- OpenTools' Newsletter
- Posts
- 👷🏻♂️Trump’s AI Tech Force
👷🏻♂️Trump’s AI Tech Force
PLUS: MIT's Robotic Assembly System | Data Centers Devour Power Grid
Reading time: 5 minutes
🗞️In this edition
Trump launches 1,000-person US tech force for AI
Sponsored: Remio - AI powered personal knowledge hub
MIT develops AI system that builds furniture from text prompts
Data centers overwhelming U.S. power grid
In other AI news –
Ant expands AI health app with family profiles and smart reminders
Creative Commons explores pay to Crawl as AI scraping grows
Nvidia doubles down on open source AI with Schedmd deal
4 must-try AI tools
Hey there,
Governments are pulling industry insiders directly into policy and infrastructure, researchers are pushing AI from screens into the physical world, and data centers are stretching America’s power grid to its limits.
We're committed to keeping this the sharpest AI newsletter in your inbox. No fluff, no hype. Just the moves that'll matter when you look back six months from now.
Let's get into it.
What's happening:
The Trump administration Monday unveiled a new initiative dubbed "U.S. Tech Force," comprising about 1,000 engineers and other specialists who will work on AI infrastructure and technology projects throughout the federal government.
Participants will commit to a two-year employment program working with teams reporting directly to agency leaders in "collaboration with leading technology companies," according to the official government website.
Those "private sector partners" include Amazon Web Services, Apple, Google Public Sector, Dell Technologies, Microsoft, Nvidia, OpenAI, Oracle, Palantir, Salesforce, and numerous others.
Tech Force shows the Trump administration increasing focus on developing America's AI infrastructure as it competes with China for dominance in a rapidly growing industry.
Once Tech Force members complete their two terms, they can seek full-time jobs with those companies, who are committed to consider programs' alumni for employment. Private partners can also nominate their employees to do stints of government service.
Annual salaries will likely fall in the range of $150,000 to $200,000, plus benefits.
Why this is important:
The 1,000-person Tech Force with private sector partners including OpenAI, Nvidia, AWS, and Microsoft creates a direct pipeline between AI companies and the federal government.
Companies committing to consider alumni for employment after two-year stints creates a revolving door between government and industry.
Private partners able to nominate their employees for government service means companies placing their people in federal agencies with direct reporting to agency leaders.
$150,000-$200,000 salaries are significantly below what these engineers earn at partner companies (often $300,000-$500,000+ for senior engineers), suggesting other incentives driving participation.
Our personal take on it at OpenTools:
This is formalizing the revolving door between Silicon Valley and the government.
Companies nominating employees for government stints while committing to hire alumni creates a direct pipeline where industry controls who staffs federal AI projects.
The "collaboration with leading technology companies" framing is Orwellian. This is companies placing their people in government to influence policy and procurement while those people retain industry loyalties.
Reporting directly to agency leaders gives Tech Force members unusual access and authority. That's not a typical government employment structure where career civil servants provide institutional knowledge and continuity.
The partner list reads like who's who of AI infrastructure: AWS, Microsoft, Nvidia, OpenAI, Oracle, Palantir. These are companies with massive federal contracts and vested interest in shaping AI policy.
This creates conflict of interest where people staffing government AI projects have loyalties to companies bidding on those same projects. The alumni pipeline back to industry after two years ensures they maintain those relationships.
This is regulatory capture by design. Industry gets to staff the government teams implementing AI policy industry just to convince the administration to centralize at federal level.
From Our Partner:
Tired of scattered notes, lost files, and forgotten conversations?
remio unifies your digital life, transforming your knowledge into a personal intelligence source.
Here’s what you’ll love about remio:
Tailored Answers: remio provides unique answers by combining with your personal knowledge.
No More Uploads to ChatGPTs: One click to sync all your files, making your entire knowledge base chatable with AI.
Master Your Meeting: Unlimited free recording with transcription, get AI summaries with key decisions.
Privacy & Security: With a "Local First" design, all your data is stored exclusively on your device.
What's happening:
MIT researchers developed an AI-driven robotic assembly system that builds physical objects by simply describing them in words. Users can prompt the system with text like "make me a chair," and it fabricates furniture from prefabricated parts.
The system uses a generative AI model to build a 3D representation of an object's geometry based on user prompt. A second generative AI model reasons about the desired object and figures out where different components should go according to function and geometry.
A vision-language model (VLM) determines how structural components and panel components should fit together. The user remains in the loop throughout and can refine the design with new prompts like "only use panels on the backrest, not the seat."
The components can be disassembled and reassembled at will, reducing waste. Researchers evaluated designs through user study and found more than 90% of participants preferred objects made by their system compared to different approaches.
