What if a Few AI Companies End up with All the Money and Power?

Key Takeaways

• Anthropic's enterprise AI revenue may already surpass OpenAI's
• Agentic coding drives high-margin profits for business-focused AI firms
• Cybersecurity applications could cement a moat for top AI model providers
• If AI tools become essential for defense, spending will skyrocket
• Concentration of AI profits raises risk of extreme economic inequality

Pulse Analysis

Agentic coding has become the AI sector’s "killer app," allowing businesses to automate software development and data analysis with a single prompt. Unlike OpenAI’s consumer‑first strategy, Anthropic has built a sales engine around enterprise contracts, translating rapid adoption into higher ARR and a clearer path to profitability. Analysts note that Anthropic’s compute efficiency further narrows the margin gap, positioning it to outpace OpenAI’s revenue trajectory and potentially dominate the lucrative B2B AI market.

The next battleground is cybersecurity, where advanced coding agents can both discover and exploit system vulnerabilities at scale. Anthropic’s delayed release of its Mythos model, cited for its ability to uncover decades‑old bugs, signals a shift toward AI‑driven defensive tools. As attackers adopt equally powerful models, organizations will be compelled to purchase premium AI solutions to stay secure, creating a self‑reinforcing revenue loop for the few firms that can supply the most capable models. This arms‑race dynamic promises multi‑digit billions in annual spend on AI‑enhanced security services.

Beyond profit, the concentration of AI capabilities raises systemic concerns. If a handful of companies control the most powerful models, they wield disproportionate influence over critical infrastructure, finance, and even political processes. Such dominance could exacerbate wealth inequality, echoing Piketty‑style scenarios where a small elite captures the majority of economic gains. Policymakers and regulators will need to grapple with antitrust, data‑privacy, and national‑security implications before the AI market solidifies into an oligopoly.