Ex-Palantir turned politician Alex Bores says AI deepfakes are a ‘solvable problem’ if we bring back a free, decades-old technique

New York Assemblymember Alex Bores, now a Democratic congressional candidate in Manhattan’s 12th District, argues cryptographic provenance (C2PA) is a practical fix for deepfakes and urges it become a default standard for images, audio and video. Bores authored the Raise Act, recently signed into law, which requires so-called “frontier” AI labs (named include Meta, Google, OpenAI, Anthropic, XAI) to publish safety plans, disclose critical safety incidents and withhold models that fail internal tests; he says compliance would be minimal. He warns that while technical provenance is essential, statutory bans on harms such as deepfake child sexual abuse remain necessary, and his stance has drawn opposition from a pro‑AI super PAC pledging millions against him.

Analysis

Market structure: Short-term winners are vendors of cryptographic provenance, enterprise security and cloud KMS services (benefit: increased demand and pricing power as platforms bolt-on C2PA-like tooling); losers are high-volume ad-driven content venues facing higher moderation/tech costs (pressure on margins for META/GOOGL over 12–24 months). Incumbents (Google, Meta) have scale to absorb compliance costs, which increases their competitive moat and raises barriers for smaller AI startups that can't easily bear reputational/regulatory risk.

Risk assessment: Tail risks include a major election-period deepfake that triggers rapid federal mandates or large class-action suits (low probability, high impact; could cut ad growth by >5% YoY for implicated platforms). Immediate (days) market moves should be muted; expect measurable pricing impact in 3–12 months as standards adoption and state/federal rulemaking clarify; over 2–5 years provenance requirements could become a new compliance line item (0.5–1% revenue drag for ad-heavy firms). Hidden dependencies: success hinges on device/browser OEM adoption (Apple/Google buy-in) and default-on UX decisions — without those, uptake stalls.

Trade implications: Tactical longs: enterprise security/provenance plays and Palantir (PLTR) as a 1–3% position for 6–12 months given gov’t/enterprise demand; tactical shorts: modest (1–2%) exposure to META/GOOGL if they issue weak guidance or face PAC-driven ad boycotts in next 3–9 months. Options: buy 3–6 month 25–30 delta puts on META/GOOGL sized to 0.5–1% portfolio risk to hedge regulatory shock; alternatively buy 6–12 month calls on PLTR for asymmetric upside.

Contrarian angles: The market may overestimate near-term damage to big tech — if compliance costs are truly ~1 FTE as claimed, negative revenue impact could be <1% and the narrative favors large incumbents. Historical parallel: HTTPS transition initially feared then concentrated trust in large platforms; provenance could similarly entrench cloud giants. Unintended consequence: strict provenance defaults may concentrate power with providers who control signing keys, creating new systemic risk and regulatory focus on custody of provenance keys.

Key Decisions for Investors

• Establish a 1–3% long position in PLTR (Palantir) over the next 2–6 weeks as a hedge against rising demand for provenance/security tooling; add on any sell-off >8% from entry or after an adverse quarterly guide.
• Initiate a 1–2% short exposure to META and a separate 1–2% short to GOOGL (or use inverse ETFs) to capture potential ad-revenue and margin pressure over the next 3–12 months; tighten stops at 15% adverse move and cover if both firms guide <3% ad growth next quarter.
• Purchase 3–6 month 25–30 delta puts on META and GOOGL sized to 0.5–1% portfolio risk to protect against a regulatory shock or major deepfake incident; roll or exit after legislative milestones (bill introduction/hearings) within 90 days.
• Allocate 2–4% to pure-play cybersecurity/cloud KMS names (e.g., CRWD, OKTA, AWS play through AMZN) over 6–18 months; increase allocation by +50% if Apple/Google announce default C2PA-like support within 6 months.
• Monitor three specific catalysts (1) federal AI/deepfake legislation text posted — act within 7 days, (2) Apple/Android adoption of provenance metadata — act within 30 days, (3) any major election-related deepfake incident — deploy hedges immediately and increase short sizes by 50%).