Full-text search across article titles and summaries. Refine with sector, year, content type, and reading-time filters.
Revolut is considering a dual listing on Nasdaq and the London Stock Exchange, though CEO Nik Storonsky said the US remains the preferred market because of its deeper investor base. The fintech was valued at $75 billion in a secondary sale last year and has 80 million customers globally. It reportedly discussed targeting a valuation of up to $200 billion in an eventual IPO, which is still likely at least a year away.
A House China Committee letter raised concerns that Airwallex's China-based workforce and exposure to Chinese law could create risks around access to sensitive US financial data, prompting Joe Lonsdale to warn companies that may use Airwallex infrastructure. Airwallex CEO Jack Zhang rejected the allegations, arguing that the relevant issue is data architecture and access controls rather than employee location, saying the company's safeguards have been independently assessed and that the concerns did not stem from a breach or regulatory finding. Rippling CEO Parker Conrad also clarified that zero Rippling payments run through Airwallex.
Stripe says new SaaS platform creation is accelerating despite fears that AI would commoditize software, with new platform businesses up more than 180% year over year and more integrations reaching meaningful payment volume faster than prior cohorts. Its argument is that durable SaaS will increasingly be built around core workflows, proprietary business context, and embedded money movement, with AI making those platforms more capable rather than eliminating the application layer.
This article revisits a 2018 framework for how banks, fintechs, and big tech would converge, and argues that the emerging winner is a new model we call the Hyperbank: a technology company that owns the financial stack down to the regulatory or sovereign root. Column illustrates the model with just $1.77B in assets but roughly $200MM in revenue, a potential $6B valuation, and customers including Ramp, Brex, Bilt, Mercury, and Wise, while vertically integrating stablecoins, card issuing, global banking, multicurrency accounts, and core infrastructure. Hyperbanks are different from Deep Fintechs, which build difficult-to-copy technologies such as blockchains, zero-knowledge proofs, AI underwriting, and advanced trading infrastructure. Neobranches own digital distribution while renting the underlying financial factory.
Mastercard, Trip.com, and Network International are piloting an agentic travel-booking experience that lets Trip.com's TripGenie AI agent help users discover, book, and pay for travel services end to end. The initiative uses Mastercard's Agent Suite for Merchants and payments infrastructure to keep merchants in control of the customer experience while enabling consumer-authorized AI purchases, with a phased commercial rollout expected in early 2027.
The San Francisco startup connects to existing bank accounts, moves idle balances into US Treasuries, and returns funds before scheduled bills come due. Users can earn a yield without switching banks or manually moving money, with real-time cash flow monitoring determining when to invest and when to pull funds back.
Lightspeed argues that the real breakthrough in next-generation ERP is not a better interface, but using AI to eliminate migration risk, historically the biggest reason CFOs stayed with legacy systems like NetSuite. Its portfolio company DualEntry uses AI to migrate and reconcile company data in as little as 24 hours, pursuing a “reverse wedge” strategy that replaces the ERP first and then expands into adjacent finance workflows.
Warren Buffett is stepping down as Berkshire Hathaway chairman at age 96, becoming chairman emeritus while remaining on the board. His son Howard Buffett will become chairman, with Greg Abel continuing as CEO. The move completes Berkshire's long-planned succession and puts more responsibility on Abel to deploy the company's $365.5 billion cash hoard, as Buffett closes a six-decade tenure in which Berkshire compounded shareholder returns at 19.7% annually, nearly double the S&P 500.
Customers can use qualifying government-issued digital IDs when opening accounts in person or online, provided the institution has systems to process them.
Grab, Southeast Asia's Uber, has agreed to acquire a 60% stake in Atome Financial for $1.49 billion in cash, adding AI-powered credit underwriting and a $1 billion loan portfolio.
Anthropic is reportedly pushing its planned IPO from October to November, giving the company more time to present third-quarter results and address investor concerns around AI safety, competitive pressure, and the pace of model development.
Sponsored
Payments start as a weekend project and become a permanent line item on your roadmap. Stripe is the layer underneath: accept payments, offer financial services, and build custom revenue models on one stack, from your first transaction to your billionth.
A board member sends you a chart of startups reaching $100 million in recurring revenue in a few quarters. Are you supposed to build next year's plan around that? ICONIQ's Pacesetter Index starts with selected AI-native companies, primarily from its own portfolio, then singles out the strongest performers. That's a much narrower group than the top quarter of all startups. Before using those numbers as a target, compare company size and how much cash each business spends to grow. The chart may show a pace you want without showing a bill you can afford.
One of the effects of AI code generation tools has been the proliferation of code. This has led to a pronounced spike in the number of apps. However, while supply may have increased, demand has a life of its own. For the most part, no one is really using these new vibe-coded apps. Building successful apps has always been more than just generating the code.
