AI's Systemic Disruption Rewrites Software, Energy, and Compute Economics

The AI Revolution's Unseen Costs: Why Today's Disruptions Are Just the Beginning

The current market is in a frenzy, shedding companies perceived as vulnerable to the AI revolution. This conversation, however, reveals that the true impact of AI is far more nuanced and systemic than a simple stock sell-off suggests. The hidden consequences lie not just in which companies fail, but in how the very definition of value, competition, and even the fundamental economics of software and energy are being rewritten. This analysis is crucial for founders, investors, and technologists who need to understand the second- and third-order effects of AI to build durable businesses and avoid being blindsided by the inevitable shifts in market dynamics. Those who grasp these deeper implications will gain a significant advantage in navigating the coming years.

The Unraveling of Software Value: When AI Becomes the Ultimate Commodity

The market's reaction to AI is a blunt instrument, punishing software companies with a broad brush. But the deeper implication, as highlighted by Anchor Crawford, is that AI fundamentally alters the cost of creating software. When AI can write code, the perceived value of traditional software businesses, particularly those reliant on per-seat pricing, is called into question. This isn't about AI directly replacing SaaS products; it's about AI enabling the creation of bespoke solutions so rapidly and cheaply that the old pricing models become untenable.

"When AI can write software, the cost to create the software plummets. I believe the market is starting to question the terminal value of these businesses. It will be more challenging. It will be more challenging for companies who are not AI native."

This creates a cascading effect. Smaller, "point solution" software companies are particularly at risk, as their niche can be easily usurped by larger, AI-native players or adjacent businesses that integrate AI capabilities. Larger, platform-like companies may survive, but their margin structures and growth rates will inevitably change. The conventional wisdom of valuing software based on recurring revenue and user count is being challenged. The question isn't whether software will exist, but what its economic underpinnings will be. The immediate pain of this market correction, driven by fear of obsolescence, is a necessary precursor to a more sustainable, AI-integrated future for software, but only for those who can adapt their business models.

The Energy Bottleneck: AI's Voracious Appetite for Power

While software faces disruption, the AI revolution is simultaneously creating an unprecedented surge in demand for energy. Jeff Lawson, founder of Twilio and now Inertia, a fusion energy startup, articulates this critical bottleneck. The current infrastructure, heavily reliant on fossil fuels, is insufficient to meet the projected needs of AI development and deployment. This isn't a distant problem; it's an immediate driver of massive investment in new energy solutions.

"Now we're facing this arms race with China. We need more power. We need energy addition. The way to have energy addition is to stop, stop getting rid of the stuff that already works."

Lawson's venture into fusion energy, aiming for grid-scale power in the 2030s, underscores the long-term nature of this challenge. The immediate consequence of this energy demand is a significant increase in capital expenditure for hyperscalers and hardware providers. Anchor Crawford points to "token growth" -- a measure of the intelligence AI is asked to perform -- increasing 14-fold year-over-year. This insatiable demand necessitates massive infrastructure investment, creating opportunities for hardware manufacturers and energy innovators. However, the long lead times for developing new energy sources mean that this demand will likely outstrip supply for years, creating a persistent constraint and a significant competitive advantage for those who can secure or generate power efficiently. The conventional approach of incremental energy upgrades will not suffice; a systemic shift is required.

The Compute-in-Space Frontier: Redefining AI Infrastructure

The convergence of AI, SpaceX, and the burgeoning private space industry presents another layer of systemic change. The proposed merger of SpaceX with Elon Musk's AI startup, XAI, signals a strategic move to fund ambitious AI development through the lucrative aerospace business. Mandip Singh highlights that XAI's current monetization strategies (consumer subscriptions and a defense contract) are not scaling as rapidly as competitors like OpenAI. The merger, and potential IPO, aims to provide the immense capital required for training large language models and developing advanced AI capabilities.

"So because all these large language models are converging and growing really fast, you have to ask yourself, how are they going to fund the next training run? And you need that sort of funding to really keep up in terms of how these models are developing."

This move points towards a future where compute infrastructure extends beyond terrestrial data centers. The idea of "compute in space," with satellite constellations and potentially space-based data centers, becomes a tangible possibility. The integration of AI into products like Tesla vehicles, as suggested by Singh, also illustrates how AI adoption can be accelerated through existing user bases, bypassing the slower growth of standalone AI products. The conventional understanding of AI infrastructure is being challenged by the potential for a decentralized, space-based compute network, offering unique advantages in terms of data access and processing power, albeit with significant upfront investment and technical hurdles. This strategic alignment, while risky, could create a powerful, integrated ecosystem that reshapes the competitive landscape.

The Algorithmic Tug-of-War: Design, Addiction, and Accountability

The legal battles surrounding social media's impact on young users, exemplified by Adam Mosseri's testimony, reveal a critical systemic tension: the conflict between product design that prioritizes engagement and user well-being. The argument is shifting from content accountability (Section 230) to the inherent design of the platforms themselves. Plaintiffs are framing social media as a "digital casino," where algorithms are deliberately engineered to foster addiction, leading to mental health issues.

"What they're arguing is harmful, um, is actually the design. So they're making this a personal injury, uh, they're making personal injury claims here, saying that it is the algorithm that prioritizes engagement, that it is the scrolling, that that is what is, that that it is personal injury."

This highlights a fundamental challenge: the business model of many social media companies relies on maximizing user time on platform, which can directly conflict with user health and well-being. The immediate payoff of increased engagement, through features like infinite scrolling and personalized content feeds, creates a downstream negative consequence of potential addiction and psychological harm. The companies' defense often centers on their efforts to protect children and provide tools for user control. However, the ongoing legal scrutiny suggests that these measures may be insufficient to address the systemic issues embedded in the product design itself. The long-term consequence of this algorithmic arms race is a growing demand for greater accountability and a potential re-evaluation of how digital products are designed and regulated.

Key Action Items

• For Software Companies:
  • Immediate Action: Re-evaluate per-seat pricing models and explore AI-driven service offerings or infrastructure APIs.
  • Longer-Term Investment (12-18 months): Develop AI-native architectures and business models that leverage AI for cost reduction and value creation, rather than viewing it as a threat.
• For Investors:
  • Immediate Action: Discern between AI-disrupted software companies and AI-native platforms; avoid broad sector-based selling.
  • Longer-Term Investment (18-24 months): Identify companies with strong AI integration, robust energy sourcing strategies, or those positioned to benefit from the growth of space-based compute.
• For Energy Innovators:
  • Immediate Action: Focus on scaling existing technologies and securing funding for pilot projects to meet surging AI-driven energy demand.
  • Longer-Term Investment (3-5 years): Accelerate research and development in advanced energy solutions, such as fusion, to address the long-term energy deficit.
• For AI Developers & Platform Companies:
  • Immediate Action: Prioritize responsible AI development, including robust guardrails and ethical considerations, to build trust and avoid regulatory backlash.
  • Longer-Term Investment (2-3 years): Explore novel monetization strategies beyond subscriptions, such as integrated commerce and specialized enterprise solutions, to fund ongoing model development.
• For Social Media Platforms:
  • Immediate Action: Proactively implement and transparently communicate enhanced child safety features and user well-being tools.
  • Longer-Term Investment (1-2 years): Consider fundamental shifts in engagement-driven algorithms to prioritize user health and reduce addictive design patterns, anticipating future regulatory pressures.
• General Strategic Imperative:
  • Requires Patience (6-12 months): Embrace solutions that involve immediate discomfort (e.g., business model changes, significant R&D investment) for durable, long-term competitive advantage.