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: Major Market Milestones, Expanding Enterprise Agents, and Regulatory Clashes
This week saw leading AI developers reach unprecedented revenue milestones and unveil massive hardware investments, underscoring how rapidly the technology is industrializing. Autonomous AI agents are actively reshaping enterprise software and the physical workforce, even as escalating regulatory battles threaten to fracture the deployment landscape.
Major Investment and Market Milestones:
Anthropic reached a $14 billion annualized revenue run rate in February with 12x year-over-year growth and a $380 billion valuation—signaling that generative AI is rapidly shifting from experimental IT budgets to core enterprise operations
OpenAI closed 2025 with a $20 billion annualized run rate and an $840 billion valuation, achieved in just 24 months—proving that enterprise willingness to pay for AI is compounding at an unprecedented pace
NVIDIA CEO Jensen Huang projected $1 trillion in purchase orders for its Blackwell and Vera Rubin architectures by 2027, though Wall Street remained cautious amid lingering concerns over AI market bubbles and uncertain long-term returns
Jeff Bezos is reportedly seeking $100 billion to launch a fund aimed at acquiring and automating legacy manufacturing firms using AI, marking a significant strategic shift toward applying frontier models directly to heavy industry
AI's Evolving Impact on the Workforce:
Meta is reportedly planning to lay off up to 20% of its 79,000-person workforce to offset the enormous costs of its anticipated $600 billion data center buildout by 2028—a stark illustration of the tradeoffs tech giants are making to fund AI infrastructure
Data center construction is creating a lucrative boom for skilled tradespeople, with 25–30% pay increases for workers moving into technical AI infrastructure roles, as the US faces a projected shortfall of nearly 500,000 construction workers by 2027
Handshake AI posted job listings seeking improv actors to train models on authentic emotional expression and tone—a signal that AI labs are moving beyond text generation into replicating complex human interaction
A Harvard Business Review study of 1.4 million prompts from 2,500 KPMG employees found that while 90% use AI regularly, only 5% qualify as highly sophisticated users, highlighting that access alone isn't enough—organizations need to actively develop AI reasoning skills across their workforce
Enhanced AI Safety and Governance Efforts:
The Trump administration unveiled a national AI legislative framework aimed at preempting state-level regulations and limiting legal liability for developers, setting up a significant battle with states over how to enforce safety guardrails and platform accountability
The US Department of Defense labeled Anthropic an "unacceptable risk to national security" after the company refused to allow its models to be used for mass surveillance or autonomous weapons—exposing a deepening conflict between commercial AI safety principles and military objectives
The US Treasury published an AI Risk Management Guidebook for financial institutions detailing 230 specific control objectives, forcing a heavily regulated sector to mature its AI governance and incident response processes
European Union parliament committees voted 101–9 to propose a blanket ban on AI "nudifier" systems, continuing Europe's pattern of legislating against specific generative AI applications deemed inherently harmful
Expanding AI Integrations Across Tech Giants:
OpenAI partnered with Amazon Web Services to distribute its models to US government agencies, including within classified top-secret environments, challenging competitors on their home turf while capturing lucrative federal contracts
Snowflake launched Project SnowWork in research preview, giving business users a conversational AI platform capable of autonomously executing multi-step enterprise workflows—turning static databases into active, agentic operating systems
WordPress.com enabled AI agents to draft, edit, and publish content directly using the Model Context Protocol, signaling a major shift in how digital media will be produced without human intervention
Visa initiated its "Agentic Ready" program in Europe, testing payment infrastructure with banks including Commerzbank to process transactions initiated entirely by AI agents—laying the financial plumbing for a future where autonomous software buys and acts on behalf of users
Strategic Hardware Developments:
NVIDIA's networking division surged to $11 billion in quarterly revenue, a 267% year-over-year increase, proving that the interconnectivity infrastructure for large GPU clusters is becoming as lucrative as the processors themselves
Frore Systems reached a $1.64 billion valuation after developing liquid-cooling technology specifically for AI chips, reflecting the industry's urgent need for thermal management as hardware grows increasingly power-hungry
Goldman Sachs estimated that AI workloads will consume 30% of total global data center capacity within two years—a structural shift that will fundamentally reshape the cloud computing landscape
From record-breaking revenue milestones to the first federal AI legislative framework, this week made clear that AI is no longer being evaluated—it's being institutionalized. The tension between rapid commercial deployment and meaningful governance will define the next phase of adoption, and the gap between how organizations access AI and how effectively they use it remains the most underappreciated challenge in the industry.
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