

AlphaEarth
#37 in Science & Bio AIGoogle · 3× · last seen Aug 10, 2026
AlphaEarth Foundations is a geospatial Foundation Model developed by Google DeepMind that compresses satellite imagery, radar, LiDAR, and climate data into compact 64-dimensional Embeddings per 10x10-meter ground pixel. The model functions as a "virtual satellite" and enables mapping, change detection, and environmental monitoring across time periods from 2017 to present. The resulting annual Embeddings are provided free of charge as a "Satellite Embedding" dataset via Google Earth Engine and Google Cloud Storage and are used by organizations such as UN FAO, Harvard Forest, and MapBiomas. Additionally, Google Maps Platform recently began offering "Custom Satellite Embeddings" for high-frequency, location-specific queries in private preview.
Features
| Compliance/Certification | No formal certification documented; accuracy peer-reviewed via AEF paper showing ~24% (23.9%) lower error rate vs. tested models |
| Deployment Model | Cloud-based (precomputed embeddings as dataset), no user-side model hosting needed, no GPU required |
| Use Case Scope | Geospatial mapping: agriculture, deforestation, urban growth, water resources, ecosystem classification, change detection |
| Integrations | Google Earth Engine, Google Cloud Storage, BigQuery, Vertex AI, Google Maps Platform (Custom Satellite Embeddings, private preview) |
| License | Creative Commons Attribution 4.0 (CC-BY 4.0), attribution required |
| Platform | Google Earth Engine, Google Cloud Storage (Cloud-Optimized GeoTIFFs), BigQuery, Vertex AI, Source Cooperative |
| Price | Free via Google Earth Engine (free for research, education, nonprofit use); GCS bucket set to 'provider pays' since July 2026 (no user cost) |
| Release Date | July 30, 2025 (announcement); annual updates since then, e.g. 2025 update completed |
More products in this category: Science & Bio AI
Sources (3)
Company Analysis: Google
Data supports strong operating momentum (Q2 2026 revenue +24% YoY; operating margin 34%; Cloud profitability at 35.6% margin), and valuation is moderate on trailing earnings (~16.7x P/E in early Sep 2026). (sec.gov) However, the near-term investment case is constrained by materially higher capital intensity (FY2026 CapEx $195B–$205B) and the demonstrated sensitivity of sentiment to free-cash-flow pressure, alongside ongoing U.S. antitrust remedies uncertainty. (investing.com) A HOLD is warranted until there is clearer evidence that elevated AI CapEx translates into sustained cash generation without impairing margins or increasing regulatory-driven structural costs.
Summary
Alphabet Inc. (Google) is a diversified digital platform company whose core monetization engine remains advertising (Google Search, YouTube, and Network), complemented by a rapidly scaling Google Cloud business and a growing set of subscription/consumer services (e.g., YouTube subscriptions, Google One). Its core competencies are (1) global-scale consumer distribution (Search/Android/Chrome/YouTube), (2) data/ML infrastructure and model development (Gemini and related tooling), and (3) performance advertising and measurement capabilities that translate intent into ROI for advertisers. Market position remains leadership in global search and a top-tier position in digital video via YouTube, with competitive advantages rooted in default distribution, product integration, and advertiser demand aggregation. In Cloud, Alphabet is positioning “Enterprise AI solutions” as a primary growth driver, leveraging its infrastructure and model stack to win incremental workloads. (blog.google) Recent months’ key developments are dominated by Q2 2026 results and AI-infrastructure investment. Alphabet reported Q2 2026 revenue of $119.8B (+24% YoY; +23% constant currency) and operating margin expansion to 34% (operating income +30% YoY). Google Cloud accelerated sharply (company-highlighted 82% YoY growth) with segment profitability improving materially (reported Cloud operating margin 35.6%). (sec.gov) However, management raised FY2026 CapEx guidance to $195B–$205B (from $180B–$190B previously), and Q2 free cash flow turned negative (reported around -$5.9B), shifting investor focus from earnings strength to cash conversion and capital intensity. (investing.com) Valuation (USD-listed shares) screens as moderate versus many mega-cap peers: trailing P/E is ~16.7x as of early September 2026 (methodologies vary slightly by provider). (ycharts.com) For