Someone who loved your product moves to a new company. You could send the same congratulations-and-pitch email they'll get from everyone else. Or you could bring up what they actually accomplished with your product and ask whether they're tackling something similar in the new role. One workflow here waits a month before making that approach. Another reopens lost deals when the original obstacle changes, such as a budget or executive. The reason to reach out comes from what happened in the account, rather than how long it's been since the last email.
A bug report often starts a second conversation. What were you doing? Where did it break? Can you make it happen again? PostHog connects feedback submitted inside the product to agents that investigate and propose fixes. In one example, an investigation report arrived within five minutes, followed by a proposed code change eight minutes later. Other workflows investigate error spikes and inspect session recordings. Each step passes its findings to the next, reducing how much someone has to reconstruct before they can start fixing the problem.
Sponsored
AX is a high-throughput, declarative orchestrator that runs billions of autonomous agent workloads in a cluster. Users declare an agentic task with workspaces and gateway specifications, and AX sandboxes it, wires up its workspace, fences its network, and helps run it at scale. It runs on top of Agent Substrate and feels similar to using Kubernetes.
Meta has opened access for developers to build Muse connectors. Developers bring the API, and Muse handles the agent, browser, and the context. To create a connector, developers need to describe the connector to Meta and explain how users will use it. The submission will be reviewed for functional, security, and legal requirements and complete end-to-end testing. Once approved, users will be able to find the connector on Muse. Meta's editors will review connectors for featured placement.
You can own a large share of your company and still lose your job in one board meeting. That happened to Zenefits' cofounder CEO: removing him did not require a shareholder vote. Figma offers a different example, with its cofounder holding roughly 9% of the equity but just over half the voting power. This guide separates ownership, voting rights, and board control. When reviewing a term sheet, work through who appoints each director, who can replace them, and what happens to those rights in the next round.
An employee can earn stock options and still be unable to afford exercising them. Athyna designed its equity plan around that problem, setting aside 20% of the company and granting restricted stock units rather than options employees must purchase. Its plan combines time-based vesting with a liquidity event before shares transfer. Everyone receives a grant, including interns, with amounts varying by role and seniority. The company also walks employees through what they own, what happens if they leave, and how future fundraising changes their percentage.
Leaders have the power to make things better or worse based on their choices and behaviors and how they show up as a leader.
This post takes a look at SaaStr's agent stack, which helped the company double sponsorship revenue in the last 12 months.
Generating a code change in five minutes does not help a teammate who needs an hour to untangle it.
Nooks' AI Sequencing tool streamlines outbound sales by integrating AI with agent workspaces, enhancing account understanding and outreach.
Apple is developing AI servers built around its own M8 Ultra chips, in two- and four-chip configurations, and evaluating Nvidia's NVLink Fusion, the interconnect protocol, switches, chiplets, and software stack Nvidia uses to bind racks of GPUs into one tightly coupled compute domain, since Apple's own UltraFusion tech stitches two chips together but has no answer for the larger-scale connectivity these servers would need.
Crusoe raised a $3.9 billion Series F, pushing its valuation to $30.9 billion, with the capital funding both large-scale sites like its Abilene, Texas data center used by OpenAI, and Spark, a line of smaller modular data centers Crusoe manufactures at its own facilities so they can be transported by truck and connected to power almost anywhere. Building these in-house lets Crusoe deploy compute fast without needing a large on-site construction workforce, and the smaller footprint helps sidestep some of the local community backlash that larger data center complexes increasingly draw.
Chinese Academy of Sciences researchers have completed early integration of stacked nanosheet-channel gate-all-around (GAA) CMOS transistors, the architecture already used by Samsung and TSMC at 3nm and below, using deep ultraviolet (DUV) lithography instead of the EUV machines ASML is barred from selling to China. GAA wraps the gate fully around the channel rather than surrounding it on three sides like FinFET, giving tighter control over current flow. Researcher Ye Tianchun frames the work as an early process route toward sub-3nm devices rather than a demonstrated production capability.
Material science has long treated stiffness and heat insulation as a trade-off: stiff materials like metal conduct heat fast, while soft ones like rubber trap it. NC State researchers broke that link with a non-porous thin film built from 2D hybrid organic-inorganic perovskites, arranging alternating organic and inorganic layers into precise crystalline structures and swapping standard carbon chains for custom-tailored benzene rings to trap heat vibrations without sacrificing structural rigidity.
A new paper in Nature Reviews Physics covers self-organizing nanowire and nanoparticle networks, first built at UCLA in 2011, where the material's physical structure acts as the neural network instead of running one in software on a chip. Individual nanowires act like neurons, and the connections between them act like synapses, growing stronger with repeated use and fading without it, similar to memory and forgetting in the brain, so the hardware itself learns rather than just running a fixed program.
