Gemini now lets you import chats and memory from other AI apps

Key event: Google Gemini now supports importing chats and memory from other AI apps via gemini.google.com/import, permitting upload of .zip exports up to 5 GB per file and up to 5 .zip files per day. Imported chats appear in Gemini's side panel with an import icon, are searchable, and can be deleted per-import or reuploaded (reuploads add new conversations and overwrite prior imports). The article explicitly names ChatGPT and Claude and provides step-by-step export instructions; import-memory requires pasting a prompt-generated result into Gemini. The feature is not available in the EEA, Switzerland, or the United Kingdom.

Analysis

Google’s import feature is a low-friction lever to turn multi-platform AI usage into first-party engagement, materially shortening the path to personalization and stickiness. Even a modest conversion — if 5–10% of active users of alternative assistants import chats — should measurably raise Gemini weekly engagement and increase ad+cloud telemetry within 3–6 months, since richer personal context amplifies upstream recommender and query substitution effects.

Second-order supply-chain effects favor firms that provide tooling and governance around data portability: identity verification, provenance checks, and ingestion-layer sanitization will be required as imports scale. That implies a potential 12–24 month revenue runway acceleration for security/observability vendors and a parallel increase in marginal costs for smaller AI startups that must harden export formats or face churn.

Key tail risks are regulatory and security: regulators in EEA/UK already restrict the feature, and a single high-profile privacy incident from cross-platform import could trigger fines and user opt-outs that reverse adoption within weeks. The upside is multi-quarter: training signal improvement and monetization of richer user state are likely to materialize over 12–36 months, but initial valuation re-rates should be tied to measurable KPIs (import volume, retention lift, enterprise uptake) rather than announcement alone.

Contrarian read: market narratives that this feature destroys rival stickiness understate the friction of messy, non-standardized exports and legal constraints on reusing imported content for model training. The real benefit is utility for power users and enterprises, not mass consumer exodus; expect a gradual, measurable revenue contribution rather than an immediate winner-takes-all shift.

Key Decisions for Investors

• Buy GOOGL (or GOOG) via a 6–12 month call spread sized to 2–3% of portfolio to capture 15–30% upside if import-driven engagement metrics accelerate; finance by selling a 10–15% OTM call to limit premium outlay. Max loss = premium; target take-profit 20–30% from entry.
• Implement a protective collar on existing GOOGL exposure ahead of regulatory scrutiny: buy 3–6 month 8–12% OTM puts and sell equivalent-duration 15–20% OTM calls. This limits downside from a regulatory/privacy shock (10–20% drawdown scenario) while keeping upside participation.
• Add a tactical 0.5–1% portfolio hedge in cybersecurity (e.g., CRWD or PANW) via 3–6 month calls to monetize the increased demand for ingestion/sanitization tools. Rationale: a privacy incident would re-rate defenders 20–40% in 1–3 months; downside limited to premium paid.
• Set data-driven entry triggers rather than calendar ones: increase long exposure if Google discloses import volume growth >5% month-over-month or a measurable retention uplift of >3–5% within the first 90 days; reduce exposure if EU/UK regulatory actions broaden or a material breach is reported.