AI Energy and Security Infrastructure Take The Stage

Nuclear-AI Infrastructure: Thursday's Critical Hearing

China builds two coal plants weekly to power AI development. America needs a different path to technological dominance. Thursday's House Science Energy Subcommittee hearing on "Powering Demand: Nuclear Solutions for AI Infrastructure" addresses exactly this challenge.

The Power Problem

AI's exponential demands have hit infrastructure reality. We are seeing weekly announcements about large datacenter build-outs, but every major AI announcement faces the same constraint: reliable baseload power. Nuclear power can deliver the massive, consistent electricity AI requires.

Key Witnesses

Pat Schweiger (Oklo CTO): Four decades in advanced reactors, recently hired to commercialize fast-reactor technology for AI infrastructure at this publicly traded nuclear startup.

Kathleen Barrón (Constellation Energy): Leads the nation's largest carbon-free power producer. Constellation already powers Microsoft datacenters through their Three Mile Island restart.

Dr. Jeremy Renshaw (EPRI): Directs AI optimization for power grid operations, showing how artificial intelligence can enhance electricity generation and distribution.

What's At Stake

China's power-abundant model lets them throw massive electricity at AI compute, accepting higher costs for worse compute because they have plenty of power. American companies face grid connection delays while our competitors accelerate.

This week’s hearing addresses whether America will build nuclear-powered AI infrastructure or watch others capture technological leadership through superior energy capacity.

Nuclear companies are seeing record investment while AI infrastructure demands hit historic levels, and DC is finally recognizing that energy security is economic, technological, and national security. Thursday's hearing showcases whether America will clear regulatory barriers and build the power infrastructure our AI ambitions require.

The AI-Cyber Convergence: America's Next Security Frontier

Thursday's House Homeland Security Subcommittee on Cybersecurity and Infrastructure Protection is having a hearing titled "Security to Model: Securing Artificial Intelligence to Strengthen Cybersecurity" — a topic that recognizes the current moment is a pivotal time in our cybersecurity field as AI transforms from experimental technology to one being used throughout government, private industry, and critical infrastructure.

The witness lineup tells the story of this transformation.

Microsoft's Federal Security CTO Steve Faehl: Brings Zero-Trust architecture expertise from the government trenches.

Securin's Kiran Chinnagangannagari: Scaled a vulnerability management platform from startup to million-dollar revenue in under twelve months.

Trellix's Gareth Maclachlan: Orchestrated the massive McAfee-FireEye integration.

Cranium AI's Jonathan Dambrot: Spun his company out of KPMG specifically to secure AI/ML pipelines.

These are operators defending systems under active attack today while building the AI capabilities that will define the era we’re entering into tomorrow.

The Strategic Reality

The witnesses represent exactly ALFA’s approach to technology challenges: using AI to strengthen cybersecurity rather than treating it as a threat vector alone.

Yesterday’s cybersecurity model—monitor breaches, respond to incidents—cannot scale to meet AI-era threats. Machine-speed attacks require machine-speed defenses.

Subcommittee Chairman Garbarino clearly understands this evolution. By focusing on how AI strengthens cybersecurity, they're moving toward "AI for security”, a recognition that America's cyber defense strategy must be AI-native, not AI-fearful.

The bottom line: America's cybersecurity future will be defined by how quickly we deploy AI-enhanced defenses, not how carefully we regulate AI development.

FIELD NOTES

Reports indicate that U.S. trade negotiators are prepared to exchange looser export controls of AI chips for much-needed rare earth minerals here in the U.S.

NEC Director Kevin Hassett said on CNBC yesterday saying that the slow-rolling of mineral exports to the U.S. could disrupt production for U.S. companies that rely on these minerals.

While a deal would still prevent the most powerful chips from entering China, the Administration acknowledges a reality that we have been describing for months: Huawei’s advances present foreign technology companies with an emerging alternative to American chips. Yet, our rare earth supply chain is still heavily concentrated in China. Growing our mining production domestically remains a huge priority. But such a significant supply chain shift takes time, and this is a practical measure to ensure American companies can continue to develop and build cutting edge products.

The Wall Street Journal has a story out this morning on how chatbots are decimating traffic on news sites as search gives way to answers.

“Social-media platforms such as Facebook and Twitter helped funnel online traffic to publishers, but ultimately pivoted away from giving priority to news. Search was a stalwart traffic driver for more than a decade, despite some turbulence as Google tweaked its powerful algorithm. Generative AI is now rewiring how the internet is used altogether.”

Here’s a related post from The Chernin Group’s Jarrod Dicker to analogize the current moment we are in:

The model around media and creative works is about to face its most challenging time yet.