Wednesday, May 20, 2026

Fintech firm Mercury hits $5.2 billion valuation after funding round, up 49% in 14 months

Fintech firm Mercury has raised $200 million, achieving a valuation of $5.2 billion, which represents a 49% increase from its previous funding round just 14 months ago. This fundraising comes shortly after Mercury received conditional approval from the Office of the Comptroller of the Currency to become a federally regulated bank, enabling it to offer more financial services and reduce reliance on partner banks. With over 300,000 customers and $650 million in annualized revenue, Mercury has bucked the fintech downturn and has been profitable for four consecutive years.

Jim Cramer: Wall Street is waking up to how wrong it got this cybersecurity stock

Jim Cramer highlighted a significant shift in Wall Street's perception of CrowdStrike, which saw its stock rise 4% to a new all-time high. He argued that fears regarding AI undermining cybersecurity companies are unfounded, asserting that the demand for cybersecurity solutions is actually growing as businesses seek to bolster defenses against escalating digital threats.

VisitorsCoverage launch highlights growing consumer openness to AI-led insurance journeys

VisitorsCoverage has introduced a ChatGPT-powered comparison app that allows consumers to research, compare, and purchase travel insurance policies through natural language conversations. This launch comes as a GlobalData survey reveals that 42% of consumers are open to receiving insurance quotes from chatbots, highlighting a growing trend towards AI-led insurance interactions despite notable resistance among some users.

Amazon.com defeats appeal claiming it aided tariff evasion

A U.S. appeals court dismissed a whistleblower's claim that Amazon.com facilitated tariff evasion by foreign fur manufacturers, ruling there was no evidence that Amazon knowingly ignored the situation. The court highlighted that the lower prices charged by these manufacturers could be attributed to legitimate factors, such as economies of scale, rather than fraudulent practices.

Intuit to lay off over 3,000 employees to refocus on AI

Intuit, the enterprise software giant known for products like TurboTax and QuickBooks, is laying off approximately 3,000 employees, or 17% of its workforce, to refocus its resources on integrating artificial intelligence into its offerings. CEO Sasan Goodarzi stated that these layoffs aim to streamline the company's structure and enhance its AI initiatives, amidst a challenging year for the tech sector, which has seen significant job cuts across multiple companies.

Bezos brushes off concerns of an AI bubble: 'You shouldn't worry about it'

In a recent interview with CNBC, Amazon founder Jeff Bezos dismissed concerns about a potential artificial intelligence bubble, asserting that the significant investments in AI will ultimately drive technological advancement. He noted that even if a bubble occurs, the influx of capital will lead to many beneficial innovations, similar to the biotech boom of the 1990s. Bezos also highlighted his new AI startup, Project Prometheus, which focuses on developing advanced AI models for engineering and manufacturing, launched with substantial funding to ensure dedicated focus on its goals.

AI search startups are blowing up

AI search startups are rapidly emerging as Google transitions to an AI-driven search experience. Notably, Exa Labs has raised $250 million at a $2.5 billion valuation, joining a wave of companies like Parallel Web Systems, which recently secured $100 million. This shift is attracting attention from established tech firms like Amazon and LinkedIn, eager to enhance their own search functionalities.

Stability AI releases a new audio model that can create six-minute songs

Stability AI has unveiled its latest audio model family, Stability Audio 3.0, which can generate professional-grade music exceeding six minutes in length. This release includes four models, with the medium and large versions capable of producing compositions lasting 6 minutes and 20 seconds, a significant improvement over the previous version from 2024. The small models will be available with open weights, allowing users to modify and utilize them freely.

Wendy's taps former Potbelly CEO to lead struggling burger chain

Wendy's has appointed Bob Wright, the former CEO of Potbelly, as its new chief executive, effective Thursday. This leadership change comes as the burger chain faces significant challenges, including five consecutive quarters of declining same-store sales and plans to close approximately 300 restaurants. Wright's experience in turning around Potbelly after pandemic-related setbacks may be crucial as Wendy's seeks to regain market share lost to competitors like McDonald's and Burger King.

It’s make or break time for AI labeling systems

Google is testing its AI labeling systems, SynthID and C2PA, in a significant expansion aimed at combating deepfakes. Announced at the I/O conference, SynthID will now be integrated into Chrome and Search, allowing users to easily verify the origins of images and videos. This collaboration with OpenAI and Meta represents a concerted effort to enhance content authenticity and transparency in the face of rising AI-generated misinformation.

If Google can’t make AI agents useful, maybe no one can

At I/O 2026, Google unveiled its latest advancements in AI agents, including Gemini Spark, which promises seamless integration with its suite of services and over 30 external applications. Building on the success of OpenClaw, Google aims to transform AI from a theoretical concept into a practical tool for everyday tasks, with agents capable of operating continuously in the background.

First-gen Chromecast streamers are suddenly failing for some users, 13 years later

First-gen Chromecast devices, launched in 2013, are increasingly failing for users, with reports surfacing that many can no longer stream content effectively. While Google ceased updates for these devices in 2023, the recent surge in failures, particularly with apps like YouTube and HBO Max, suggests a significant decline in functionality, marking what appears to be the end of an era for this once-popular streaming device.

