medium 3 is particularly strong in software development tasks. In coding tests like HumanEval and MultiPL-E, it matches or surpasses both Claude 3.7 Sonnet and OpenAI’s GPT-4o models. According to third-party human evaluations, it outperforms Llama 4 Maverick in 82% of coding scenarios and exceeds Command-A in nearly 70% of cases. Mistral Medium 3 is optimized for enterprise integration. It supports hybrid and on-premises deployment, offers custom post-training, and connects easily to business systems.
FICO’s AI framework, enables Bradesco Bank to execute instantaneous credit and fraud decisions once managed by entire departments
FICO and Bradesco Bank are collaborating closely to embed responsible, real-time AI decisioning into core financial operations, enhancing fraud detection, credit modeling and customer experience. “For FICO, responsible AI consists of four major pillars,” said Scott Zoldi, chief analytics officer of FICO. “It’s around building the AI responsibly, which means building it correctly. It’s around explainability and interpretability. It’s around ethics, ethical AI, and then accountability and auditability of that AI.” One such case study in responsible, purposive AI adoption is Brazil’s Bradesco Bank. With FICO’s AI framework, the bank executes instantaneous credit and fraud decisions once managed by entire departments. However, with this power comes an absolute demand for transparency and accuracy. Customers won’t tolerate unexplained decisions, especially in financial services. For Bradesco Bank, responsible AI isn’t a bonus — it’s business-critical, according to Rafael Cavalcanti, CDAO & CRM director at Bradesco Bank. “I do think that this golden age brings high expectations of customers as well,” he said. “They want to have that same experience that they have when using streaming solutions or while doing online shopping. They want banks to do that. We need to combine these techniques, and at the same time, understand that what we do has a core value that is privacy and responsibility, and that’s what we do in Bradesco.”
Structify’s AI platform combines visual language model with human oversight to simplify data preparation by letting users create custom datasets by specifying the data schema, selecting sources, and deploying AI agents to extract that data through navigating the web
Startup Structify is taking aim at one of the most notorious pain points in the world of artificial intelligence and data analytics: the painstaking process of data preparation. The company’s platform uses a proprietary visual language model called DoRa to automate the gathering, cleaning, and structuring of data — a process that typically consumes up to 80% of data scientists’ time. At its core, Structify allows users to create custom datasets by specifying the data schema, selecting sources, and deploying AI agents to extract that data. The platform can handle everything from SEC filings and LinkedIn profiles to news articles and specialized industry documents. What sets Structify apart, is their in-house model DoRa, which navigates the web like a human would. This approach allows Structify to support a free tier, which will help democratize access to structured data. Structify’s vision is to “commoditize data” — making it something that can be easily recreated if lost. Finance teams use it to extract information from pitch decks, construction companies turn complex geotechnical documents into readable tables, and sales teams gather real-time organizational charts for their accounts. A key differentiator for Structify is its “quadruple verification” process, which combines AI with human oversight. This approach addresses a critical concern in AI development: ensuring accuracy. What differentiates Structify, according to CEO Alex Reichenbach, is its combination of speed and accuracy. Reichenbach claimed they had sped up their agent “10x while cutting cost ~16x” through model optimization and infrastructure improvements.
Verizon launches 3-year price lock guarantee on all mobile and home internet network plans targeting the growing number of households with connected devices
The average Verizon internet household now manages 18 connected devices and consumes 656 GB of data monthly, up 6% YoY, according to the latest Verizon Consumer Connections Report. Verizon is offering an industry-first 3-year price lock guarantee on all myPlan (mobile) and myHome (home internet) network plans, available to both new and existing customers. The offer meets a growing consumer demand for financial stability, control and simplicity — especially as families weigh technology investments alongside everyday priorities. Verizon is also now offering a free phone to new and existing wireless customers on any myPlan when they trade in a device — regardless of its condition. Plus, get even more immediate savings when bundling home internet, taking $15 off the total monthly bill. Families can stay confidently connected — with steady rates on core services like calling, data and texting. Customers can also save over 40% on five of the most popular subscription services Plus, free satellite messaging on qualifying devices. Verizon’s 3-year price lock guarantee offers predictable, multi-year stability across connectivity costs. As families upgrade their homes and devices, Verizon’s price lock guarantee offers something rare in tech: predictability.
Akamai creates firewall purpose built for unique AI threats unauthorized queries, adversarial inputs, and large-scale data-scraping attempts
Akamai Technologies announced Firewall for AI, a new solution that provides multilayered protection for AI applications against unauthorized queries, adversarial inputs, and large-scale data-scraping attempts. Combined with other new enhancements such as API LLM Discovery, Akamai Firewall for AI provides customers with a holistic set of AI-driven capabilities. AI models contain valuable proprietary knowledge and sensitive datasets, making them prime targets for attackers. Akamai Firewall for AI addresses this as a purpose-built security solution designed to protect AI-powered applications, LLMs, and AI-driven APIs from emerging cyberthreats. By securing inbound AI queries and outbound AI responses, the firewall closes security gaps that generative AI technologies introduce. Key features of Firewall for AI include: Multilayered protection: Blocks adversarial inputs, unauthorized queries, and large-scale data scraping to prevent model manipulation and data exfiltration. Real-time AI threat detection: Uses adaptive security rules to dynamically respond to evolving AI-based attacks, including prompt injection and model exploitation. Compliance and data protection: Helps ensure AI-generated outputs remain safe and align with regulatory and industry standards. Flexible deployment options: Deploys via Akamai edge, REST API, or reverse proxy, enabling seamless integration into existing security frameworks. Proactive risk mitigation: Filters AI outputs to prevent toxic content, hallucinations, and unauthorized data leaks.