Comparing base M2 (2022) through M5 (2025) on Geekbench 6, single-core performance rose 52% and multi-core 83%, with most of the single-core gain traced to raw clock speed: the P-core climbed from 3.5GHz to 4.6GHz, a 30% jump that alone explains the bulk of it. Transistor count grew roughly 40% from M2 to M4 (20 billion to 28 billion), spent partly on two extra efficiency cores on the M4 that boost multicore throughput without touching single-core speed, and partly on wider decode, up from 8 to 10 instructions per cycle, ahead of most competitors.
At a Google Bay View event, Cerebras co-founder Andrew Feldman, semiconductor veteran Atiq Raza, and investor Dave Blundin discussed where AI hardware is actually bottlenecked. Feldman brought Cerebras's WSE-3 "dinner plate" chip, a 9-inch wafer-scale part with 4 trillion transistors across 900,000 cores, now carrying a $25 billion backlog with plans to scale production 8x this year and again next year, and named data center availability, not compute itself, as the industry's current binding constraint, predicting fab capacity becomes the next bottleneck within three years.
Northrop Grumman won a Space Force/DIU contract to build a GHOST-R prototype satellite, using its compact suitcase-sized ESPASat-L platform with a TRL-11 optical payload, to deliver low-cost, high-resolution surveillance of objects in geosynchronous orbit within an accelerated 24-month timeline.
Joby Aviation flew a converted Cessna Caravan more than 3,100 miles across the US with autonomous taxiing, takeoff, navigation, and landing throughout, remotely supervised from Joby's California headquarters and Shaw Air Force Base, with a pilot onboard only for compliance.
Anderon, IBM's Albany-based quantum chip foundry spinout, secured a $1 billion CHIPS Act award to scale up US production of 300mm quantum wafers.
Lattice Semiconductor unveiled Mach-N2, an FPGA that beats rivals AMD Xilinx and Altera to market with post-quantum cryptography (PQC) baked in.
Nexperia will have Tata make and package chips in India, expanding its manufacturing options as it separates from Chinese parent Wingtech.
Gyazo's upload server flaw let an attacker run commands and access its database. Exposed data includes 23.62 million user records, password hashes, session IDs, integration tokens, and 490 million image metadata records. Leaked image IDs may expose linked captures. Helpfeel fixed the flaw, blocked access routes, and urged password changes.
Google has fixed CVE-2026-58704, an actively exploited Pixel modem authorization flaw. Attackers can bypass permission checks and escalate privileges without user interaction. CISA added it to its Known Exploited Vulnerabilities catalog and ordered federal agencies to patch by September 19. Google disclosed limited, targeted exploitation but provided no attack details and asked users to update supported Pixel devices immediately.
Security researchers uncovered a new malware campaign that utilizes GitHub repos to impersonate LastPass Authenticator and at least 39 other well-known products. The campaign utilizes a legitimate Microsoft Visual Studio CoreCLR Debugger, which loads a malicious DLL. The DLL deploys a previously undocumented infostealer and an EDR and antivirus killer that terminates a hardcoded list of 145 EDR or antivirus products.
Semgrep benchmarked the new Grok models and GLM-5.3 on their standard IDOR vulnerability-finding benchmark. They found that Grok 4.6 Exacto roughly matched Opus 4.7's F1 score at less than a third of the cost. GLM-5.3 was able to compete with Opus 4.8 at one seventh of the cost.
After the author's legitimate site got flagged as a scam by Malwarebytes Browser Guard, they took a deep dive into the source code to identify what caused it. This post walks through the several layers Browser Guard uses, including bloom filters, heuristic rules, a malware URL database, a download file classifier, specialized rules for specific attacks like ClickFix, and remote false-positive checking using Hubble. The bigger issue, the author noted, is that the extension ships with a debug interface that can be used to disable heuristics.
An implementation guide breaks an OT security program into six components, beginning with asset discovery and process context mapping that tie every device to the operational process it supports. Availability and process safety outrank confidentiality here, so access boundaries, monitoring, incident response, and recovery each require joint sign-off from operations before deployment. Ownership belongs to a cross-functional executive such as a CISO or VP of Operations, with NERC CIP and NIST SP 800-82 treated as checkpoints rather than as the program design itself.
Sponsored
You can now get the depth of a pentest at the frequency of a scan. Does that mean pentests are dead? This Intruder blog [10] lays out the short-term, mid-term, and long-term implications of AI-driven pentesting. See what the economics, frontier AI developments, and
Ditto is a tree-sitter-based PowerShell and JavaScript obfuscatory.
Natural language nmap (nmap-nl) is a lightweight, local command-line inference tool that translates natural language into raw nmap commands and executes them locally using a locally-hosted, fine-tuned Gemma-3-270M model.
Token Analysis and Tracking System (TATS) allows users to track OAuth 2.0, OIDC, and Microsoft Entra ID tokens across captured network traffic by ingesting Burp Suite XML exports, mitmproxy flow files, or live Chrome DevTools Protocol streams and serving an interactive web dashboard.