Man wins $835K after sheriff jailed him for a month over Charlie Kirk post

Larry Bushart, a retired cop from Tennessee, secured an $835,000 settlement after being jailed for 37 days due to a Facebook post featuring a Trump meme. His legal team argued that the arrest was an attempt to censor his speech, violating his First Amendment rights. Bushart expressed relief at the vindication of his rights and the financial support will aid his retirement, which was jeopardized by the incident.

Samsung preps 6.47-inch Galaxy S27 ‘Pro’ to make 4 flagship models next year

Samsung is preparing to launch the Galaxy S27 Pro, which will feature a 6.47-inch OLED display, as part of a new strategy to expand its flagship lineup to four models. This move comes after the company previously opted to limit its offerings to three devices, suggesting a willingness to take risks in a challenging manufacturing environment.

ADT ‘Blu’ is a confusing new DIY security system that ditches Google Nest integration

ADT has introduced Blu, a new DIY security system that aims to provide a more affordable option compared to its existing ADT+ system. However, Blu's lack of Google Nest integration and its similar pricing to ADT+ raise questions about its value proposition, leaving potential customers confused about its advantages.

Presentation: The AI Gateway: Scaling Centralized Inference Across Decentralized Teams

Meryem Arik, CEO of Doubleword, presented on the significance of AI model gateways in addressing the 'inference chaos' that modern engineering teams encounter. She highlighted the need for a balance between empowering decentralized teams to select appropriate models and ensuring centralized oversight for security and cost management. Arik introduced open-source solutions like LiteLLM and Doubleword as essential tools for streamlining AI infrastructure and improving operational efficiency.

OpenAI Outlines WebRTC Architecture for Low-Latency Voice AI at Scale

OpenAI has unveiled a new architecture leveraging WebRTC to optimize low-latency voice AI at scale. By replacing the conventional media termination model with a relay-transceiver design, the company aims to enhance global reach and connection speed while minimizing operational complexity in Kubernetes environments. This innovative approach focuses on maintaining session state in a dedicated transceiver layer, allowing for more efficient media routing and improved user experience.

AI companies and data brokers even resort to fake forms to keep selling our data

A recent privacy study has uncovered that data brokers and AI companies are using deceptive tactics to thwart consumers' attempts to opt out of personal data sales. An audit of 38 major firms, including industry giants like Google, Meta, and OpenAI, revealed practices such as fake opt-out forms and convoluted processes that make it difficult for users to protect their privacy. The findings underscore the urgent need for stronger federal privacy regulations in the U.S.

Jeff Bezos says bottom half of earners should pay zero in income taxes

In a recent CNBC interview, Jeff Bezos argued that the bottom half of earners should not pay any federal income taxes, stating that their contributions are negligible compared to the overall tax revenue. He highlighted the financial strain on lower-income workers, suggesting that instead of taxing them, the government should apologize for asking them to contribute.

Google tests the app market version of the SaaSpocalypse

Google's AI Studio now allows users to create simple Android apps directly from text prompts in their browser, potentially making traditional apps from the Play Store less relevant. This shift towards personal software could disrupt the app market, as users may prefer to generate tailored solutions for specific needs rather than searching for existing applications. While Apple maintains strict security measures against similar tools, Google's approach could redefine how users interact with software on Android devices.

Trump Mobile may be leaking customer addresses

As the T1 Phone from Trump Mobile approaches its release, the company is facing serious allegations of data insecurity. Two YouTubers have reported that their personal information, including addresses and phone numbers, was leaked from the Trump Mobile website, revealing a much lower number of orders than previously claimed.

Designing a Multi-Agent System for Engineering Support at Scale: A Case Study From Grab

Grab's Analytics Data Warehouse team has implemented a multi-agent AI system to streamline engineering support across its extensive data platform. This initiative aims to automate tasks such as SQL debugging and data warehouse troubleshooting, thereby freeing up engineers to concentrate on more strategic development efforts. The system is designed to handle a high volume of internal requests, significantly enhancing operational efficiency and reducing the time spent on routine support tasks.

Apple’s three goals for iOS 27 sound like a big win for users

With the unveiling of iOS 27 approaching, Apple is reportedly focusing on three main goals: enhancing AI features, refining the Liquid Glass design, and implementing crucial bug fixes and battery improvements. These updates are seen as essential for addressing user concerns and revitalizing Apple's standing in the AI space, especially following an underwhelming performance with Siri in previous releases.

Yearslong fight over users' right to tweak smart TV software heads to trial

The Software Freedom Conservancy (SFC) is taking Vizio to trial over the company's refusal to release the complete source code for its Linux-based smart TV operating system, which has limited user control and raised privacy concerns. The outcome of this case, set for August 10, could have significant implications for user rights across the smart TV industry, especially as many systems are built on similar Linux foundations.