Is your security strategy ready for a world with no room for error? 💥 The 2026 Cloudflare Threat Report is out, and the findings are clear: adversaries are accelerating with AI, exploiting over-privileged SaaS integrations, neutralizing MFA through token theft, and pushing hyper-volumetric DDoS attacks to new levels. Join Cloudflare’s Cloudforce One experts to hear what the report means for 2026 and beyond. 📅 July 29 at 12 PM CT Register here:

Another Swiss bank folds crypto into its regular banking app BancaStato, the state-owned cantonal bank of Ticino, has launched regulated crypto trading through Sygnum, letting clients buy, hold, and sell $BTC, $ETH, $LTC, and $SOL inside the same apps they use for everyday banking. Sygnum's API plugs straight into the bank's Avaloq core system, a first for Avaloq's SaaS environment, with client assets held off balance sheet in Sygnum custody. The setup joins 25+ institutions on Sygnum's B2B platform, including PostFinance and SocGen-FORGE. @sygnumofficial says the model takes banks "from decision to live offering in months rather than years," and with its fresh MiCA license covering the EU, the plug-in path to bank crypto looks set to run well beyond Switzerland.

Jensen Huang's speech quick summary: -Vertically integrated but horizontally open strategy enables broad ecosystem reach -Computing demand increased 1M times in last 2 years - Current demand visibility: $1T through 2027 (vs $500B through 2026 last year) - Grace Blackwell delivers 35-50x performance improvement over Hopper (vs typical 1.5x Moore’s Law progression) - Vera Rubin platform: 7 chips, 5 rack-scale computers, integrated AI - Tokens are new commodity - data centers becoming “token factories” - Groq Integration & Specialized Computing - OpenClaw & Nvidia Nemo Claw: enterprise-secure version with policy engines, network guardrails - Every enterprise needs “OpenClaw strategy” - SaaS companies becoming “agentic as a service” - Confidential computing enables secure model deployment across regions - Pattern: NVIDIA brings customers to cloud partners through integrated acceleration - Expanding beyond data centers: space computing (Vera Rubin Space One) - Six frontier model families: Nemotron (language), Cosmos (world models), Groot (robotics), others ...

CT missed the most important earnings call of the week. It wasn't a crypto company. It was @Cloudflare Everyone caught the wallet announcement. The real alpha was in the call itself. Today Cloudflare monetizes the internet through subscriptions: security services, AI platform spend, pools of funds. A mix of SaaS and IaaS economics. The CEO was explicit that AI agents are about to break that model. Cloudflare sits in front of roughly 20% of internet traffic. Here's what that traffic looks like from the call: - AI agent requests up 1,700% YoY - Agents crossed 50% of total network traffic this quarter. First time in history non-human traffic is the majority. Management admitted it happened faster than their own models - Their projection: if trends hold, non-human traffic outnumbers human traffic 1000x within 5 years The monetization shift is the key part. The ad-supported internet doesn't work when the visitor is an agent. Cloudflare's CEO answer: block malicious bots for free, charge good agents a tiny fee per request. Fractions of a penny. They want to be the ones defining that layer. Now the throughput math here: - Cloudflare handles ~500M requests per second - They estimate 1 to 10% is monetizable via micro/nanotransactions - That means 10M TPS on day one, scaling to 100M TPS Visa peaks at ~20k TPS The CEO's framing: "we're building this while others compete with Visa." Three to four orders of magnitude beyond card rails. No existing payment network can settle this. It has to be something new. Two conclusions I keep coming back to: - Being short L1 throughput is being short agentic workflows. If agent traffic gets monetized per request, the settlement layer needs to scale orders of magnitude beyond anything live today. - The fee math for L1s flips. Base fees have collapsed across ETH, SOL, everywhere. MEV is getting internalized by apps. Hard to build a base fee revenue case at human scale. But at 10M TPS and $0.001 per transaction, you're looking at ~$315B a year in base fees alone. At 100M TPS the number gets silly. Stablecoins and crypto are the end-game here for Agentic finance @jerallaire @circle