Pinterest now lets shoppers use a “more expansive visual ‘vocabulary’ to ‘describe’ their style ideas.- generating the words they can use to figure out what they like about the image, and then further explore and shop
Pinterest is introducing new features to let users find ideas that match their personal tastes. The new visual search tools will first be available for women’s fashion content in the U.S., Canada and the U.K., eventually moving to more categories. The new tools let users find what they’re searching for without using words, by breaking down and decoding images so users can quickly search and shop for the details of an outfit they like. “Whether it’s an overall aesthetic, a color palette, a specific fit, or product category, when users view a Pin, we’ll now generate the words they can use to figure out what they like about the image, and then further explore and shop,” the company said. To help users identify and select the objects that they want more easily, the company says it has added a new animated glow. Also new is a “refinement bar” to help users narrow down their search results and find things that match their style. The company said the new search options are powered by visual language models (VLMs), a form of generative artificial intelligence (AI), to offer users a “more expansive visual ‘vocabulary’ to ‘describe’ their style ideas.” “Our visual search technology represents a shift in how users interact with and discover inspiration,” Dana Cho, Pinterest vice president of design, said. “We’re not simply delivering search results — we’re curating a personalized journey of discovery that empowers individuals to find their unique style, and shop it too.”
The Hispanic Organization of Mortgage Experts (HOME) announced the launch of La Reina, the first-ever shoe created for women in mortgage
The Hispanic Organization of Mortgage Experts (HOME) announced the launch of La Reina, the first-ever shoe created for women in mortgage. La Reina, which is Spanish for “the Queen,” was designed in partnership with AliveShoes and manufactured in Le Marche, Italy. Nicole Yelland, HOME’s chief communications and strategy officer, told that the shoe pays homage to women in the mortgage industry who often have to resort to heels at business events. Each pair of shoes features a metallic gold sole to honor Hispanic and Latino culture. Besides the sole, the shoe is black and “high-end leather intended to go well with a suit,” Yelland said. “The other part of wanting to have a shoe that honors Hispanic and Latino culture is just to help people see it right. But people can’t know what they don’t see.” “Inspired by global voices like Bad Bunny, who use their platforms to push culture forward, we are shining a light on these powerful forces and giving them the recognition they deserve,” he added. “Launching during Mother’s Day week, La Reina will continue celebrating them throughout the year.”
Claude’s web search API to allow the AI assistant to conduct multiple progressive searches using earlier results to inform subsequent queries complete with source citations
Anthropic has introduced a web search capability for its Claude AI assistant, intensifying competition in the rapidly evolving AI search market where tech giants are racing to redefine how users find information online. The company announced that developers can now enable Claude to access current web information through its API, allowing the AI assistant to conduct multiple progressive searches to compile comprehensive answers complete with source citations. Anthropic’s technical approach represents a significant advance in how AI systems can be deployed as information gathering tools. The system employs a sophisticated decision-making layer that determines when external information would improve response quality, generating targeted search queries rather than simply passing user questions verbatim to a search backend. This “agentic” capability — allowing Claude to conduct multiple progressive searches using earlier results to inform subsequent queries — enables a more thorough research process than traditional search. The implementation essentially mimics how a human researcher might explore a topic, starting with general queries and progressively refining them based on initial findings. Anthropic’s web search API represents more than just another feature in the AI toolkit — it signals the evolution of internet information access toward a more integrated, conversation-based model. The new capability arrives amid signs that traditional search is losing ground to AI-powered alternatives. With Safari searches declining for the first time ever; we’re witnessing early indicators of a mass consumer behavior shift. Traditional search engines optimized for advertising revenue are increasingly being bypassed in favor of conversation-based interactions that prioritize information quality over commercial interests.
Stripe services can now natively run on Verifone payment devices, to support commerce use cases such as self-service checkout and tableside ordering
Through this partnership, Stripe services will natively run on Verifone payment devices, offering Stripe customers even more flexibility and choice in deploying durable and performant in-person payments. Merchants can now support commerce use cases such as self-service checkout and tableside ordering by leveraging Verifone’s payment devices with Stripe Terminal, an in-person payments solution. Verifone and Stripe support digital wallets, QR codes on-reader, digital or printed receipts, and interactive screens for tipping, loyalty programs, customer signatures, and more. Stripe customers can choose from a broad range of Verifone devices—from multilane to handheld readers with built-in printers—designed to meet the needs of any in-person commerce experience. With powerful device management tools and dedicated support from Stripe and Verifone, merchants can confidently manage and scale their in-person payments with ease.