AI is reviving tech sectors that VCs had all but forgotten

Health and wellness VC deals jumped to $678.0M across 23 transactions in Q4 (vs an eight-quarter average of $332.0M and 16 deals), driven by AI-native consumer and provider tools. Cybersecurity hit a record $643.1M across 15 deals with average valuations rising to $273.4M (vs prior average $129.1M); notable financings include Function Health’s $300M Series B at a $2.5B valuation (11.5x step-up), 7AI’s $130.6M Series A, and Braveheart Bio’s $185M debut round. PitchBook’s ETI shows elite VCs redeploying into healthtech, cyber, biotech and enterprise SaaS where the pitch is “AI-native” outcome-driven products rather than point solutions.

Analysis

The emergence of AI-native founders is changing where and how capital gets allocated: the gating factor is high-quality data + continuous feedback loops, not just a UI or workflow tweak. That structural pivot favors vendors that own compute, vectorized data platforms, and unified observability — because the marginal dollar of startup spend goes to infrastructure that shrinks model iteration time, not to more headcount. Expect disproportionate demand for GPUs, inference-optimized chips, managed feature stores, and MLOps tooling over the next 6–24 months as these startups scale from prototypes to production.

Second-order winners include incumbents that can embed outcomes into contracts (health systems, CROs, MSPs) and roll-up operators that can standardize AI-enabled service delivery across locations; losers will be legacy seat-based SaaS and narrow-point telehealth plays whose economics don’t support outcome guarantees. Key risks that could reverse this funding rotation are non-linear: a regulatory shock on health AI liability, a major adversarial attack undermining trust in generative cyber tooling, or a rapid jump in inference costs that reintroduces an “AI cost curve” constraint. Time horizons matter — sentiment and valuation moves occur in quarters, while durable clinical and regulatory validation plays out over years.

The market is underpricing two nuances: (1) outcome-based pricing compresses revenue visibility but expands gross margins and potential take-rates if models generalize; (2) AI-native entrants will accelerate M&A for incumbents looking to buy capabilities rather than build, creating a wave of strategic exits that will reprice both private and public comps. That suggests tactical public exposures to infrastructure and scaled integrators, paired with selective shorts in single-product telehealth and legacy SaaS names where monetization must materially rebase to justify multiples.

Key Decisions for Investors

• Long NVDA (6–12 months): buy shares or 9–12 month calls to capture continued GPU/inference demand from cross-sector AI adoption. Risk: a recession or a sudden drop in AI spend could compress multiples; reward: asymmetric given constrained supply and secular usage uplifts.
• Long CRWD or PANW (3–9 months): add exposure to proven cloud-native cyber vendors that can commercialize AI detection as a subscription outcome. Consider 6–9 month call spreads to control premium. Risk: rising competition from startups and potential margin reinvestment; reward: expansion of ARR and higher gross margins as AI features increase customer stickiness.
• Long SNOW (6–12 months): buy shares to play data + vector infra monetization as startups centralize feature stores and observability; use 1–2% position sizing. Risk: SaaS pullback reduces consumption-based revenue; reward: re-rating if consumption accelerates and partner-led integrations materialize.
• Long UNH (12–24 months) vs short TDOC (3–12 months) pair: overweight a large payer/provider that can capture AI-enabled cost savings and scale clinic integrations, short legacy telehealth operator lacking differentiated outcomes. Risk: policy/regulatory outcomes could impair ARPU upside; reward: captures margin recovery in large-cap stability vs valuation haircut for point-solution telehealth.
• Event hedge: buy 3–6 month put protection on a 2–3% notional of AI longs to guard against sudden regulatory/adverse-adjudication shocks in health or cyber that could cascade through sentiment within weeks.