Last week, Anthropic launched Claude Code Security, and cybersecurity stocks such as CrowdStrike, Palo Alto Networks, Check Point, and others fell. Shares of traditional software companies also dropped sharply this week. IBM fell 13% in a single day after Anthropic said its AI tools can help with modernizing COBOL. A viral Substack post outlined a hypothetical scenario where AI rewrites the economics of knowledge work, and food delivery, payments, and Indian IT services stocks consequently fell (as of Thursday, most had recovered).
Investors are clearly worried. But where others panic, I see opportunity.
New developments with the Claude platform have major implications:
The rise of AI as digital labor.
Massive pressure on SaaS companies.
Verticalization accelerates.
Enterprise buying behavior changes.
A new infrastructure layer becomes critical.
The advice I’m giving founders right now
Claude for Everything is both a threat and a gift. A threat if you’re building something shallow. A gift if you’re building something with a deep MOAT.
Here’s the advice I’m giving founders right now:
Build systems of work, not systems of record.
Own a vertical, deeply.
Accumulate proprietary data loops fast.
Own painful, multi-step workflows, distribution, and the customer relationship.
Price for outcomes.
Design trust and safety from the start, with human-in-the-loop controls. And don’t build on a single model provider.
The market selloff is a signal, not a verdict
The market selloff this week is an overreaction to disruption anxiety. That anxiety is legitimate. But the winners in this next era won’t be determined by which incumbents survive the news cycle — they’ll be the companies, new and old, that figure out how to own execution inside the workflows that matter most.
The next decade won’t be defined by general chat interfaces. It will be defined by AI embedded invisibly within how work actually gets done (“Work as a Service”) by domain- and vertical-specific agents.
Signals Shaping the Future of AI:
Infrastructure
AMD and Meta sign multi-year deal to deploy up to 6 gigawatts of AMD Instinct GPUs. The partnership spans multiple GPU generations, with first shipments starting in 2H 2026 using a custom MI450-based chip and 6th Gen EPYC CPUs, aligning roadmaps across silicon, systems, and software. Click here
Meta signs multi-billion-dollar AI chip rental deal with Google. The multi-year agreement will see Meta use Google’s Tensor Processing Units to train and run advanced AI models, as it diversifies beyond Nvidia and AMD amid escalating AI infrastructure spending. Click here
OpenAI faces compute constraints as “Stargate” buildout stalls. Reports indicate delays around OpenAI-owned data center efforts, keeping near-term reliance on external infrastructure partners and available capacity. Click here
Meta halts development of its most advanced in-house AI chip. After reported design challenges, the company pivoted to a less complex chip, highlighting how difficult full-stack AI hardware integration remains, even for hyperscalers. Click here
Enterprise
OpenAI partners with McKinsey, BCG, Accenture, and Capgemini to scale its Frontier AI agent platform. The consulting giants will help enterprises deploy and govern OpenAI’s new agent orchestration system, signaling a push to dominate enterprise AI workflows and compete directly with Anthropic and traditional SaaS vendors. Click here
Perplexity launches “Computer,” a $200/month multi-model AI agent platform. The new system orchestrates 19 models, including Claude, Gemini, GPT-5.2, Grok, and Veo, to autonomously complete complex, multi-step tasks in the background for Max subscribers. Click here
Google launches Nano Banana 2 to lower enterprise AI image generation costs. The Gemini 3.1 Flash Image model brings Pro-tier reasoning and text rendering to Flash pricing, cutting per-image costs roughly in half while adding 4K output and built-in provenance watermarking. Click here
Cloudflare says it used Anthropic’s Claude to reimplement 94% of the Next.js API in one week. The experimental open-source project, Vinext, replaces Vercel’s Turbopack with Vite to ease deployment beyond Vercel, with roughly $1,100 spent on AI tokens. Click here
OpenAI and Amazon strike multi-year AI partnership backed by $50 billion investment. AWS will become the exclusive third-party cloud provider for OpenAI Frontier and co-develop a stateful runtime on Bedrock to power enterprise AI agents at scale. Click here
Block is cutting more than 4,000 jobs as it pivots to an AI-first operating model. Jack Dorsey said the fintech company will shrink from over 10,000 employees to fewer than 6,000, arguing that AI tools and smaller teams enable a fundamentally different way of building and running the business. Click here
Capital Flows
OpenAI raises $110 billion in a record private funding round. Amazon committed $50 billion, Nvidia $30 billion, and SoftBank $30 billion, valuing OpenAI at $730 billion pre-money and expanding its AWS and Nvidia compute partnerships. Click here
Encord raises $60 million to scale data infrastructure for physical AI. The San Francisco startup, now valued at $550 million, is building a multimodal data platform for robotics, autonomous vehicles, and drone companies, positioning itself as a challenger to Scale AI in sensor-heavy AI workflows. Click here
MatX raises $500 million in Series B funding to develop AI training chips competing with Nvidia. The startup, founded by former Google TPU engineers, plans to manufacture its processors with TSMC and begin shipments in 2027, aiming to significantly improve LLM training efficiency. Click here
Anthropic acquires AI agent startup Vercept and will shut down its product. Vercept, which raised about $50 million and built a cloud-based computer-use agent, will sunset its platform as parts of the team join Anthropic to advance Claude’s agent capabilities. Click here
Research
Mizzou researchers develop an AI model to encode 3D protein structures. The GCP-VQVAE system translates protein geometry into AI-learnable representations and reconstructs them with near-atomic accuracy, aiming to speed up drug and protein design workflows. Click here
