Welcome to this week's edition of Overclocked!

In this issue, Washington just laid out a national AI blueprint, and workers are asking whether “efficiency” is code for pink slips. Let’s break it down ⬇️

In today’s newsletter ↓
🏛️ White House drops major AI Action Plan
🏢 Companies blame “efficiency” while layoffs rise
🧩 Big Tech faces a UK probe over fresh AI alliances
🛒 Walmart rolls out “super agents” for shopping
✍️ Challenge of the week: write your own AI transparency pledge

🫡 What America’s AI Action Plan Really Means

The White House released a 28-page AI Action Plan aimed at cementing U.S. leadership while policing model risks. Crafted with input from 14 agencies and dozens of firms—including open letters of support from Microsoft, Google, and Anthropic—the blueprint rests on three pillars: supercharge innovation, protect people, and lead globally.

💰 Money on the Table

The AI Action Plan as currently constructed, doles out billions of dollars for GPUs, disinformation testing, and AI training for federal workers: 

• $5 billion in “Compute for America” grants over four years to subsidize GPU clusters for universities and startups.
  

• $1.2 billion for secure model evaluation labs at NIST and DARPA, targeting bio-risk and disinformation testing.
  

• $800 million to retrain federal workers on AI-driven tools.

Funds would be allocated in the FY-2026 budget, pending congressional approval.

🛡️ Guardrails & Governance

The plan orders the FTC to draft explainability rules for consumer-facing AI by Q1 2026 and directs OSHA to study workplace safety for human-robot collaboration. A national Model Registry—voluntary at first—will log training data provenance and safety evaluations. Anthropic praised the registry as “a step toward transparent frontier research.”

🌐 A Geopolitical Sprint

Officials frame the plan as a bid to “out-pace and out-compete” China in foundational models. A proposed Trusted Compute Corridor would fast-track export licenses for chipmakers building fabs in allied nations. Conversely, Section 9 threatens “heightened scrutiny” of U.S. venture funding flowing into Chinese AI labs.

🤔 What’s Missing?

Privacy groups argue the plan lacks teeth: no federal pre-emption means state rules still vary widely. Labor advocates note the blueprint mentions “displacement support” but sets no new safety net. And the AGI clause? Absent—leaving the definition of “frontier model” to future rule-making.

If Congress funds even half these proposals, expect an arms-race for grant dollars and a wave of compliance jobs. If not, the plan may join a shelf of unimplemented tech roadmaps. Either way, the document signals that AI policy is now a ballot-box topic heading into 2026.

🏢 Companies Downplay AI’s Role in Layoffs

A CNBC analysis of regulatory filings and internal memos finds that at least 38,000 tech workers have been laid off since January, with companies citing “re-orgs” or “efficiency.” What’s often missing: the word AI.

🔍 Reading Between the Lines

• Microsoft’s July staff memo said cuts “align resources with strategic growth,” even as the firm posted record profits and doubled AI capex to $14 billion.
  

• Disney’s spring restructuring trimmed 7,000 roles while announcing an AI-centric “Imagineering Futures” lab.
  

• IBM told investors AI will create more jobs than it removes—then froze hiring for 7,800 back-office posts it expects software to automate.
  

A study from Capacity Media estimates 46% of roles eliminated in recent tech layoffs overlapped tasks large language models can now perform. Executives, however, rarely link cuts to automation, fearing backlash from employees and regulators.

💬 Worker Sentiment

“I spent six months teaching a chatbot how to do my job,” one former marketing analyst told NBC, “then my position was ‘redundant’—go figure.” A leaked Slack channel at a major cloud vendor shows staff joking about being “prompted out of a paycheck.”

📈 The Transparency Question

Labor economists urge the Securities and Exchange Commission to require disclosure when AI materially affects headcount. Until then, investors get vague “productivity” language while workers connect the dots themselves.

One bright spot: some firms pair layoffs with AI-upskilling stipends—$3,000 at Amazon Web Services, $5,000 at Shopify—though uptake remains modest.

The takeaway? If AI isn’t listed in the press release, check the earnings call. Odds are the bots got a promotion when humans lost theirs.

The Weekly Scoop 🍦

🧩 UK antitrust probe targets Big Tech’s newest AI alliances

🧪 AI lab discovers protein-folding tweak that could speed drug design 

🎙️ Sam Altman tells podcast host “AI scares me, too” 

🛒 Walmart bets on AI “super agents” to streamline online shopping 

💻 Researcher jailbreaks Amazon’s coding tool; wipes mock customer data 

🌏 McKinsey bans China projects involving generative AI

💡 Weekly Challenge: Draft Your Personal AI Transparency Statement

1. ✍️ Map Every AI Touchpoint – list all tools you use both personally (Spotify DJ, Snapchat My AI, Ring doorbell) and professionally (ChatGPT, Grammarly, Excel Copilot, Salesforce Einstein).
   

2. 🔍 Rate the Impact – tag each as assist (spell-check), co-create (draft emails, slide decks), or replace (automated reporting, chatbots that answer customers).
   

3. 🛡️ Write Two Short Policies –
   

   • Personal: when will you tell friends, followers, or family a post, photo edit, or résumé tweak was AI-enhanced?
     

   • Business: what must you disclose to teammates or clients when AI generates content, code, or analysis? Include audit steps (e.g., “human fact-check required before publish”).
     

4. 📢 Publish & Share – drop the work policy in Slack or Confluence; pin the personal version to your social bio or a notes app. Transparency builds trust.
   

5. 💬 Gather Feedback – ask a colleague and a friend: “Would this level of disclosure satisfy you?” Iterate.
   

6. 🔄 Review Quarterly – calendar a 15-minute check-in every three months. Remove tools you ditched; add the new ones.

That’s it for AI news updates this week! Policy blueprints, hidden layoffs, and transparency challenges—AI’s impact is everywhere. Which headline hit home? Hit reply and share.

Zoe from Overclocked