Can Anthropic truly overtake a giant like Alphabet in such a short timeframe? While Anthropic has already surpassed OpenAI and reached a historic valuation of $1.2 trillion, the deeper question is: what prevents an even stronger competitor from disrupting Anthropic next? WhatAI argues that in a market defined by exponential growth, no lead is permanent. 1. The Logic Behind the 2028 Crossover This projection reflects a fundamental structural shift in the global economy: >Ad-Based vs. Agent-Based Economy: Alphabet’s growth remains tied to the mature search advertising market, where revenue per interaction is measured in cents. Anthropic is capturing the “AI Computing Tax” — as AI agents replace manual search, value per interaction jumps from advertising clicks to dollars in compute and automation. >Velocity of Enterprise Adoption: Claude’s reputation for high reliability and controllability has driven deep integration into enterprise workflows. Reaching ~$30B ARR by early 2026 demonstrates that demand for frontier-grade, trustworthy models is scaling faster than traditional SaaS. 2. The Winner’s Curse & Emerging Competition WhatAI’s core warning is especially relevant here: today’s leader can quickly become tomorrow’s laggard. >Architectural Breakthroughs: Dominance currently rests on Transformers. A leap to more efficient non-Transformer architectures (e.g., advanced state-space or hybrid models) could instantly reset the field in favor of a new entrant. >Vertical Specialization: While Anthropic excels at general reasoning, specialized “sovereign models” purpose-built for biotech, structural engineering, finance, or other high-stakes domains could carve away the most valuable market segments. >Open-Source Deflation: Meta’s Llama series and other open-source efforts continue to exert powerful downward pressure on pricing. Near-parity with Claude would rapidly undermine Anthropic’s premium positioning. 3. Critical Reality Checks Even aggressive projections must confront hard limits: >Energy Ceiling: Unlimited scaling is physically constrained by power grids, data center cooling, and energy infrastructure. >Regulatory Friction: Approaching trillion-dollar revenue will trigger intense antitrust scrutiny and AI-specific regulations far heavier than those faced by early internet companies. Conclusion WhatAI’s central thesis holds: the AI era marks a paradigm shift from indexing information to generating intelligence. Anthropic currently leads in reliability and enterprise trust, yet its moat remains shallow. The next, potentially stronger disruptor is likely already training on next-generation hardware or algorithms we have not yet standardized. In exponential markets, sustained victory belongs only to those who can repeatedly innovate beyond architectural, physical, and regulatory bottlenecks.

Beyond the Rails: "Starstream Circle" Special Route Coming Soon Dear passengers, welcome aboard Hethereau's Anomaly Metro "Ghost Train". The Special Route "Starstream Circle" will be opening shortly. The train is about to depart. We wish you a pleasant journey. ▸ Duration: September 24, 2026, 05:00 – October 8, 2026, 04:59 (UTC+8) ▸ Rewards: During the "Starstream Circle" Special Route event, clear event challenges for a chance to earn rewards, including Annulith, Fons, and Train Logs. ▸ Requirement: Complete the main story quest "Prologue - A New Destination - Head to Eibon Antique Shop." *For more details, head to the in-game event page. #NTE# #NevernesstoEverness#

Beyond 9YA at Trust Wallet House. An evening above the Singapore skyline at @token2049 for builders, founders & creators. Ecosystem conversations, drinks, a DJ and the 9YA toast. Oct 6th, 6pm SGT Join us 👇

Beyond the Model: Training a 397B Agent with SkyRL | Charlie Ruan (阮世麟) What does it take to make RL work at 397B scale—beyond training the model itself? Join Charlie Ruan (@charlie_ruan) for a deep dive into SkyRL and the hard-won lessons of training a 397B office-work agent—from fixing environment and harness bugs to system tuning and the final hero run. Hosted by @qingke_ai and simulcast on Zhihu Frontier. 📅 Sept 19 · 10 AM Beijing / 2 AM UTC 📺 Watch live: