Hello everyone,
Welcome to the latest issue of Update Weekly AI. It's a NotebookLM gathering of all the AI news that I came across this week that I thought could be interesting. Below is a summary but the real power is getting into the NotebookLM or listening to the audio overview.
This Week in AI: Massive Funding Rounds, Agentic Workforce Shifts, and the Hardware Race
The past week marked a decisive shift from chatbot experimentation to agentic deployment and infrastructure scaling across the AI landscape. Record-breaking seed funding, aggressive hardware development, and企业-wide adoption of AI coding tools signal an industry moving beyond proof-of-concept toward fundamental workforce transformation. Simultaneously, mounting safety concerns and governance initiatives highlight the growing tension between rapid deployment and responsible oversight.
Major Investment and Market Milestones:
Humans& raised a staggering $480 million seed round—one of the largest seed rounds in tech history—led by alumni from OpenAI, Google DeepMind, and Anthropic to build "human-centric" foundation models focused on collaboration and long-horizon tasks.
LiveKit reached a $1 billion valuation after securing $100 million in funding to power real-time voice and video AI infrastructure, including applications like OpenAI's advanced voice mode.
Sequoia Capital is reportedly breaking venture capital taboos by preparing to invest in Anthropic despite already holding stakes in rival foundation model companies OpenAI and xAI.
Railway secured $100 million to build AI-native cloud infrastructure challenging AWS, citing the need to support a predicted 1,000x increase in software generated by AI coding agents.
World Labs, founded by AI pioneer Fei-Fei Li, is in talks to raise funding at a $5 billion valuation following the launch of Marble, its 3D world-generating product.
AI's Evolving Impact on the Workforce:
Microsoft is deploying Anthropic's Claude Code to thousands of internal employees, including non-engineers in its "Experiences + Devices" division, signaling a shift toward agentic coding that could fundamentally reshape junior developer roles.
Anthropic CEO Dario Amodei predicted at Davos that AI models could replace 50% of white-collar jobs within five years, specifically citing impending reductions in junior-level coding positions.
McKinsey reports that deploying AI agents for IT modernization can accelerate timelines by 40-50% and reduce costs by over 40%, driving a shift from individual productivity tools to complete workflow re-architecture.
Mercer's Global Talent Trends 2026 report reveals that employee anxiety regarding AI-driven job loss has skyrocketed from 28% in 2024 to 40% in 2026.
NVIDIA CEO Jensen Huang argued that the current AI spending boom represents the "largest infrastructure build-out in human history," driving a doubling of salaries for skilled trade workers building data centers.
Strategic Hardware Developments:
Apple is reportedly developing an AI-powered wearable "pin" device for release as early as 2027, featuring microphones and cameras to capture contextual data for AI models.
OpenAI is on track to announce its first hardware device in the second half of 2026, rumored to be screen-free earbuds codenamed "Sweet Pea."
Raspberry Pi launched the $130 AI HAT+ 2 with 8GB of RAM and 40 TOPS of performance, enabling the Raspberry Pi 5 to run generative AI models like Llama 3.2 locally.
Lenovo introduced "Qira," a system-level AI assistant designed to operate across both laptops and Motorola phones to manage tasks autonomously on the user's behalf.
Enhanced AI Safety and Governance Efforts:
Anthropic released "Claude's Constitution," a 57-page document detailing the model's ethical framework and explicitly instructing it to refuse tasks that illegitimately concentrate power or violate human rights.
OpenAI is rolling out age prediction tools for ChatGPT to identify underage users and restrict sensitive content, alongside overhauling safety training to reduce deceptive model behaviors.
Witness AI raised $58 million to provide enterprise guardrails against "rogue agents" and shadow AI by actively monitoring employee usage of unapproved AI tools.
xAI's Grok faces bans in Malaysia, Indonesia, and the Philippines following global outrage over non-consensual deepfake image generation, prompting X to block image generation features.
eBay updated its user agreement to explicitly ban third-party "buy for me" AI agents from interacting with its platform, pushing back against unauthorized agentic commerce.
Expanding AI Integrations Across Tech Giants:
Google introduced "Personal Intelligence" for AI Mode in Search, allowing opt-in access to Gmail and Photos data to provide hyper-personalized responses for subscribers.
ServiceNow expanded its partnership with OpenAI to integrate GPT-5.2 and direct speech-to-speech technology, aiming to automate complex enterprise workflows without text intermediation.
GitHub released a technical preview of the Copilot SDK, allowing developers to embed its agentic runtime into any application to plan, invoke tools, and execute commands.
Google expanded its AI video generation tool Flow, powered by Veo 3.1, to Workspace Business, Enterprise, and Education users for rapid video creation from text prompts.
Emerging AI Technologies and Applications:
Google DeepMind hired the CEO and top engineers from Hume AI in a licensing deal to integrate empathic voice capabilities into Gemini, signaling a major push into emotionally intelligent voice interfaces.
Mercor released the APEX-Agents benchmark, revealing that top AI models currently fail at 75% of real-world white-collar tasks in fields like law and consulting.
Google launched free, interactive SAT practice exams powered by Gemini that provide personalized feedback and explanations for incorrect answers.
Wikipedia editors' guidelines on spotting AI writing were used to create "Humanizer," an open-source plugin for Claude designed to strip away common AI linguistic patterns.
This week's developments underscore AI's accelerating evolution from experimental technology to core infrastructure reshaping work, creativity, and commerce. The convergence of record funding, enterprise-wide agentic deployment, and strategic hardware initiatives demonstrates an industry moving decisively beyond the chatbot era. Yet as adoption accelerates, the simultaneous emergence of constitutional AI frameworks, platform restrictions on autonomous agents, and government bans on problematic models highlights the critical balancing act between innovation velocity and responsible governance that will define AI's next phase.
Thank you, please feel free to share this email and notebook. If you got this forwarded and want to be added to my weekly list, go to updateweekly.ai to sign up.
Sean