Michael Burry adds a new Microsoft bet, doubles down on software

Michael Burry disclosed new positions in Microsoft and increased holdings in MSCI, PayPal, and Adobe while also raising bearish exposure, with put options now about 5% of his portfolio. He framed software and payment stocks as oversold despite AI disruption concerns, and said he did not sell any software names. The note is more sentiment and positioning-driven than fundamental, but it could support near-term volatility in software and fintech shares.

Analysis

The important signal is not the individual buys; it is the divergence between implied dispersion in software and the index-level bid for downside protection. That combination usually shows up late in a tape: crowded momentum can keep higher-quality software names pinned if fund flows keep chasing large-cap AI exposure, but the broader market hedge suggests the marginal buyer is no longer comfortable paying for perfect outcomes. In that setup, MSFT is the cleanest relative winner because it can absorb multiple compression better than lower-quality names while still retaining AI optionality.

The second-order effect is that the market is starting to discriminate between software with pricing power and software with narrative-only AI exposure. IBM and NOW weakness matters because it can spill into every enterprise vendor with mediocre net retention or long implementation cycles; if those names continue to trade poorly, the market is likely to further penalize ADBE and PYPL on duration and take-rate risk, not just sector beta. MSCI is more insulated than the others because recurring index/data revenue tends to re-rate better during volatility spikes, making it the most defensive member of this basket.

The contrarian read is that this is less a call on absolute fundamentals than on timing: Burry appears to be buying where sentiment has already done most of the damage, while paying for portfolio-level convexity in case the rally rolls over. That means the thesis can work in two distinct windows — a 2-6 week mean-reversion trade if software gets a relief bid, or a 3-6 month relative-value trade if AI optimism narrows to only the highest-quality platforms. The main risk is that index concentration and passive flows overwhelm valuation for longer than expected, leaving beaten-down software cheap but dead money.

For the broader market, the presence of puts at only ~5% of portfolio suggests caution, not outright crash positioning. That is consistent with a regime where tails are being hedged, but not enough to imply immediate stress; if volatility spikes, the first beneficiaries should be high-beta downside leaders rather than the software longs, so the hedge is about funding and drawdown control more than a precise timing signal.

Key Decisions for Investors

• Long MSFT / short NOW for 1-3 months: express the view that the market will reward balance-sheet quality and AI monetization over execution-risk software; target 8-12% relative outperformance with a tight stop if NOW re-accelerates on bookings.
• Long MSCI as the defensive quality name in the group; hold 2-4 months and use any broader market vol spike to add. Risk/reward is attractive because recurring revenue and low cyclicality should compress downside while maintaining upside if rates fall.
• Sell PYPL calls or structure a call spread against a long equity position over the next 6-8 weeks: the stock needs a credible catalyst to break the overhang, and upside is likely capped unless payment volume or margin commentary inflects. Best if implied vol remains elevated.
• Avoid chasing ADBE outright; if wanting exposure, use a pair trade long MSFT / short ADBE for 2-3 months to isolate AI platform winners from application-layer multiple compression. The trade works if the market continues to reward infrastructure over workflow software.
• Use index downside hedges rather than single-name shorts for the broader market view: 1-2 month SPY or QQQ puts funded by selling put spreads in the names with strongest balance sheets. The point is convexity into a volatility event, not a structural bearish outright.