The framework could be useful for rapid prototyping complex objects like aerospace components and architectural objects. Longer term, it could be used in homes to fabricate furniture locally without bulky products shipped from central facilities.
Why this is important:
90% user preference versus alternatives validates that AI-driven component placement matches human design intuition better than algorithmic approaches (placing panels on all horizontal surfaces or randomly).
Reusable, disassemblable components reducing waste addresses sustainability concerns with traditional manufacturing. Local fabrication without shipping from central facilities extends this environmental benefit.
Lowering barriers to design tools democratizes physical object creation. Non-experts can prototype without mastering CAD software requiring extensive expertise.
Our personal take on it at OpenTools:
This is an impressive research demonstration but overstates practical impact.
"Make me a chair" producing functional furniture from prefabricated parts is a constrained design problem, not general fabrication. The system works because the component library is predefined and limited. Scaling to arbitrary objects requires massive component variety.
Reusable components reducing waste is positive but requires standardized parts inventory. IKEA already does this without AI. The AI contribution is determining component placement, not the sustainability model.
Aerospace and architectural applications mentioned as future use cases are stretched. Those domains require precision, material specifications, load calculations, and regulatory compliance this system doesn't address. Furniture from prefabricated parts is a tractable problem. Aerospace components are not.
The real contribution is VLM-driven component placement reasoning. Using vision-language models to map functional requirements to geometric configurations is a novel approach. But framing this as democratizing design overstates accessibility gains.
What's happening:
A shuttered Pennsylvania coal plant is becoming a massive AI data center campus with seven 30-acre gas-fired power stations. The Homer City facility will consume as much electricity as all homes in Philadelphia's urban area.
By 2030, data centers could use 10%+ of U.S. power, a 60-150% increase from today. That's enough to power 16 Chicagos annually (430 trillion watt-hours).
Tech companies abandoning clean energy pledges. Google admits its 2030 zero-emissions goal is now "very difficult." Natural gas is filling the gap, according to the International Energy Agency.
Projects nationwide: Texas Panhandle's 5,800-acre complex with nuclear reactors, Cheyenne's 10-gigawatt campus (enough for 20× Wyoming's housing), Meta's $30B Louisiana facility stretching Manhattan's length.
PJM grid monitor serving 65 million customers urged federal regulators to block more data centers indefinitely, warning the grid can't handle additional load.
Why this is important:
10% of national power consumption is an infrastructure crisis. The grid wasn't built for this demand surge, and expanding it takes decades.
"Unretiring coal" plants contradicts every climate commitment made by tech companies. Google, Microsoft, and Meta pledged 100% renewable energy. They're now betting on natural gas and hoping carbon capture technology materializes.
PJM's call to block new data centers signals grid capacity is maxed out in the eastern U.S., home to major population centers and existing tech infrastructure.
Our personal take on it at OpenTools:
This is what the AI boom actually costs.
Energy demand forecasts show data centers producing carbon equivalent to New York, Chicago, and Houston metro areas combined by the mid-2030s. That's not incremental, it's catastrophic for climate goals.
The "we'll innovate our way out" argument is wishful thinking. Fusion power and carbon capture are decades from deployment at scale. Natural gas is the reality, and emissions are accelerating.
Tech companies face impossible math: AI requires massive, uninterrupted power. Renewables are intermittent. Grid batteries don't scale. So they choose natural gas and hope nobody notices the contradiction.
PJM blocking new data centers is the canary. If the eastern seaboard can't handle more load, where does expansion go? Rural areas with weaker grids and fewer protections?
Chinese fintech giant Ant upgrades AI health app to tap booming eldercare, wellness demand – App AQ integrates Apple and Huawei devices, adds family health profiles and smart reminders to boost user engagement.
Creative Commons announces tentative support for AI ‘pay-to-crawl’ systems – Creative Commons has come out in favor of “pay-to-crawl” technology, a system to automate compensation of website content when accessed by machines, like AI web crawlers.
Nvidia buys AI software provider SchedMD to expand open-source AI push – The chip designer doubles down on open-source technology and steps up investments in the artificial intelligence ecosystem to fend off rising competition.
Jason AI - A tool for automating B2B conversations and bookings
Voxify - A tool that allows users to create realistic voice-overs in multiple languages and accents
VModel - AI-powered tool that uses virtual models to showcase clothing and accessories on e-commerce platforms
WellyBox - A tool that helps users track and manage their receipts and invoices
We're here to help you navigate AI without the hype.
What are we missing? What do you want to see more (or less) of? Hit reply and let us know. We read every message and respond to all of them.
– The OpenTools Team
How did we like this version? |
Interested in featuring your services with us? Email us at [email protected] |