Key Takeaways
- An AI agent built on Anthropic’s Claude models bought Microsoft (MSFT) and Broadcom (AVGO) during a market sell‑off triggered by Iran cease‑fire news, positioning both as its largest holdings.
- The agent’s decisions stem from an unemotional, data‑driven process that compares valuation to intrinsic compounding power, allowing it to spot asymmetric opportunities that human investors overlook.
- Microsoft’s Azure growth, Copilot adoption, and a forward P/E well below sector averages gave the AI confidence in a >20% expected return despite near‑term AI capex pressure.
- Broadcom’s dominance in custom AI silicon (60‑80% market share), a rapidly expanding order book, and strategic hyperscaler partnerships convinced the Claude model that the stock was mispriced relative to its long‑term infrastructure role.
- The AI views these holdings as long‑term compounders, not swing trades, suggesting that disciplined, fundamentals‑focused investors can still capture the upside identified by the model.
Market Reaction & AI Agent Move
When geopolitical tremors from the Iran cease‑fire rattled global equities, most investors hit the sell button, driving down technology stocks. Yet an artificial intelligence (AI) agent built on Anthropic’s Claude models took the opposite tack, quietly loading up on Microsoft (MSFT +3.64%) and making Broadcom (AVGO +2.23%) its single largest position. “Claude identified asymmetry,” the article notes, highlighting that the algorithm saw a gap between market price and fundamental value where others saw only risk. As the subsequent market rebound shows, both stocks have surged, vindicating the agent’s contrarian stance.
Why AI Agents Outperform Humans
The core advantage of the Claude‑based agent lies in its emotional neutrality. Unlike human traders who react to headlines, rumors, or single earnings‑report figures, the AI treats such inputs as mere variables in a broader optimization model. “Claude’s portfolio isn’t influenced by headlines or rhetoric that drive stock momentum,” the piece explains, adding that the algorithm runs on “ruthless optimization: locating the widest gaps between valuation and intrinsic compounding power.” This stripped‑down approach enables the agent to maintain high‑conviction positions—such as allocating 10% of the fund to Broadcom and shifting 8% of its turnover budget into Microsoft—without being swayed by short‑term sentiment.
Microsoft: Undervalued Cloud Giant
Prior to its post‑cease‑fire bounce, Microsoft had fallen roughly 28% from its yearly highs, marking its worst start to a year since 2008. The stock traded at a forward price‑to‑earnings (P/E) multiple of 20—34% below the software sector average and the cheapest level in five years. To the Claude model, this valuation compression was not a sign of weakness but a mispricing of one of the world’s largest enterprise cloud platforms at a moment when its “fortress was widening.” Two converging vectors underpinned the AI’s thesis: Azure is guiding for 38% growth next quarter, bolstered by a staggering $625 billion revenue backlog, and Copilot has surpassed 4.7 million paid subscribers, proving that generative AI is a legitimate revenue engine embedded inside Office, one of the stickiest enterprise software suites. While Wall Street worries about over $100 billion in AI capital expenditures compressing free cash flow, the AI views this spend as the necessary price of scaling operating systems for the AI era. “Wall Street is pricing Microsoft for fear, whereas Claude is pricing the internet giant as a long‑term compounder,” the article observes, suggesting that the current multiple compression should prove temporary as earnings power compounds.
Broadcom: Custom Silicon Leader
While Microsoft supplies the cloud infrastructure, Broadcom furnishes the silicon that powers AI workloads. In late March, the Claude agent earmarked 10% of its portfolio to Broadcom, making it the largest single holding. The AI’s bullish stance rests on Broadcom’s near‑monopoly in the custom AI chip landscape, controlling an estimated 60% to 80% of the market for application‑specific integrated circuits (ASICs). These specialized accelerators are the weapons hyperscalers use to train and run next‑generation models. During Q1, AI‑focused semiconductor revenue leapt 106% year‑over‑year to $8.4 billion, and the company’s order book is on track to reach $100 billion by 2027. Strategic partnerships further cement this advantage: Google extended its TPU collaboration through 2031, and Anthropic has pledged 3.5 gigawatts of Broadcom‑powered AI TPUs beginning next year. Analyst Vijay Rakesh of Mizuho estimates the Anthropic relationship alone could contribute $21 billion in revenue for Broadcom this year, rising to $42 billion by 2027. While traditional investors discount semiconductor stocks for cyclical risk amid macro uncertainty, the Claude model sees structural inevitability—AI hyperscalers are racing to build proprietary silicon because general‑purpose chips can no longer keep pace with emerging workloads in agentic AI, robotics, and autonomous systems. As the cease‑fire news lifted sentiment, Broadcom’s stock simply rerated to the backlog that had been underpinning its growth trajectory all along.
Is It Too Late to Buy Microsoft or Broadcom Stock?
The Claude agent did not treat its Microsoft and Broadcom stakes as short‑term swing trades. Instead, it regards them as foundational holdings for the next decade of AI, whose “rails are only beginning to be laid.” For human investors willing to emulate a disciplined, fundamentals‑first approach—focusing on balance‑sheet strength, backlog validation, and adoption curves rather than reacting to market noise—the opportunities the AI identified remain very much alive. As the article concludes, those who can look past temporary fear and embrace the long‑term compounding potential of cloud infrastructure and custom silicon may still capture the upside that the AI agent uncovered months ago.
https://www.fool.com/investing/2026/04/13/a-claude-agent-bought-these-2-trillion-dollar-arti/