Anthropic publishes an AI Fluency Index. The index tracks how humans collaborate with AI systems, shifting the focus from raw usage metrics to measurable workflow integration. Click here
Inception launches Mercury 2 diffusion model. The new model is positioned as faster and cheaper than transformer-based rivals for certain workloads, reviving the cost-performance debate for inference-heavy AI applications. Click here
Policy
Anthropic refuses Pentagon request to remove AI safeguards tied to a $200 million defense contract. CEO Dario Amodei said the company will not lift protections against mass domestic surveillance or the use of fully autonomous weapons, despite threats to terminate the partnership and to label Anthropic a supply chain risk. Click here
U.S. officials say DeepSeek used NVIDIA’s Blackwell chips to train a new AI model despite export controls. The chips were reportedly deployed at a data center in Inner Mongolia, raising fresh concerns about the enforcement of U.S. AI semiconductor restrictions. Click here
Pentagon moves to develop AI tools for China-focused cyber operations. The U.S. Department of Defense is advancing efforts to build and deploy artificial intelligence systems to support cyber operations targeting China, as part of a broader push to integrate AI into military and intelligence workflows. Click here
Global AI Strategy
Japan’s Rapidus secures $1.7 billion to advance domestic 2nm chip production. The government-backed venture raised ¥267.6 billion from the state and firms including Sony, Toyota, and SoftBank as it targets mass production of 2-nanometer chips by 2028.Click here
South Korea approves Google’s request to export detailed map data overseas. The government reversed its longstanding restriction on high-resolution geographic data exports, clearing the way for Google to improve Google Maps functionality in the country after years of limited service. Click here
OpenAI will make London its largest research hub outside the U.S. The company cited the UK’s talent pool and research ecosystem as it expands its AI footprint, aligning with Britain’s push to position itself as a global AI superpower. Click here
Talent Signals
Each week, we spotlight key roles tied to the themes shaping this week’s AI headlines, connecting talent to the companies driving the news.
Wayve is a UK-based autonomous AI and self-driving startup building flexible, learned driving systems aimed at commercial robotaxi launches. After raising $1.2 billion from partners including Mercedes-Benz, Stellantis, Nissan, Nvidia, Microsoft, and Uber, Wayve is expanding its team across AI research, perception, and software engineering. Open roles are listed on its careers page. Click here
Sierra builds AI agents for enterprise customer experience, enabling brands to deploy autonomous service agents that handle complex, multi-step interactions. As companies shift from chatbots to outcome-oriented digital labor, Sierra is hiring across AI systems, product, and go-to-market roles. Open roles are listed on its careers page. Click here
Arcade builds AI-powered tools that generate demos, visuals, and product storytelling assets in minutes, helping companies compress product communication and GTM workflows. As teams race to explain and sell products faster, Arcade’s platform is becoming a go-to SaaS layer for interactive demos and visual content automation. Open roles are listed on its careers page. Click here
You can see all the opportunities at Mayfield-backed AI companies here, and across the broader ecosystem here.
Social Signals
The most important conversations in AI are unfolding across social media, where top voices are shaping the next wave of signals and strategy. Here are some of the top social signals and their takes from the past week.
Jack Dorsey (Click here) — “Today we’re making one of the hardest decisions in the history of our company: we’re reducing our organization by nearly half, from over 10,000 people to just under 6,000. We’re not making this decision because we’re in trouble. Our business is strong. But something has changed. We’re already seeing that the intelligence tools we’re creating and using, paired with smaller and flatter teams, are enabling a new way of working which fundamentally changes what it means to build and run a company.” In a post viewed more than 53 million times, Dorsey announces a major workforce reduction at Block and frames it as a structural shift toward AI-enabled productivity rather than a response to weak performance. He argues that intelligence tools allow the company to operate with fewer layers and smaller teams, signaling how AI is beginning to reshape cost structures, operating leverage, and the capital intensity of scaling technology businesses.
Guillermo Rauch (Click here) — “If you thought your company’s edge was ‘how fast you ship’, you’re in for a rude awakening. Everyone can ship fast now. Obviously, not everyone can ship tastefully, with quality and restraint in mind. That’s the new edge.” Rauch, CEO of Vercel, argues that AI tools like Claude, Cursor, and v0 have compressed prototyping from weeks to minutes, making speed widely accessible rather than differentiated. As execution becomes commoditized, the advantage shifts to the final 20 percent, polishing, subtracting the unnecessary, and exercising taste in UX, reliability, and design. The discussion reflects a broader transition where AI handles much of the build phase, but human judgment, product sense, and quality control determine which products stand out and endure.
Nick Turley (Click here) — “ChatGPT just crossed 900M weekly users and 50M paying subscribers. The part I love most is seeing how people use it differently. For a lot of people, ChatGPT is where they start with AI.” Turley highlights the scale milestone as evidence that ChatGPT has become the default on-ramp to AI for writing, research, trip planning, shopping, and task execution. He notes that broader usage feeds back into faster iteration, improved reliability, and more natural responses, underscoring how distribution, engagement, and paid conversion are compounding alongside product improvement at a global scale.