EUR framing, investors should translate USD price/earnings into EUR using the prevailing EURUSD reference rate (ECB: 1 EUR = 1.1592 USD on 11 Sep 2026), noting that the multiple itself is currency-neutral but EUR returns will be FX-sensitive. (ecb.europa.eu) Short- to medium-term outlook is a trade-off: operating performance and Cloud profitability momentum are strong, but elevated AI CapEx and related depreciation/opex can pressure free cash flow and near-term valuation support. A second major overhang is regulatory/antitrust remedies risk in U.S. search, where DOJ communications indicate “significant remedies” were obtained (timing/implementation uncertainty remains). (justice.gov)
Key Takeaways
- Q2 2026 showed strong top-line and profitability: revenue $119.8B (+24% YoY) and operating margin 34% (+2pp YoY), indicating continued operating leverage despite AI investment. (sec.gov)
- Google Cloud is a key incremental earnings driver: Cloud growth accelerated (company-highlighted 82% YoY) and profitability improved substantially (reported Cloud operating margin 35.6%). (sec.gov)
- Capital intensity is the near-term swing factor: FY2026 CapEx guidance increased to $195B–$205B and Q2 free cash flow turned negative, raising execution and cash-return questions. (investing.com)
- Valuation is not stretched on trailing earnings: trailing P/E around ~16.7x in early September 2026, which can provide downside support if earnings durability holds. (ycharts.com)
- Regulatory outcomes remain a material risk: DOJ updates point to meaningful remedies in U.S. search, with potential implications for distribution economics and default placement arrangements. (justice.gov)
Action Ideas
12–24 month accumulation thesis focused on (1) sustained double-digit consolidated revenue growth and margin resilience, (2) Cloud profitability scaling (already demonstrating high segment margin), and (3) a valuation multiple that is moderate on trailing earnings versus many large-cap growth peers. The key condition is that incremental AI CapEx translates into durable Cloud/ads product improvements rather than structurally lower cash conversion. Data points supporting the setup: Q2 2026 revenue +24% YoY, operating margin 34%, and Cloud margin 35.6%, alongside a trailing P/E ~16.7x (early Sep 2026).
Horizon: 18 mo.
6–12 month hold for investors prioritizing confirmation on cash conversion. The operating P&L is strong, but the market’s near-term debate is centered on whether AI-driven CapEx and operating costs normalize fast enough to restore positive and growing free cash flow. Maintain exposure while monitoring (a) CapEx trajectory versus guidance, (b) Cloud margin sustainability, and (c) any incremental commentary on AI monetization in Search/YouTube.
Horizon: 9 mo.
3–6 month risk-reduction thesis for portfolios with strict free-cash-flow discipline or low tolerance for regulatory uncertainty. The factual basis is that Alphabet has entered a phase of materially higher capital intensity (FY2026 CapEx $195B–$205B) and has already posted a negative free cash flow quarter, which can increase valuation sensitivity to rates and sentiment even if earnings remain strong.
Horizon: 6 mo.
Contrarian Insights
- • The market focus on negative Q2 free cash flow may overstate structural deterioration: the same quarter showed operating margin expansion to 34% and strong Cloud profitability (35.6% margin), implying the core earnings engine is not weakening even as investment ramps. (sec.gov)
- • Consensus narratives often frame AI in Search as a monetization risk; Alphabet’s own investor materials emphasize AI-driven engagement and AI-powered ad tools improving advertiser ROI, suggesting AI features can be a performance-ad tailwind rather than purely a cannibalization vector (though measurement will be key). (blog.google)
Sources (7)
- https://www.sec.gov/Archives/edgar/data/1652044/000165204426000066/googexhibit991q22026.htm
- https://www.spglobal.com/market-intelligence/en/news-insights/research/2026/07/alphabet-postq-ai-growth-accelerates-as-spending-weighs-on-cash-flow
- https://blog.google/alphabet/investor-presentation-june-2026/
- https://www.ecb.europa.eu/stats/policy_and_exchange_rates/euro_reference_exchange_rates/html/index.en.html
- https://www.justice.gov/opa/pr/department-justice-wins-significant-remedies-against-google
- https://www.justice.gov/atr/case/us-and-plaintiff-states-v-google-llc
- https://ycharts.com/companies/GOOGL/pe_ratio